University of Southern California Dissertations and Theses

Regulatory dissonance in the global development of drug therapies: a case study of drug development in postmenopausal osteoporosis

(USC Thesis Other)

Regulatory dissonance in the global development of drug therapies: a case study of drug development in postmenopausal osteoporosis

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

REGULATORY DISSONANCE IN THE GLOBAL DEVELOPMENT OF DRUG
THERAPIES: A CASE STUDY OF DRUG DEVELOPMENT IN
POSTMENOPAUSAL OSTEOPOROSIS
by
Neal E. Storm

A Dissertation Presented to the
FACULTY OF THE USC SCHOOL OF PHARMACY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF REGULATORY SCIENCE

December 2013

Copyright 2013 Neal E. Storm
2

DEDICATION
This thesis is dedicated to my father, Martin W. Storm, who for so long supported all of my
endeavors, academic and otherwise, unconditionally. It was he who inspired me to continue my
academic pursuits, wherever they led me, and continue to challenge myself in ways I never
thought possible. He did not live during a time when many professionals had the opportunity to
continue their education while working, nor did he know regulatory science first hand. However,
he understood immediately its potential as a new field of study, and understood well that if I was
to follow the future course of this new profession, that I must take advantage of this unique
educational opportunity. That was the best advice ever. Now that I am finally completing this
doctoral program, I wish only to share the outcome of this academic effort, which he inspired,
with him, since I know that he, more than anyone else in the world, would have appreciated and
savored this moment. (Martin W. Storm: 1938 – 2013)
3

ACKNOWLEDGEMENTS
I extend thanks to the survey and focus group participants for their thoughtful perspectives and
insight, and give special thanks to Drs. Arkadi Chines and Ogo Egbuna whose advice and
personal involvement was instrumental in shaping the ultimate design of this study. I am also
deeply appreciative of my thesis committee for their provocative input and guidance.
Certainly, none of this would have been possible without the strength and vision of my advisor,
Dr. Frances Richmond, through whose fortitude and energy the USC Regulatory Science Program
was created, filling a unique educational gap for a cadre of regulatory professionals that now
possess these credentials. I thank Dr. Richmond for cultivating the evolution of my thought
process, the endless stream of advice, for helping bring structure to this broadest of topics, and for
providing wisdom when the going was the toughest.
4

TABLE OF CONTENTS
DEDICATION ................................................................................................................................. 2
ACKNOWLEDGEMENTS ............................................................................................................. 3
LIST OF TABLES ........................................................................................................................... 8
LIST OF FIGURES ......................................................................................................................... 9
ABSTRACT ................................................................................................................................... 11
CHAPTER 1: OVERVIEW OF THE STUDY ............................................................................. 12
1.1 Introduction ................................................................................................................................. 12
1.2 Statement of the Problem ............................................................................................................. 20
1.3 Purpose of the Study .................................................................................................................... 21
1.4 Importance of the Study ............................................................................................................... 21
1.5 Limitations, Delimitations, Assumptions ..................................................................................... 23
1.6 Definitions .................................................................................................................................... 26
1.7 Organization of the Study ............................................................................................................ 28
CHAPTER 2: LITERATURE REVIEW ...................................................................................... 29
2.1 Introduction ................................................................................................................................. 29
2.2 Regulatory Dissonance in Clinical Trials Gu idelines .................................................................. 30
2.2.1 Regulatory Guidelines/Standards for Postmenopausal Osteoporosis ......................................... 30
2.2.2 Background on United States Guidance ..................................................................................... 34
2.2.3 Background on European Union Guidelines .............................................................................. 41
2.2.4 Comparison of Key Guidelines................................................................................................... 45
5

2.2.5 Additional Guidelines of Interest ................................................................................................ 50
2.3 Regulatory Dissonance in Postmenopausal Osteoporosis Drug Approval Precedents ............... 52
2.3.1 Cyclic Etidronate Disodium ....................................................................................................... 54
2.3.2 Tibolone ...................................................................................................................................... 55
2.3.3 Strontium Ranelate ..................................................................................................................... 56
2.3.4 Intact PTH 1-84 .......................................................................................................................... 58
2.3.5 Bazedoxifene .............................................................................................................................. 59
2.3.6 Lasofoxifene ............................................................................................................................... 60
2.3.7 Calcitonin-containing Agents ..................................................................................................... 62
2.3.8 Summary ..................................................................................................................................... 63
2.4 Ethical Controversies in Postmenopausal Osteoporosis Drug Approval Standards ................... 63
2.4.1 Placebo- versus Active-Controlled Studies ................................................................................ 64
2.4.2 Treatment Duration for Pivotal Trials ........................................................................................ 71
2.5 Recent Trends in a Changing Regulatory Landscape .................................................................. 73
2.6 Status on Developing a Common Benefit:Risk Framework ......................................................... 76
2.7 Relative Benefit:Risk of Available Therapies for Postmenopausal Osteoporosis ........................ 79
2.8 Status of Global Harmonization of Indication-specific Guidance ............................................... 84
2.9 Framing of Study of Regulatory Dissonance ............................................................................... 88
2.10 Summary and Research Direction ............................................................................................... 90
CHAPTER 3: METHODOLOGY ................................................................................................ 91
3.1 Introduction ................................................................................................................................. 91
3.2 Development of Initial Survey ...................................................................................................... 91
3.3 Survey Deployment and Analysis ................................................................................................. 95
6

CHAPTER 4: RESULTS .............................................................................................................. 97
4.1 Focus Group ................................................................................................................................ 97
4.1.1 Focus Group Outcome ................................................................................................................ 97
4.2 Analysis of Survey Results ........................................................................................................... 98
4.2.1 Responses and Profiles of Respondents .................................................................................. 98
4.2.2 Impact/Implications of Regulatory Dissonance ..................................................................... 103
4.2.3 General Causes/Factors Contributing to Regulatory Dissonance .......................................... 113
4.2.4 Alignment of Regional Indication-specific Postmenopausal Osteoporosis Guidance ........... 115
4.2.6 Potential Policy Solutions to Reduce Regulatory Dissonance ............................................... 121
4.2.7 Cross-Tabulations .................................................................................................................. 125
CHAPTER 5: DISCUSSION ...................................................................................................... 126
5.1 Regulatory Dissonance Between DRAs ..................................................................................... 126
5.2 Methodological Considerations ................................................................................................. 126
5.3 Potential Challenges of Regulatory Dissonance ........................................................................ 130
5.4 Potential Advantages of Regulatory Dissonance ....................................................................... 140
5.5 Regulatory Dissonance and Implications for Business Strategy ................................................ 141
5.6 Regulatory Dissonance and Implications for Policy Development ............................................ 149
5.7 How Best to Effect Change in the Area of Regulatory Dissonance of Indication-specific
Regulatory Requirements on a Global Level? ........................................................................................ 152
5.8 Conclusions and Future Directions ........................................................................................... 153
LITERATURE REFERENCES ................................................................................................... 155
APPENDICES ............................................................................................................................. 173
I. Survey on Regulatory Dissonance in the Global Development of Pharmacotherapies for
Postmenopausal Osteoporosis ............................................................................................................ 172
7

II. Cross Tabulation Data Analysis .................................................................................................... 176
III. Complete List of Postmenopausal Osteoporosis Therapies Approved by EMA Since
1995…………………………………………………………………………………………………..177
IV. Complete List of Postmenopausal Osteoporosis Therapies Approved by FDA Since
1995……………………......................................................................................................................178

8

LIST OF TABLES
Table 1: Significant Novel Therapies in Postmenopausal Osteoporosis (US versus EU) ............................. 32
Table 2. Key Therapeutic Agents in Development for Postmenopausal Osteoporosis ................................ 33
Table 3. Guide to FDA Action – Clinical vs. Nonclinical Study Outcomes ................................................ 41
Table 4. Primary Differences in the Elements of the FDA and EMA Guidance for Developing Agents for
Postmenopausal Osteoporosis ....................................................................................................................... 49
Table 5. Key Safety Events from Clinical Trials and Postmarketing Safety Reporting Therapeutic Product
Class of Drugs for Postmenopausal Osteoporosis ......................................................................................... 83
Table 6. Questionnaire Instrument: Breakdown of Areas of Inquiry ........................................................... 91
Table 7. Description of Focus Group - Professional Backgrounds .............................................................. 93
Table 8. Focus Group Agenda ...................................................................................................................... 95
Table 9. Question 14 – Open Responses from Respondents ...................................................................... 109
Table 10: Question 15 – Open Responses from Respondents .................................................................... 111
Table 11. Question 17 – Open Responses from Respondents .................................................................... 113
Table 12. Question 18 – Open Responses from Respondents .................................................................... 114
Table 13. Question 31 – Open Responses from Respondents .................................................................... 120
Table 14. Question 32 – Open Responses from Respondents .................................................................... 122
Table 15. Question 33 – Open Responses from Respondents .................................................................... 123
Table 16. Question 35 - Open Responses from Respondents ..................................................................... 125

9

LIST OF FIGURES
Figure 1. Question 4 - Please indicate the region in which you have acquired the majority of your
regulatory/clinical experience? Please check all that apply ........................................................................ 100
Figure 2. Question 5 - What phases of drug development have you supported for therapeutic products
intended for the treatment/prevention of PMO? Please check all that apply .............................................. 102
Figure 3. Question 6 - How many different therapeutic products for PMO have you supported during your
regulatory/clinical career? ........................................................................................................................... 102
Figure 4. Question 7 - In your role, did your job responsibilities require you to interact directly with any of
the following? Please check all that apply .................................................................................................. 103
Figure 5. Question 8 - In your experience, has regulatory dissonance increased the challenges associated
with developing drug therapies in PMO? .................................................................................................... 104
Figure 6. Question 9 - In your experience, has regulatory dissonance increased the clinical development
timelines of PMO drugs (ie, delays to study start, lengthier clinical studies)? By how much time? .......... 105
Figure 7. Question 10 - In your experience, has regulatory dissonance in drug registration requirements
increased the regulatory authority review times of marketing authorization applications of PMO drugs?
Please estimate time added to standard application reviews? ...................................................................... 106
Figure 8. Question 14 – In your experience, does regulatory dissonance impact patient access to important,
new treatments for PMO (ie, products not approved or delayed to market)? .............................................. 108
Figure 9. Question 15 - In your experience, does your company consider regulatory dissonance to be an
important factor when assessing the commercial probability of success or the net present value (NPV) of a
new potential PMO agent ............................................................................................................................ 110
Figure 10. Question 16 - In your opinion, how does regulatory dissonance impact industry innovation in
drug research and development (as it relates to PMO)? .............................................................................. 112
Figure 11. Question 18 - In your experience, which of the following lead to the most regulatory dissonance
in the setting of PMO? Please check the items that best apply ................................................................... 114
10

Figure 12. Question 26 -In your view, do intrinsic factors relating to the investigational agent under study
(e.g., safety profile, mechanism of action, novelty of product) impact the level of regulatory dissonance?
..................................................................................................................................................................... 117
Figure 13. Question 27 - Amongst the areas identified above, which in your view results in the greatest
degree of regulatory dissonance when designing pivotal, phase 3 studies in PMO? .................................. 118
Figure 14. Question 32 - In your view, which of the following potential mechanisms would be most
beneficial in reducing regulatory dissonance? Please check one item. ...................................................... 122
Figure 15. Question 33 - In your view, which of the organizations listed below should take the lead in
implementing mechanisms to reduce regulatory dissonance? Please check one item. ............................... 123
Figure 16. Question 34 - Given your drug development experience, do you think that the mechanism you
selected above to reduce regulatory dissonance in PMO would be helpful if applied in other disease areas?
..................................................................................................................................................................... 124
Figure 17. The Risk-Adjusted New Present Value Formula as Proposed by Stewart et. al. ...................... 147
Figure 18: Modified Regulatory Risk-Adjusted New Present Value Formula ........................................... 148
11

ABSTRACT
The purpose of regulatory indication-specific guidance documents, standardized regulatory
review procedures and benefit:risk assessments are to ensure the safety and effectiveness of drug
products and to set a common regulatory bar for industry. However, nationally-developed
regulatory standards for clinical trial conduct and drug approval become potentially problematic
when companies must operate across differing regional constituencies and address the
requirements of various stakeholders to attain market authorizations globally. When this occurs,
companies face dissonance in regulatory review and approval requirements and standards. The
absence of a clear global regulatory pathway complicates the development of novel therapies
since it is assumed to result in added uncertainty, time, costs and delays to the development of
important treatments for patients. The findings indicate that regulatory dissonance is
multifaceted, with perhaps greatest impact on application review outcomes, clinical trial design
and timelines, but also on areas as diverse as allowable labeling claims and reimbursement. The
degree of downward pressure on drug development and innovation from regulatory dissonance
may be underestimated by industry and may be disease state dependent. The results also suggest
that dissonance cannot be reduced by a single mechanism of harmonization. Integration of
advisement and decision-making procedures across Drug Regulatory Authorities (DRAs) and
harmonization of national regulatory guidelines may be most efficacious in reducing regulatory
dissonance. Additional benefit may result if these are implemented in parallel with other
recognizable methods of harmonization, such as standardization of benefit:risk methodology and
through expanded use of mutual recognition pacts.
12

CHAPTER 1: OVERVIEW OF THE STUDY
1.1 Introduction
As biopharmaceutical companies develop novel treatments for disease, they look to Drug
Regulatory Authorities (DRAs) for appropriate guidance upon which proposed strategies for drug
development and commercialization can be based. Such regulatory guidance ensures that drug
manufacturers are informed of currently accepted scientific and regulatory standards for
chemistry, manufacturing and controls, nonclinical testing procedures and clinical trial conduct.
This allows the manufacturer to generate a data package for a particular constituency (designated
for example in the United States (US) as a New Drug Application or Biologics License
Application, or in the European Union as a Marketing Authorization Application) that will
appropriately support regulatory decisions regarding the safety and efficacy of a therapeutic
product. Thus, DRAs play a pivotal role in setting testing standards for therapeutics, so that
uniform, informed and structured decisions can be made for the purposes of product registration
on the basis of the benefits versus risks of a novel agent for its intended indication.
Developing a legal/regulatory framework that supports new drug development and is suitable for
all of the relevant stakeholders (i.e., patients, medical community, and drug sponsors/applicants)
is a daunting task for the DRA of any country. Regulators are often placed in a position where
they must react to innovative products posing unique scientific and regulatory questions that have
not had to be answered previously. Although product innovation and basic research remain the
domain of industry and academia, DRAs play an active role in overseeing and facilitating
efficient product development, with appropriate safeguards, on behalf of the public. In fact, the
existence of food and drug regulatory authorities, and the regulatory/legal authority granted them,
are due in large part to reactionary responses by governments to public health crises stemming
13

from poor corporate stewardship, such as the introduction of inefficacious or harmful products
into the market place (Hamowy, 2010).
The unique challenges posed by novel drug products ultimately affect the direction of new trends
in regulatory science. Regulatory authority guidance and the establishment of regulatory
precedence “lags” innovation, so that available regulatory guidance may not necessarily speak to
the issues raised by novel products. Likewise, trends in regulatory science are also affected by
the types of public and private incentives for new research, and by the need to develop policy
within a landscape of differing medical, ethical, religious, cultural, sovereign legal and political
views within the medical community and public at large. Notably, few if any of these factors are
controlled directly by any single regulatory authority, but they do shape the decisions by these
agencies. Furthermore, these factors can vary from one country to another, so that the sponsor of
an investigational agent must often cope with differing rules and differing interpretations of rules
from one constituency to another.
Because regulatory approaches may differ across constituencies, sponsors seeking to develop
drugs for a specific disease indication must have a deep understanding of the regulatory landscape
of each country in which they intend to seek market authorization. Even at the pre-clinical stages
of development, sponsors must begin to develop a cohesive global strategy that bridges potential
regulatory differences across regions, and in doing so, must consider regional regulatory
precedence from public sources such as (1) regulations, (2) regulatory guidelines, (3) approved
product labeling, (4) scientific literature, (5) public assessment reports for marketed products, and
(6) published proceedings from public advisory/scientific advice committee meetings, all of
which may substantiate the particulars of any given benefit:risk assessment leading to approval.
In cases where publicly disclosed information is insufficient, case-specific scientific advice may
be obtained through confidential sponsor-regulator scientific advice meetings or correspondence.
14

These interactions typically occur at major milestones of development, such as at the end of phase
2 clinical development (i.e., EOP2 meeting), or in advance of submitting a marketing
authorization application (i.e., ‘pre-filing’, or pre-NDA/BLA or pre-MAA meeting). Although
sponsors typically obtain scientific advice in the form of protocol reviews and advice letters as
they proceed throughout product development, ‘EOP2’ and ‘pre-filing’ meetings constitute the
most significant milestones at which regional regulatory authorities provide comprehensive
guidance on the completeness of the proposed nonclinical and manufacturing programs, the
adequacy of the proposed, pivotal phase 3 clinical protocol(s), and the completeness of the overall
clinical registration program to support marketing authorization. In essence, these meetings
further refine the basic set of regulatory requirements set forth by regional guidelines by applying
them to case-specific technical issues. It is important that the drug sponsor consider the
guidelines of the relevant DRA (if available) in their proposals. If the sponsor’s proposals deviate
significantly from the guidelines, the sponsor should supply an appropriate rationale for the
proposed deviation(s), anticipating agency concerns (Kumar & Mao, 2010). Importantly, drug
sponsors who do seek scientific advice meetings during development often benefit by improving
the probability of approval of their products (EMEA, 2004a).
The mechanism by which access is gained to regulatory advice also varies across regions in ways
that may be related to the structure, organization, funding and expertise of the regulatory
authority in question. A biopharmaceutical company cannot assume that the quality of scientific
review, level of transparency, and timely access to medical review staff across regions will be
equivalent. A company seeking approval in the US can anticipate that the quality of scientific
review at the US FDA will be relatively high and consistent for analogous products within a
given product class, that “an effective mechanism to discuss questions and issues with the
agency” exists (Solberg & Richmond, 2012), and that the meeting outcomes will be respected. In
15

contrast, a similar application for approval to the Chinese Food and Drug Administration (CFDA)
might encounter lengthy reviews of inconsistent scientific and clinical quality. Although it might
be easier to consult informally with Chinese regulators, because informal meetings with review
staff may be arranged by simply walking into CFDA’s Beijing offices during certain times, the
outcomes of these meetings are informal and are not minuted. The specialized scientific
competencies required of reviewers, in addition to the resources needed to review extensive
dossiers, “puts pressure on small authorities to keep up with international standards set by
agencies such as the US FDA and the EMEA” (Hill & Johnson, 2004). Even at the FDA, funding
levels have impacted the ability of science-based regulation to keep pace with scientific discovery
(von Eschenbach, 2012). Thus, divergent levels of scientific competencies, staffing, funding and
procedural standards may act to further increase regulatory dissonance across divergent DRAs.
Challenges for global market entry are compounded by the increasing size, and thus the costs, of
pivotal phase 3 safety and efficacy studies. This development, among other reasons, has
prompted industry to increasingly conduct multi-regional clinical trials (MRCTs) in emerging
market and developing countries to lower operational costs and facilitate enrollment. In fact, a
recent evaluation of data collected from the ClinicalTrials.gov registry indicates that “one third of
the trials (157 of 509) are being conducted solely outside the United States and that a majority of
study sites (13,521 of 24,206) are outside the United States,” with “many of these trials being
conducted in developing countries” (Glickman, 2009). With this degree of cost control being
implemented in the clinical trial setting, it is surprising then that an equal level of attention has
not been paid by industry and DRAs to regulatory mechanisms that might simplify, and where
possible, align cross-regional regulatory advice and approval procedures that would achieve the
same end. A case in point, the most recent Report on Key Achievements for the FDA’s Critical
Path Initiative (2009), which defines the “FDA’s strategy for driving innovation in the way its
16

regulated products are developed, evaluated, and manufactured,” no mention is made of the
harmonization of guidelines, scientific advice, or review procedures with other DRAs (FDA,
2009a). However, harmonized indication-specific clinical studies guidelines and scientific advice
mechanisms are arguably one of the most significant spheres in which DRAs have the power to
facilitate drug development and at the same time increase the safety of marketed products. At a
2011 Workshop on Strengthening a Workforce for Innovative Regulatory Science in Therapeutics
Development sponsored by the Institute of Medicine, Andrew Dahlem (Vice President and Chief
Operating Officer of Eli Lilly & Co.) cited harmonization of global regulatory expectations as a
critical need for the regulatory science process (Olson & Claiborne, 2012).
The harmonization of the regulatory process for therapeutics, the ‘holy grail’ of regulatory
science, is less a scientific than an administrative challenge, albeit a massive one related to the
ability of global regulators to collaborate to a degree that has not yet been accomplished. It
would be optimal if all DRAs shared a common set of expectations (at the level of the individual
drug class and indication) for the (1) requirements of clinical trial conduct, (2) the intended
labeling statements and claims those studies are designed to support, and (3) common procedures
and assumptions for assessing benefit:risk during product review. However, coming to such
agreements across constituencies involves overcoming significant political, cultural, competency,
methodological and administrative barriers, and will likely result from gradual, stepwise
developments over time leading to greater regulatory convergence. For instance, Hill and
Johnson note that
Harmonization within the EU took a number of years to develop to its current status: the first
European Pharmaceutical Directive (65/65/EEC) was issued in 1965, and it was not until the
1990s that effective methods for sharing regulatory processes and structures were really in place.
(Hill &Johnson, 2004)
17

Hill and Johnson also underscore the challenges of harmonization using the example of the joint
medicines regulatory authority between Australia and New Zealand.
The first steps in this process were taken in the early 1990s, and although it is hard to imagine
two countries with more similar philosophy and backgrounds, it is only in the past two to three
years that joint decision-making has really commenced (Hill &Johnson, 2004)
Harmonization of regulatory guidelines is recognized to be a highly costly and time consuming
endeavor (Hill & Johnson, 2004). However, the significant impact of regulatory dissonance on
the various stakeholders of drug review and approval standards is palpable, especially to the drug
sponsor, yet remains relatively unstudied in the regulatory science literature. In this manner, the
cost and burden of regulatory dissonance has been effectively shifted to industry without public
dialogue as to whether this policy is the most appropriate means to address regulatory dissonance.
As discussed in Chapter 2, progress is being made in the area of harmonized guidelines through
groups such as the International Conference on Harmonization of Technical Requirements for
Registration of Pharmaceuticals for Human Use, or ICH. However, the scope of these activities
to date has been focused primarily on broad, practical aspects of product registration,
circumventing the issue of harmonized indication-specific guidance. This is likely the result of
the fact that medical practice and standard of care can vary significantly across regions, making
this a challenging area to regulate.
Agents for the treatment and prevention of postmenopausal osteoporosis (PMO) provide one
model for studying the impact of divergent regulatory standards across regional constituencies,
and mechanisms to facilitate regulatory harmonization. Osteoporosis (stemming from the Greek
osteo and porosis, meaning bones and hole, respectively) is a systemic skeletal disease
characterized by a failure in the homeostasis of bone resorption and deposition, reflected
clinically by an increased risk of fragility fracture, which may occur following low impact
18

activities. Postmenopausal osteoporosis results from hormonal changes associated with
advancing age in women, with disease risk being influenced by genetic, nutritional and lifestyle
factors (Pharmaceutical Medicine, 2012). Accepted definitions of osteoporosis are generally
consistent with the following put forth by the World Health Organization:
A disease characterized by low bone mass and microarchitectural deterioration of bone tissue,
leading to enhanced bone fragility and a consequent increase in fracture risk. (WHO, 1998)
The WHO has also defined the diagnostic criteria for osteoporosis based on normal skeletal mass
values for the healthy population and defines osteoporosis as “a value for bone mineral 2.5
Standard Deviation (WHO, 1998) or more below the young adult mean (t-score ≤ -2.5),” and
severe osteoporosis as “a value for bone mineral 2.5 SD or more below the young adult mean in
the presence of one or more fragility fractures” (WHO, 1998). The societal burden of resultant
fracture-associated disability can be easily understood when considering the number of
individuals affected by the disease: 50% of all women over 50 will experience an osteoporosis-
related fracture leading to disability, which is often associated with reduced quality of life, and is
associated with increased mortality, especially in patients experiencing hip and vertebral fractures
(Levine, 2011).
Two main standards of regulatory guidance are available for PMO drugs: the FDA’s 1994 Draft
Guidance (FDA, 1994), and its European counterpart (CPMP/EWP/552/95 rev 2) which was
developed by the European Medicines Agency (EMA) and its Committee for Medicinal Product
for Human Use (CHMP) (EMA, 2007). Although these guidelines have several common
features, they differ in several notable ways (a policy analysis of these guidelines is provided in
Chapter 2). The FDA and EMA also differ significantly in their thinking about benefit:risk
analysis, as judged by regulatory review outcomes for novel investigational PMO agents. Six
novel osteoporosis therapies approved by EMA have not been approved subsequently by the
19

FDA. The converse is the exception, and currently all significant novel therapies marketed in the
US are authorized for distribution in the EU (under differing trade names in some cases).
The reasons behind these divergent drug review opinions is challenging to ascertain, because the
FDA, as well as other DRAs, rarely disclose the basis for negative drug review decisions to the
general public. In addition, the primary outcome data from studies leading to a particular
negative opinion may go unpublished, and if published, “selective reporting of data” (Rising,
2008), claimed by some researchers, may lead to a distorted picture of a regulatory outcome.
Thus, the circumstances surrounding negative review outcomes are often ambiguous. In addition
to these differences, considerable debate can be identified in the medical communities of the US
and EU around certain features of these guidelines, which impact the drug sponsor’s ability to
implement the clinical recommendations of these guidelines. In particular, some members of the
medical community have argued that the guidelines proposed by the DRAs have ethical problems
related to the use of placebo controls.
The purpose of regulatory guidance documents and standardized regulatory review procedures
are to ensure the safety and effectiveness of drug products. Indication-specific clinical
development guidelines also create a fair and level regulatory bar for industry, resulting in a more
predictable and controlled regulatory environment conducive to efficient drug development,
increasing the probability of success (POS) for a new therapeutic product. However, the value of
these guidelines is degraded when regulatory dissonance exists in the requirements, procedures
and approval standards across regions and between key stakeholder groups. With the basic
molecular mechanism by which an investigational agent works and the underlying nature of a
disease remaining relatively constant across regions, the decision on whether an investigational
agent may receive marketing authorization for use in a particular indication does change across
regions, and sometimes with extraordinary divergence. Either we must accept the limits to global
20

harmonization of regulatory standards by government agencies, or we must understand the
entropy that is constantly working against harmonized regulatory approval standards with the aim
of developing common inter-regional policies and methodologies to counteract its impact.
Understanding these costly differences of opinion is of critical importance to industry applicants.
1.2 Statement of the Problem
The criteria for the registration of therapeutic products often differ across various
national/regional DRAs. Indication- and/or disease-specific regulatory standards for product
registration remain largely unaligned. It is assumed that divergent indication-specific product
guidance and review outcomes, henceforth referenced as regulatory dissonance, adds
significantly to an unpredictable regulatory landscape, increasing the cost of developing novel
therapeutics with the downstream effect of increasing costs of approved therapies. The assumed
dampening effect of regulatory dissonance on innovation and product access for patients makes
this public health concern appear worthy of further investigation. We do not know, for example,
how additional development complexities and costs increase research and development
expenditures. We also do not understand whether these issues translate into an increase in drug
costs for patients, or have an upstream impact on the capital outlay of healthcare systems. We
also do not fully understand the ethical implications of enrolling extra study subjects into ever
larger and more complex clinical studies designed to address the multitude requirements of
DRAs. The primary way that we can gain more insight into the implications of regulatory
dissonance must come from the affected parties.
Regulatory dissonance can be defined as divergent regulatory guidance / recommendations /
standards set forth by regional drug regulatory authorities and medical committees responsible for
overseeing the drug development process (i.e., ethics committees, institutional review boards
(IRBs), study investigators). Thus, the term regulatory dissonance is not limited to potentially
21

divergent guidance/recommendations of regional DRAs, but is inclusive of the
guidance/recommendations provided by those stakeholders with oversight roles in clinical trial
conduct and drug development.
1.3 Purpose of the Study
This exploratory study aims first to inform the researcher about the sources of regulatory
dissonance from the standpoint of the stakeholders for a particular disease state, postmenopausal
osteoporosis. It then will use survey methods to examine the consequences of regulatory
dissonance on the drug sponsors who must comply with the regulatory requirements. Thus its
particular goals are to:
(1) Analyze the potential intrinsic and extrinsic determinants of regulatory dissonance by
evaluating the perspectives of the primary stakeholders affected by indication-specific
regulatory guidelines and review procedures,
(2) Gauge the impact and implications of regulatory dissonance (i.e., “downstream
consequences”) on the drug development process from the point of view of one of these
stakeholders, the drug sponsor, and,
(3) Discuss whether the study’s results are sufficiently robust and valid to inform on wider
policy initiatives which seek to develop new and/or improve existing regulatory
mechanisms / frameworks to attain greater harmonization of indication-specific guidance.
1.4 Importance of the Study
The results developed in this survey will give useful insight into aspects of global regulatory
dissonance that challenge the prospects of globally harmonized indication-specific clinical
guidelines and drug development procedures, with the hope that the results obtained from this
case study may be generalizable to policy endeavors whose impact and scope are applicable to
other disease areas. The study intends to illuminate areas in which greater harmonization is
possible, and thus seek practical solutions for more efficient and effective pathways for product
22

registration. The hope is that by studying these relationships, the results of this study will
contribute to a useful framework for policy makers to better understand regulatory dissonance.
This is a first step leading towards the creation of mechanisms to reduce its impact and improve
global harmonization of indication-specific drug development approaches.
The regulatory burden of divergent regulatory requirements is borne by all stakeholders in the
drug development process, including manufacturers, DRAs, and patients, the end users benefiting
from medicinal therapies. The ‘Great Recession’ (Rampell, 2009) and sovereign debt burden has
refocused public attention on high healthcare costs at a time when “the number of drugs that will
go off patent is increasing, which will cause companies to become even more cost-conscious”
(Olson & Claiborne, 2012). As a result, governments will be compelled to evaluate all available
mechanisms to reduce the cost-structures of health care administration, including the regulatory
dissonance inherent in today’s de-centralized drug development process. The harmonization
process has potential benefits to different stakeholders in this regard. From the standpoint of the
DRAs, better mechanisms of collaboration between agencies may increase data sharing, improve
pharmacovigilence and risk management activities, and reduce the associated administrative
burden to individual DRAs. From the standpoint of industry, it may lead to streamlined
protocols, fewer treatment arms, and fewer study subjects exposed to investigational agents.
Such harmonization might therefore lower drug development costs and time to market, and may
enhance safety for study subjects by reducing their exposure to investigational agents.
By understanding the roles of the primary stakeholders of indication specific regulatory
standards, one may also begin to analyze the areas of regulatory dissonance that cannot for one
reason or another be changed. For sponsors, this understanding could allow the company to
assess the impact of these factors early in the strategic planning phase. For the medical
community, it is important to understand how regulatory dissonance impacts the clinical
23

assessment of breakthrough products. For DRAs, an understanding of how individual regulatory
determinants contribute to regulatory dissonance will allow DRAs to target the most pressing
areas in need of greater harmonization, in order to facilitate drug development.
1.5 Limitations, Delimitations, Assumptions
This study has several real and potential limitations. First, the majority of survey participants are
regulatory or clinical scientists primarily working for large, multi-national pharmaceutical
companies in countries which participate in or recognize ICH activities, so that the collective
responses may be biased by the fact that they may represent only one view, that of industry.
Also, given that the primary researcher of this study is currently employed by a multi-national
biopharmaceutical company, experimenter expectancy is a potential source of bias.
DRAs, the medical community and drug sponsors may hold different views on certain questions
explored within this work. Although regulatory professionals interact directly with DRAs (FDA,
EMA, Health Canada, PMDA, CFDA, etc.) and clinical development scientists interact directly
with investigational review board (IRB) and ethic committee (EC) members, direct feedback is
not sought in this study from regulators and/or members of the medical community involved in
drug development (i.e., study investigators, IRB/EC members, health organizations).
Some respondents might be hesitant to share opinions different than those of their respective
organizations, since many respondents who are qualified to share their views may be constrained
by the confidential nature of their activities and the types of information that they are able to
share. Therefore, respondents will be informed that all data will be pooled, and if specific
responses are quoted, will not be attributed to an individual. However, these safeguards may not
be sufficient to allay concerns regarding anonymity and thus may impact the nature of the
responses. The geographic delimitation for a majority of potential respondents (European Union
24

and United States) has the potential to bias responses towards the views of the pharmaceutical
industry in Western nations or regions participating in ICH. To minimize this effect, respondents
from other regions (Australia, Canada, China, Japan, etc.) will be actively sought out during the
recruitment phase of this exploratory study.
An assumption that might bias survey results is that medical standards of care and notions of
accepted medical judgment are difficult areas to align and harmonize, thus making indication-
specific guidance debatably the most challenging area to harmonize. This survey was delimited
to questions relating to the specific case of a single indication (PMO) in order to control certain
confounding variables. The result is a reduction in disparities that might be seen if views were
obtained from respondents working across different therapeutic indications, because their
particular problems with dissonance might vary from that faced by those developing PMO drugs.
As a result however, the ‘generalizability’ of the results of this survey may not hold for other
indications/disease states.
The results of this survey may be influenced by the fact that the survey pool is small and many
participants may be difficult to recruit or may not be known at the time of initial survey
dissemination. Thus, it may become necessary to recruit additional participants through a process
of “referral” or “snowball” sampling, which itself may introduce bias into the study since it may
lead to the recruitment of like-minded surveyees. The results may also be influenced by the
narrow time window of the trial, and thus may be highly influenced by contemporary regulatory
events, especially those involving the FDA, EMA, MHRA and Health Canada, and ongoing
scientific debate at the time of the survey. Examples of such regulatory events include product
approvals, issuance of new regulatory authority guidance, or specific feedback given by DRAs on
the products managed by the surveyees. The impact of relevant events that occur after research is
underway will be discussed in the results chapter of this thesis.
25

The focus of the study will be delimited to an evaluation of regulatory dissonance in the clinical
development of PMO agents, and will only peripherally consider the nonclinical aspects of
developing PMO agents. Thus it will summarize important areas of nonclinical regulatory
requirements only briefly for benefit of the reader, and as they relate for product registration. In
addition, although most therapeutics appropriate for PMO have other potential uses, this thesis
will focus solely on the harmonization of guidance/guidelines assisting industry in the
development of products for the indication of primary or postmenopausal osteoporosis. Thus,
this thesis will not include specific evaluation of dissonance in other on-/ off-label uses of these
agents in other indications. Such indications include 1) glucocorticoid-induced osteoporosis
(GIOP) or other drug-induced osteoporosis (e.g., hormone ablation therapies such as aromatase
inhibitors or anti-androgen therapies) (Mazziotti, 2010), 2) male osteoporosis, 3) Paget’s disease,
4) heterotopic ossification, 4) prevention of skeletal-related events (SREs) from solid tumors, 5)
treatment of bone metastases in patients with advanced breast cancer (Stopeck, 2010), and 6)
treatment of other cancer-related conditions (Lewiecki & Bilezikian, 2012) such as castration-
resistant prostate cancer (Fizazi, 2011). Agents approved for the treatment of postmenopausal
osteoporosis are also being considered in orphan drug indications such as giant-cell tumor of the
bone (Thomas, 2010) or osteogenesis imperfecta. For approved anabolic agents such as
teriparadide (Forteo/Forsteo), whose primary mode of action results in significant increases in
bone mass, acceleration of fracture healing in specific fracture types is being investigated, a
domain once reserved for devices and device/drug combination products (Dent, 2012).
26

1.6 Definitions
Abbreviation or Term Definition
ANZTPA Australia New Zealand Therapeutic Products Agency
ASBMR American Society for Bone and Mineral Research
AUC Area Under the Curve
BRF Benefit:risk Review Framework
BTM Bone Turnover Marker
CDE
Chinese Center for Drug Evaluation
CFDA China Food and Drug Administration (People’s Republic of China, or
PRC), formerly SFDA
CHMP Committee for Medicinal Products for Human Use
CIOMS Council for International Organizations of Medical Sciences
DABA Dual Acting Bone Agent
DG ENTR EU Directorate-General for Enterprise and Industry
DoH Declaration of Helsinki
DRA Drug Regulatory Authority
EC Ethics Committee
EFPIA European Federation of Pharmaceutical Industries Association
eMC Medicines Compendium (MHRA)
EMEA European Agency for the Evaluation of Medicinal Products
EMA European Medicines Agency
EPAR European Public Assessment Report
FDA United States Food and Drug Administration
FDAAA Food and Drug Administration Amendments Act of 2007
FRAX WHO Fracture Risk Assessment Tool
FR Federal Register
GCP Good Clinical Practice or Global Harmonization Group
GAO United States Government Accountability Office
GMP Good Manufacturing Practice
HC Health Canada
HRT
Hormone Replacement Therapy
ICH
International Conference on Harmonization of Technical Requirements
for Registration of Pharmaceuticals for Human Use
IDRAC International Drug Regulatory Affairs Compendium (global regulatory
database)
IOF International Osteoporosis Foundation
IRB Investigational Review Board
ISCD International Society for Clinical Densitometry
27

Abbreviation or Term Definition
JBMR Journal of Bone and Mineral Research
MHRA Medicines and Healthcare products Regulatory Agency (United
Kingdom)
MOA Mechanism of Action
MRA Mutual Recognition Agreement
MRCT Multi-regional Clinical Trial
NEJM New England Journal of Medicine
NIH National Institutes of Health
NME/NCE New Molecular Entity / New Chemical Entity
NOF National Osteoporosis Foundation
OHSRP Office of Human Subject Recruitment and Protection
OECD Organization for Economic Cooperation and Development
OIRA US Federal Office of Information and Regulatory Affairs
OTA Office of Technology Assessment, U.S. Congress
PANDRH Pan American Network on Drug Regulatory Harmonisation
PHP Primary Healthcare Provider
PMDA Japanese Pharmaceuticals and Medical Devices
PMO Postmenopausal Osteoporosis
PMSB/ELD Pharmaceutical Medical Safety Bureau/Evaluation and Licensing
Division (Japan)
PhRMA Pharmaceutical Research and Manufacturers of America
POS Probability of Success
PORS Probability of Regulatory Success
PTH Parathyroid Hormone
RANKL Receptor activator of nuclear factor kappa-B ligand
REMS Risk Evaluation and Minimization Strategy
RHI Regional Harmonization Initiative
ROW Rest-of-World (i.e., outside the United States)
SBA Summary Basis of Approval
SBD Summary Basis of Decision (Health Canada)
SD Standard Deviation
SERM Selective Estrogen Receptor Modulator
TGA Therapeutic Goods Administration
UMBRA Unified Methodologies for Benefit-Risk Assessment
VHP Voluntary Harmonisation Procedure
WHO World Health Organization

28

1.7 Organization of the Study
This research study has five main components. Chapter 1 summarizes and defines the issue to be
studied and the limitations of the research approach. Chapter 2 summarizes the relevant
academic and regulatory literature that frames the problem in sufficient detail to inform on the
exact nature of the issue. Chapter 3 describes the methodology used to investigate and analyze
the problem, including a description of the survey development, plans for its implementation, and
a data analysis plan. Chapter 4 will include an analysis of the resulting survey data and treatment
of information from the individual subject interviews, and Chapter 5 will conclude the study with
a high-level discussion of these data, including analysis of the relevance and generalizability of
the results beyond the specific area of study.
29

CHAPTER 2: LITERATURE REVIEW
2.1 Introduction
For most of the established history of regulatory science, regulations governing the safety and
efficacy of drugs were established locally for the citizens of a single country. A regional US-
centered framework was appropriate when most US-manufactured drugs were distributed and
sold only within the US market. Today the footprint of drug distribution has become multi-
national, but the regulatory frameworks that currently exist are still largely national or regional in
focus. As a result, drug manufacturers are faced with a multitude of regulatory requirements that
may be complementary, additive, or contradictory. No governing body or committee has the
authority to resolve such dissonance at the global level. Thus the individual drug manufacturer,
with its imperfect visibility into regulatory authority decision making, must apprehensively
occupy that role.
Prior to launching a new molecular or chemical entity (NME or NCE) into commercial
distribution, its manufacturer typically must submit simultaneous applications for marketing
authorization to assure its entry into multiple markets. The need to respect diverse country-
specific laws, regulations, standards and cultural attitudes must be recognized in order to achieve
seamless implementation. Sponsors must integrate often divergent guidelines and opinions from
different DRAs in order to develop a unified drug development strategy that is acceptable to the
DRAs in different constituencies. In so doing, the individual drug sponsor is in the unique
position to appreciate the often divergent filing requirements of each constituency. However, the
regulatory knowledge and experience gained by industry stakeholders during the drug review
process is often kept confidential, as this competitive intelligence aids industry rivals.
30

To understand the differences between constituencies, information is available in three types of
literature, (1) existing regulatory guidances / standards written by DRAs and officially recognized
consensus bodies, (2) drug approval precedents outlined in summary basis of approval
documentation, which clarify regulatory thinking, and (3) academic and trade articles written by
clinical practitioners, primarily in the private sector.
2.2 Regulatory Dissonance in Clinical Trials Guidelines
2.2.1 Regulatory Guidelines/Standards for Postmenopausal Osteoporosis
To understand the challenges of setting consistent approval standards for a particular
indication/disease state, one must understand the state of science with respect to the available
classes of agents that can be used to treat the condition. However, knowing current science is
often not enough to predict how regulators will approach certain products when they are
submitted for marketing authorization. Regulations and guidance documents typically lag
science, because new science and new drugs can raise issues not anticipated by prior review
experience. Thus regulators can find it difficult to apply consistent rules that are applicable
across a range of products with different mechanisms of action (MOAs). A multitude of drugs
are currently marketed in PMO globally (Table 1). As such, these precedents serve as models of
potential regulatory pathways for PMO agents currently in development (Table 2). PMO agents
belong to one of two main classes that act on the bone either as bone ‘catabolic’ agents, which
reduce the resorption of bone (i.e., anti-resorptives), or bone ‘anabolic’ agents (i.e., osteoanabolic
agents). Additionally, a few compounds are purported to act in both ways (e.g., strontium
ranelate). Specific classes of marketed PMO agents include hormone replacement therapy
(HRT), such as second-line estrogens and selective estrogen receptor modulators (SERM),
hormone analogues (intact or recombinant parathyroid hormone and calcitonin-containing
agents), strontium salts of maleate or ranelic acid, and first-line anti-resorptive drugs, which may
31

be either small molecule agents or monoclonal antibodies (Chen & Sambrook, 2012). Recently,
an ever increasing number of generic small molecule entrants of currently approved anti-
resorptive drugs have been marketed (alendronate, zoledronic acid; not listed in Table 1). Of
these classes, antiresorptive agents such as first-in-class agent Fosamax (alendronate), as well as
others in this class such Actonel (risedronate), Boniva (ibandronate), Prolia (denosumab), Reclast
(zoledronic acid), or the parathyroid hormone analog Forteo (teriparatide, 34 N-terminal amino
acids), are the most widely used agents for this condition and are considered the current standard
of care (Chen & Sambrook, 2012). It also appears that these agents are generally the most
effective in reducing future risk of fracture. Given the scarcity of direct head-to-head data for
available PMO agents (explained in Chapter 2.4.1), a systematic review of randomized,
controlled clinical studies and large observational studies (Comparative Effectiveness Report No.
53) was conducted by Southern California Evidence-based Practice Center (RAND Corporation)
for the US Department of Health and Human Services. The report suggests that the anti-
resorptives noted above, with the exception of Boniva, are the most effective therapies available
for reducing the risk of both vertebral and nonvertebral fractures in postmenopausal women with
osteoporosis. The report also concludes that these same products (sans Forteo and Boniva) are
also effective in preventing hip fractures in women with postmenopausal osteoporosis (Crandall,
2012), but that Boniva (ibandronate), another anti-resorptive, and raloxifene, a SERM, are only
effective in reducing vertebral fracture risk (Crandall, 2012). Strontium ranelate is also a
standard treatment, but is only available outside of the US (Kanis, 2013). The reasons for this are
discussed in Chapter 2.3.2.
32

Table 1: Significant Novel Therapies in Postmenopausal Osteoporosis (US versus EU)
United States European Union
Tradename (chemical or
molecular name)
Year Approved
(Withdrawn
from Market)
Tradename (chemical or
molecular name)
Year Approved
(Withdrawn
from Market)
Calcimar
(salmon calcitonin)
1

1984 (1999 &
2007)
A

Calcimar
(salmon calcitonin)
1

1999 (2012)
A

Premarin/Prempro/Premphase
2
1986/2003
A
Premarin
2
1988
A

No US Equivalent Not FDA
Approved
Didronel (etidronate)
3
1987
B

No US Equivalent Not FDA
Approved
Livial (tibolone)
2
1987
B

Miacalcin Nasal Spray
(synthetic salmon calcitonin)
1

1995
A
Miacalcin Nasal Spray
(synthetic salmon
1 1
1995 (2012)
A

Fosamax (alendronate sodium)
3
1995
A,B
Fosamax (alendronate)
3

di )
3
1995
A,B

Evista (raloxifene HCl)
4
1997
A
, 1999
B
Evista (ralxifene HCl)
4
1997
A
, 1999
B

Actonel
(risedronate sodium)
3

1999
A,B
Actonel
(risedronate sodium)
3

1999
A,B

Forteo (teriparatide)
5
2002
A
Forsteo (teriparatide)
5
2003
A

Boniva (ibandronate sodium)
3
2003
A,B
Bonviva (ibandronate)
3

3
2004
A,B

No US Equivalent Not FDA
Approved
Protelos/Osseor
(strontium ranelate)
5

2004
A

No US Equivalent Not FDA
Approved
Preotact
(intact parathyroid
h)
6
2006
A

Reclast (zoledronic acid)
3
2007
A,B
Aclasta (zoledronic acid)
3
2005
A

No US Equivalent Not FDA
Approved
Conbriza (bazedoxifene)
4
2009
A

No US Equivalent Not FDA
Approved
Fablyn
(lasofoxifene tartrate)
4

2009 (2012)
A

Prolia (denosumab)
8
2010
A
Prolia (denosumab)
8
2010
A

A
PMO Treatment /
B
PMO Prevention
1
Recombinant calcitonin, calcitonin-containing agents (antiresorptive agent). Note: marketing
authorization recently withdrawn in EU (2012)
2
Hormone replacement therapy (conjugated/equine or synthetic estrogen)
3
Bisphosphonate (antiresorptive agent)
4
Selective Estrogen Receptor Modulator (SERM)
5
Recombinant Parathyroid Hormone (PTH 1-34))\
6
Dual mechanism of action: antiresorptive and anabolic bone formation
7
Intact Parathyroid Hormone (PTH 1-84)
8
Receptor activator of nuclear factor kappa-B ligand (RANKL) inhibitor
33

Table 2. Key Therapeutic Agents in Development for Postmenopausal Osteoporosis
A,B

Chemical or Molecular
name / Laboratory
Name
Drug Sponsor Current Development
Phase / Expected
Primary Completion
ClinicalTrials.gov
or EudraCT
Identifier
c

NB S101
1
Osteologix II / August 2007 NCT00409032
Aprela (BZA/CE)
2
Pfizer (Wyeth) III / September 2008
Submitted July 2012 to
EMA, December 2012 to
FDA
NCT00242710
ACE-011(sotatercept,
ActRIIA-IgG1)
3

Acceleron Pharma I / November 2008 NCT00709540
DP 001 (2MD)
(Vitamin D
3
stimulant)
Deltanoid
Pharmaceuticals
II / December 2008 NCT00715676
Cholecalciferol/strontium
ranelate/vitamin D3
(S-06911)
Servier III / March 2010
(First Enrollment)
EUCTR2009-
0145270-18-PL
ONO-5334
4
Ono II / July 2010 NCT00532337
MK5442
5
Merck II / June 2011 NCT00996801
SMC021
6
Nordic Biosciences
/ Novartis
III / August 2011 NCT 00525798
Oral PTH Analogue
7
Unigene /
GlaxoSmithKline
II / October 2011 ableNCT01321723
AMG 167
8
UCB/Amgen I / February 2012 NCT01101048
Blosozumab (LY2541546)
8
Lilly II / May 2012 NCT01144377
RN564
9
Pfizer I / May 2012 NCT01293487
ZT-034
7
Zelos Therapeutics
Inc.
II / July 2012 NCT01604057
Odanacatib (MK0822)
4
Merck III / November 2012 NCT00529373
BPS804
9
Novartis II / January 2013 NCT01406548
BA058
7
Radius / Novartis III / October 2014 NCT01343004
Romosozumab
(CDP 7851, AMG 785)
8

UCB / Amgen III / October 2015 (Placebo)
III / July 2015 (Active)
NCT01575834
NCT01631214
A
Data from ClinicalTrials.gov and the EU Clinical Trials Register (EudraCT).
B
Drugs in Clinical Development for Osteoporosis – Summary and Table (Pharmaceutical Medicine, 2012)
c
NCT No. allows data search of study described in column 3 in ClinicalTrials.gov registry
1
Strontium Malonate
2
Bazedoxifene/conjugated estrogen combination
3
Osteoblast inhibitor, osteoclast stimulants
4
Cathepsin K inhibitor (second generation antiresorptive agent)
5
Calcilytic agent (calcium-sensing receptor antagonist)
6
Oral calcitonin (antiresorptive agent)
7
Analogue of parathyroid hormone (anabolic bone forming agent)
8
Sclerostin antibody (anabolic bone forming agent)
9
Anti-DKK1 monoclonal antibody (inhibitor of Wnt signaling)

34

Publicly available regulatory authority guidance documents provide an efficient vehicle for
communicating a DRA's current thinking in order to provide consistent, transparent, scientifically
valid, and peer-reviewed regulatory recommendations for a particular class of drugs or disease
indication. In PMO, there are only two main indication-specific guidance documents that have
been primarily responsible for providing the necessary guidelines for developing the novel and
marketed treatments described above:
• Guidelines for Preclinical and Clinical Evaluation of Agents Used in the
Prevention or Treatment of Postmenopausal Osteoporosis (Draft, Withdrawn)
(FDA, 1994)
• Guideline on the Evaluation of Medicinal Products in the Treatment of Primary
Osteoporosis (CPMP/EWP/552/95 rev 2) (EMEA, 2007)
These guidance documents establish a minimum set of requirements for drug registration, in
essence establishing a level playing field for competing manufacturers. Regulatory guidance is
typically non-binding for either the drug sponsor or regulatory agency; deviations from the
recommended approaches may be proposed with adequate scientific and regulatory justification.
Further clarification of alternative approaches may be learned from case/precedent information in
the literature (see Chapter 2.3).
2.2.2 Background on United States Guidance
An understanding of the regulatory guidance governing postmenopausal osteoporosis (PMO)
drugs can perhaps best be achieved by reviewing the FDA’s Draft PMO Guidance, that is the
most established guidance document and was first published in 1979 (FDA, 1994). The 1994
Draft FDA Guidance document is still in draft form. In fact, it was never finalized and was
withdrawn in 2009 (FDA, 2009b). However it is generally recognized as the most up to date
reflection of US FDA thinking (Storm, personal communication). The evolution of this guidance
is summarized in a paper titled “The Food and Drug Administration Osteoporosis Guidance
35

Document: Past, Present, and Future,” written by the former Deputy Director of FDA’s Division
for Metabolic and Endocrinology Products (DMEP), Dr. Eric Colman (Colman, 2003). This
review provides a unique perspective from an FDA insider on the revisions over time to this long-
standing guidance document (Colman, 2003), and describes how FDA’s registration requirements
have evolved substantially alongside clinical trial precedence.
The recommendations in the FDA’s 1994 Draft PMO Guidance were established at a time when
most drugs in development or approved agents were either anti-resorptives, estrogens or estrogen
receptor modulators (SERMs). Thus the FDA’s guidelines were influenced by the unique
biological characteristics of this first generation of PMO agents. For newer investigational agents
whose mechanism of action and treatment effects may differ from those of approved precedents,
different approaches to clinical evaluation may require alternative ‘outside the box’ solutions.
For instance, one might consider assessing osteoanabolic agents over shorter treatment durations,
when these products are expected to exert their greatest osteoanabolic effects, thus limiting
exposure to a narrower therapeutic window when the benefit:risk is expected to be most optimal.
The original FDA PMO Guidance (1979) responded to a public health “need for effective and
safe drugs to treat osteoporosis,” noting the “special challenges” of clinical trial design in this
indication (Colman, 2003). The most important questions addressed by this guidance involved
the challenges with (1) the quantitative assessment of skeletal bone in vivo, (2) the size and
duration of treatment needed to demonstrate a drugs’ effect, and (3) scientific uncertainty
associated with correlating bone mass measurements and fracture risk (Colman, 2003; Crandall,
2012). To address the state of skeletal bone, the primary efficacy outcome of interest, the 1979
FDA PMO Guidance identified skeletal mass (i.e., bone mineral density, or BMD), but did not
recommend a particular imaging technique for this assessment among several available at the
time (Colman, 2003). The incidence of fragility fracture was clearly recognized as a clinically
36

relevant outcome for treatment, but was not identified as a required study endpoint for drug
registration in the FDA’s1979 PMO Guidance. It was understood that a spontaneously occurring,
event-driven endpoint such as fragility fracture would “require a relatively large number of
patients to provide statistically significant results” (Colman, 2003; Crandall, 2012). The result
would have been a significant increase in the cost and logistical challenges of conducting clinical
studies, thus it was considered that a requirement to assess fracture risk reduction would impede
progress towards innovative therapies. In fact, reliance upon skeletal mass alone to assess
efficacy made sense at the time because it was considered to be an appropriate surrogate marker
of fracture risk reduction. Thus, the FDA 1979 PMO Guidance made a critical compromise with
industry by stating that “where there is evidence that bone formed during therapy is normal,
adequate and well-controlled studies showing a favorable effect on bone mass will provide
reasonable evidence of effectiveness of the drug in the management of osteoporosis” (Colman,
2003). To address the second issue of duration of treatment, the guidance recommended that the
pivotal phase 3 study supporting product registration be an extension of the phase 2 study, which
was itself required to be 24 months in duration (Colman, 2003). A recommendation for the
duration of the phase 3 extension was not specified. The phase 2/phase 3 extension study was
recommended to be of a randomized, placebo-controlled design (Colman, 2003).
Calcimar (calcitonin) was approved in 1984 under these original FDA guidelines relying upon
available bone mass data from the pivotal clinical study: a combination of total body calcium
(TBC) and bone mineral content data (Colman, 2002). Despite lingering questions surrounding
the lack of evidence of fracture reduction, Calcimar was approved with an agreement to further
evaluate fracture efficacy in a phase 4 postmarketing commitment study (Colman, 2002), an
approach consistent with this earlier version of the FDA’s PMO guidance.
37

A second version of the FDA’s PMO Guidance came into effect in 1984, with relatively minor
changes. New to the guidance was an added option for sponsors to register an agent based on
evidence of clinical effectiveness from an active-controlled pivotal study (given that an approved
PMO agent, salmon calcitonin, was now on the market with which to compare), and recognized
prevention of osteoporosis as a separate approvable indication from that of treatment of
osteoporosis (Colman, 2003; Crandall, 2012). In addition, the 1984 revision added a requirement
for add-on supplemental vitamin D and calcium for all study subjects, and acknowledged the
validity of using a new technique, dual-energy photon absorptiometry (DPA), for assessing BMD
(Colman, 2003; Crandall, 2012).
An assumption set forth in the guidance documents of 1979 and 1984, that increased BMD was
an adequate surrogate marker for the clinically relevant endpoint of fracture risk reduction, was
critically challenged as a result of emerging clinical experience with two novel PMO drugs,
sodium fluoride and cyclic etidronate disodium, that were developed in the 1980s. The results of
a four-year phase 3, placebo-controlled study in 202 postmenopausal women evaluating the
safety and efficacy of sodium fluoride showed a 35% increase in BMD in the sodium fluoride
treated group vs. the placebo treated group, but oddly the number of new vertebral fractures
observed in the two groups remained statistically similar, and a statistically significant increase in
non-vertebral fracture was observed in the treatment group (Riggs, 1990; Colman, 2003;). Later,
in a separate phase 3 study with disodium etidronate (Didronel), the FDA’s analyses showed that
significant increases in BMD in the treatment arm were oddly coincident with an increased rate of
in new vertebral fracture during year three in patients receiving the active treatment (Colman,
2003). These results made it plain that “a pharmacologically induced increase in bone mass did
not necessarily equate with reduced fracture risk” (Colman, 2003), and forced a reevaluation of
1979 and 1984 paradigm for registering PMO agents in the US. As a result of these two datasets,
38

the FDA concluded that (1) the efficacy of novel osteoporosis drugs could not be assessed
reliably without a more rigorous assessment of the incidence of vertebral fracture and (2) the
duration of pivotal phase 3 studies must change; long-term fracture efficacy and safety was now
recommended after three years (Colman, 2003). These two significant realizations set the stage
for the FDA’s 1994 revision to the PMO guidance. Thus, PMO agents that were already
approved at the time of FDA’s 1994 Guidance revision were not immediately required to
retroactively reassess their products under these newer, more rigorous regulatory requirements
(i.e., salmon-calcitonin, estrogens).
FDA’s 1994 Draft Guidance document, titled Guidelines for Preclinical and Clinical Evaluation
of Agents Used in the Prevention or Treatment of Postmenopausal Osteoporosis, marked a major
shift in the development of regulatory thought with respect to regulation of PMO drugs (FDA,
1994). It outlined both preclinical and clinical considerations upon which marketing
authorization has been based for all PMO agents on the US market today, and called for several
significant new requirements. First, the FDA now required that long-term statistically and
clinically significant improvements in bone mineral density (BMD) be maintained following a
required three year duration of continuous treatment in a single, randomized, double-blind study
(placebo or active-controlled) (FDA, 1994). In addition, at least “a trend (p<0.2) toward
decreased fracture incidence and no deterioration in the third year” must be demonstrated (FDA,
1994). Second, the assessment of BMD was now relegated to a secondary endpoint for novel,
non-estrogen agents since BMD was no longer considered by the FDA as a reliable marker of
effectiveness for novel agents (for estrogens, BMD was still considered an appropriate primary
endpoint given epidemiologic evidence “that estrogen therapy reduces the risk of vertebral and
non-vertebral") (FDA, 1994). Third, assessments of bone quality were also needed in addition to
demonstrations of anti-fracture efficacy and improvements of BMD at year three. Bone quality,
39

defined as an appraisal of bone architecture, mass and strength, is ethically and logistically
problematic since it may only be assessed through the histological evaluation of tissue sections
obtained through painful and invasive biopsy procedures. Moreover, sampled skeletal sites vary
in location in the same subject since no single site can be biopsied twice (FDA, 1994), increasing
the variability in data. Due to these limitations, the FDA’s guidance recommends that biopsies be
required in only a small subset of study participants (FDA, 1994), and at finite time points
(typically paired biopsies, once prior to treatment and at the end of 3 years treatment). The
histomorphometrical data from these samples must reveal no abnormality of bone (FDA, 1994).
In addition, clinical histomorphometric data must be supplemented with nonclinical assessments
of bone quality. In addition to new requirements described above for an indication of treatment
of osteoporosis, the FDA’s 1994 Draft Guidance document proposed that an indication of
prevention of osteoporosis could be supported by an assessment of BMD at the end of 24 months
of continuous treatment in a randomized, placebo-controlled study, with the caveat that the
product has already obtained approval by the FDA for treatment of osteoporosis (thus, allowing
fracture efficacy data to be bridged).
Shortly following the issuance of the FDA’s 1994 Draft Guidance document, a significant new
guidance was issued in 1998 that effectively required two-adequate and well-controlled studies as
the evidentiary standard for most drug and biological product registrations in the United States
(FDA, 1998). Ethical exceptions to this rule, where a single pivotal trial would be allowed,
would include only those conditions whose endpoints evaluate “a clinically meaningful effect on
mortality, irreversible morbidity, or prevention of a disease with potentially serious outcome”
(FDA, 1998). This 1998 FDA guidance cites the prevention of “osteoporotic fractures” as one
such endpoint where this ethical exception applies, due in large part to newly available anti-
resorptive therapy with demonstrated anti-fracture efficacy (alendronate was approved in 1995).
40

It is clear that the FDA, as early as 1998, understood the developing ethical quandary of
conducting placebo-controlled studies, with a primary endpoint of osteoporotic fracture, in a
marketplace where effective treatments were now widely available to patients (Chapter 2.8,
Ethical Controversies in PMO Drug Approval Standards).
The FDA’s 1994 Draft Guidance also called for parallel nonclinical assessments of bone quality
in vivo to corroborate clinical evidence and to bridge this potential gap in understanding of a
drug’s effects on bone quality, because the clinical assessment of bone quality and assumed
variability provided limited information on its own. Nonclinical studies of bone quality were
required in two species: the adult overiectomized rat, which mimics estrogen deficiency
following menopause in humans, and a higher order remodeling species to mimic the bone
biology of humans (FDA, 1994). These studies must also be carried out for a duration based
upon the rate of bone turnover of the species (approximately 16 months in the rat) and must
include two dose levels (one equivalent to an optimal response, and a high dose five multiples
above the optimal response dose level) (FDA, 1994). Although these studies are not considered
exact models of the condition in humans, the FDA does consider the results informative (FDA,
1994).
If abnormal bone quality is observed in these studies despite a demonstration of fracture efficacy
at three years, the FDA’s 1994 Draft Guidance states that the pivotal clinical study should be
continued for at least two more years as an open label extension study post-approval (i.e, in phase
4 ). However, if fracture efficacy is not proven at 3 years and abnormal bone quality is observed
in the nonclinical studies, the product will not be approved under any circumstance. These
circumstances are illustrated in Table 3 below, which is excerpted from the FDA’s 1994 Draft
Guidance (FDA, 1994).
41

Table 3. Guide to FDA Action – Clinical vs. Nonclinical Study Outcomes
1

Clinical
Studies
Preclinical Studies (2 Species)
Increased
BMD
Normal Abnormal
Fracture
Efficacy
Proven at 3
years
 Approved Drug for Marketing
 Phase IV Continuation Not
Needed
Approve Drug for Marketing
 Continue Phase IV Open Studies for at
Least 2 Years
Not Proven
Fracture
Efficacy at 3
Years
 Approve Drug for Marketing
 Continue Controlled Studies for 2
More Years
 NO DRUG APPROVAL
 Continue Controlled Study for 5 Years to
Determine Safety and Fracture Efficacy
1
Excerpted from FDA, 1994
2.2.3 Background on European Union Guidelines
The history of guidelines for PMO agents is shorter in the European Union (EU) than the US,
coinciding with the relatively younger age of a pan-European regulatory agency. The European
Medicines Agency, which was originally known as the European Agency for the Evaluation of
Medicinal Products, or EMEA, was established in 1995 (a name change from EMEA to EMA
occurred in December 2009). The EMEA’s PMO guidance, The Guideline on the Evaluation of
Medicinal Products in the Treatment of Primary Osteoporosis (CPMP/EWP/552/95; Rev 2), was
released shortly thereafter in 1997 under its original title of Note for Guidance on Involutional
Osteoporosis in Women (EMEA, 1997) by the Committee for Medicinal Products for Human
Use, or CPMP, the committee responsible for issuing the EMA’s opinions on medicinal products
intended for human use. Of note, the original 1997 CPMP (later CHMP) guidance was issued in
the same general time period as the FDA’s currently accepted 1994 Draft Guidance.
The original CPMP Guideline was in many respects consistent with the FDA’s 1994 Draft
Guidance. It recommended that the rate of new axial [i.e.,vertebral] or peripheral fractures as the
primary efficacy variable for the treatment of osteoporosis (EMEA, 1997). The original CPMP
42

Guideline viewed BMD assessment as important secondary endpoint variable, stating that “BMD
cannot serve as a satisfactory surrogate end-point for the documentation of clinical relevant
efficacy” (EMEA, 1997), which is consistent with the FDA’s position on BMD surrogacy. The
original CPMP Guideline also recommended that randomized treatment should be conducted for
at least three years or more, and recognized separate indications for treatment and prevention of
postmenopausal osteoporosis, points aligned with FDA’s 1994 Draft Guidance. In addition, the
original 1997 guidance strongly recommended bone histomorphometric data to rule out
significant toxicity (rather than to assess bone quality), but did not specify the relative numbers of
biopsies required to satisfy the requirement (EMEA, 1997).
Thereafter, Revision 1 of the CHMP Guideline (EMEA, 2001) was adopted in 2001. The main
impact of this change was that applicants would be requested to assess the effect of the
investigated drug on the rates of both spinal and femoral (not all non-vertebral) fractures (EMEA,
2001). The requirement to power and design a study to demonstrate a statistical reduction in hip
fractures, in addition to vertebral fractures, was deemed burdensome by some sponsors, including
the makers of strontium ranelate, since “a placebo-controlled study based on rare events such as
hip fractures as a primary criterion would have exposed a much larger population to the test
product while the safety and efficacy of the tested product in an aged population was not totally
established” (Servier, 2004).
Within a few years, several of the precepts of Revision 1 were no longer considered current
(Reginster, 2006). Based on the development of agents for osteoporosis with novel properties,
combined with new methods for identifying patients at risk of fracture, a workshop coordinated
by the Group for the Respect of Ethics and Excellence in Science (GREES) concluded that
several areas of the Version 1 guidance should be reconsidered. These included:
43

(1) Indication being granted to new chemical entities (treatment of osteoporosis in women at high
risk of fracture instead of prevention and treatment of osteoporosis),
(2) The requirements of showing an anti-fracture efficacy on all or on major nonvertebral
fractures (instead of the hip),
(3) The duration of pivotal trials (2 years instead of 3) and
(4) The possibility of considering bridging studies for new routes of administration, new doses or
new regimens of previously approved drugs (Reginster, 2006)
EMEA’s 2007 Revision 2 Guideline addressed these and other concerns raised during the
consultation period. The duration of randomized, pivotal phase 3 studies was shortened to 2
years (EMEA, 2007) because experience with several agents seemed to indicate that 2 years was
generally appropriate to assess long-term safety and efficacy of novel agents (EMEA, 2006), and
it was thought that animal data could adequately predict potential risks to bone quality (Reginster,
2006). However, with long-term safety in mind, the EMEA Revision 2 Guideline also require
data showing that fracture prevention be maintained after 3-5 years; these data could be submitted
after marketing authorization.
In addition, the recommendations for the primary efficacy variable were again adjusted. EMEA
now required that effect be demonstrated based on the incidence of patients with new fractures,
including (1) clinical or morphometric vertebral fractures and (2) hip or major non-vertebral
fractures (i.e., those of the pelvis, distal femur, proximal tibia, ribs, proximal humerus forearm,
and hip) (Ormarsdottir, 2008). Thus, a key innovation introduced by the 2007 EMEA Revision 2
Guideline was its “recognition of the importance of non-vertebral fracture other than hip fractures
and the use of major non-vertebral fractures (including hip) as a primary end-point” (EMEA,
2006). This new guidance gave sponsors the option to eliminate the prolonged evaluation of hip
fracture.
44

Importantly, Revision 2 also removed the distinction between the prevention and treatment of
osteoporosis, thus removing the option of a separate prevention indication and creating a single
indication for the treatment of osteoporosis in men at increased risk of fracture (Ormarsdottir,
2008; EMEA, 2007). Although this change was consistent with the recommendations made by
GREES, some drug sponsors disputed this recommendation since prevention of disease was
viewed to be a valid aim of treatment (EMEA, 2006). In addition, Revision 2 added “absolute
fracture risks as a criterion for subject entry to studies,” recognizing “that factors other than BMD
contribute to fracture risk” (EMEA, 2007). As result, sponsors must justify the inclusion criteria
for the clinical study population on the basis of independent risk factors including “age, BMD,
prior fractures, a family history of hip fracture, high bone turnover, low body mass index, current
tobacco use, and alcohol abuse” (Ormarsdottir, 2008). The standard method for establishing 10-
year fracture risk is the FRAX tool, developed by the WHO (Kanis, 2013), was a driver of this
particular change to the guidelines. In addition to these significant changes, Revision 2 affirmed
that labeling for new formulations or new routes of administration of already approved drugs
could be supported by conducting comparative assessments of BMD alone for a minimum of one
year (i.e., clinical bridging studies), and additional details on the design of such studies was
provided.
The 2007 EMEA Revision 2 Guideline retained from Revision 1 the requirement for applicants to
assess loss of effect due to altered bone structure, and required that data to characterize the effects
following treatment withdrawal be submitted in the post-marketing period. Also similar to
Revision 1, Revision 2 reiterates its preference to use placebo-controlled studies whenever
possible (EMEA, 2007, Ormarsdottir, 2008). This position is consistent with that of the GREES,
which agrees that this design offers greater efficiency and is also “ethically acceptable in patients
with osteoporosis but without prevalent fractures” (Ormarsdottir, 2008). Further, both versions
45

place applicants on notice to consult the ICH E10 guidance on the Choice of Control Group and
Related Issues in Clinical Trials if they wish to conduct a study using an active control product,
and suggest extra precaution in planning and conducting studies of non-inferiority trial design
(EMEA, 2007). Moreover, Revision 2 of the EMA’s guidance requires sponsors to justify clearly
their choices of margin for non-inferiority and choice of comparator in advance of study start,
and indicates that a placebo arm may be required even if a study design with an active control is
employed (EMEA, 2007).
An important issue that is not addressed by the Revision is the challenge of selecting the most
appropriate standard for the comparator agent from the armamentarium of approved products – a
decision and risk that must be taken by the sponsor alone. With the additional guidance on the
conduct of active-controlled studies, the groundwork has been set to allow the EMA to begin
requiring such studies from sponsors of investigational PMO therapies. In fact, it is publically
known that the most recent PMO therapy to enter into phase 3 clinical trials, romosozumab
(a.k.a., AMG 785), is conducting one placebo (FRAME study; ClinicalTrials.gov No.
NCT01575834) and one active controlled fracture efficacy study with alendronate
(ClinicalTrials.gov No. NCT01631214).
2.2.4 Comparison of Key Guidelines
Table 4 outlines the primary differences between the 1994 FDA Draft Guidance and the 2007
EMEA Revision 2 Guideline. Although many common features exist between the two standards,
some notable exceptions can also be identified.
The FDA and EMA requirements differ in the allowable types of indications that may be sought
in PMO. Although both guidance documents generally recognize that treatment of osteoporosis
is a valid indication to seek, a separate indication of prevention of osteoporosis for osteopenic
46

patients was no longer recognized by EU regulators for registration purposes in the 2007 EMEA
Revision 2 Guideline. Additionally, although FDA’s 1994 Draft Guidance clearly permits a
labeled indication statement for the treatment of postmenopausal women with osteoporosis,
FDA’s thinking has evolved to allow for a third indication statement for the treatment of PMO
patients at a high risk for fracture, defined as a history of osteoporotic fracture, or multiple risk
factors for fracture, or patients who have failed or are intolerant to other available osteoporosis
(Forteo, 2012; Prolia 2013). The ‘high-risk’ treatment clause is generally reserved for products
the FDA considers to possess either known or theoretical safety risks of special concern, and has
been applied to only two classes of PMO pharmacotherapy: teriparatide (Forteo; PTH 1-34) and
denosumab (Prolia). As a consequence, PMO therapies granted a ‘high risk’ treatment indication
have not subsequently received the prevention indication, in effect placing a limitation on the use
of these therapies in lower risk, osteopenic patients. The FDA’s bifurcation of available
treatment indications, in addition to the prevention indication presumably still allowed under the
FDA’s 1994 Draft Guidance, differs from the 2007 EMEA Revision 2 Guideline where the
prevention and treatment indications have been streamlined under a single indication to reduce
redundancy: treatment of osteoporosis in postmenopausal women at increased risk of fracture
(EMEA, 2007). In essence, the EMA’s position is that if the aim of pharmacotherapy is to
prevent the incidence of future fracture, it is redundant and confusing to speak of either a
prevention or treatment indication, since they are one of the same. Further, since a labeled
indication statement generally reflects the population studied, understanding how these indication
statements should be implemented in practice into clinical study design features, such as inclusion
and exclusion criteria, is not an inconsequential question.
In addition, the 2007 EMEA Revision 2 Guideline requires randomized trials that demonstrate
anti-fracture efficacy at the two-year rather than 3-year mark required in the FDA’s 1994 Draft
47

Guidance. For the primary efficacy variable, EMA requires anti-fracture efficacy using separate
assessments of spinal and non-spinal fractures (preferably in separate studies) in contradistinction
to the FDA requirement that focuses only on the incidence of new vertebral fractures. In
addition, EMA but not FDA requires data on fracture incidence after treatment withdrawal
(EMA, 2006).
The two DRAs also differ somewhat in their recommendations related to patient populations used
for study in the pivotal phase 3 study. The patient populations recommended by the FDA’s 1994
Draft Guidance are ambulatory out-patients ( ≥ 5 years postmenopausal), ≥ 60 years of age with
symptoms and signs of PMO (i.e., bone pain and height lose), with vertebral fracture and/or
lumbar vertebral BMD ≥ 2 S.D. below mean peak BMD. EMA on the other hand recommends a
PMO patient population at increased risk of fractures, i.e., a 10 year fracture risk of 15-20% in the
spine, 5-7.5% in the hip, and 10-15% major non-vertebral fractures.
Importantly for sponsors of registrational phase 3 clinical studies, FDA and EMA guidance
documents remain silent on the ethical issues associated with conducting placebo-controlled
studies (to be discussed below), and thus leave to the drug sponsors the task of resolving the
multiplicity of potential study design related issues related to ethical concerns. Both guidelines
appear to favor placebo-controlled studies given the challenges of conducting active controlled
studies, a point re-emphasized to a much greater degree in the EMA Guideline. This is to some
degree surprising, given the recent general trend towards trials utilizing active controls, a point
underscored by new guidelines for such studies on both sides of the Atlantic (EMA, 2010; FDA,
2010a). As a result of such challenges, the approval precedence for novel agents in the EU, as in
the US, has been based solely on placebo-controlled studies up until this point. Active controlled
studies have been conducted, but they have been primarily designed to compare effects on BMD,
not on rates of fragility fracture, and thus are generally not suitable for purposes of product
48

labeling as they do not meet minimum regulatory requirements of demonstration of fracture
efficacy. Of interest from the standpoint regulatory dissonance, active-controlled comparator
studies which utilize BMD as the primary endpoint have not generally been suitable for purposes
of product labeling in the United States, as they do not meet minimum regulatory requirements of
demonstration of fracture efficacy, but are allowed in some cases to be reported in prescribing
information elsewhere.
In terms of supportive nonclinical data requirements, 2007 EMEA Revision 2 Guideline, like the
FDA’s 1994 Draft Guidance, calls for an assessment of bone quality in adult overiectomized rats,
but its requirements for study duration and dose levels differ from those in the FDA’s guidance.
The EMA recommends that the duration of study be equivalent to 5 remodeling cycles in humans
and requires 3 dose levels (one equivalent to an expected half-maximal response, optimal
response, and a high dose a multiple above the optimal response dose level) (EMEA, 2007).
Table 4 provides a high level summary of the differences to familiarize the reader.
49

Table 4. Primary Differences in the Elements of the FDA and EMA Guidance for
Developing Agents for Postmenopausal Osteoporosis
Area 1994 FDA Draft Guidance 2007 EMA Guideline (Revision 2)
Potential
Indications
Treatment and/or prevention of
women with PMO at high risk for
fracture
Treatment of PMO in women (or men) at
increased risk of fracture
Design Randomized, double-blind,
placebo >>> active-controlled
Randomized, double-blind,
placebo >>> active-controlled
Duration of
Treatment /
Assessment
of Safety
 Treatment of osteoporosis: ≥ 3
yrs; continued post-marketing to 5
yrs. if anti-fracture efficacy not
proven at 3 yrs. (non-estrogens)
 Prevention of osteoporosis: ≥ 2
years
1

 Treatment of osteoporosis: 2 yrs;
maintenance of effect after second year (3-5
yrs.)
Primary
Endpoint
 Demonstration of anti-fracture
efficacy (new vertebral fractures)
 Demonstration of anti-fracture efficacy
(new spinal and non-spinal fractures). For
non-spinal fractures, either hip or major
non-vertebral fractures may be assessed.
Secondary
Endpoints
 Statistically significant
improvements in BMD
 Demonstration of normal bone
quality a subset of subjects (bone
biopsy)
 BMD from areas studied for fracture
incidence
 Efficacy at first year
 Biochemical variables reflecting bone
turnover
 Withdrawal of effect data (submitted post-
registration)
Nonclinical  Demonstration of normal bone
quality in adult overiectomized rats
and a larger remodeling species
2
◦ Duration: 16-mos. (based on
rate of bone turnover of species)
◦ Dose levels: 2 dose levels (1
optimally effective dose, one dose
5x greater)
 Demonstration of normal bone quality in
adult overiectomized rats and in animal with
oestrogen deficiency (induced by
ovariectomy and characterized by evaluable
cortical bone remodeling)
◦ Duration: equivalent to 6 remodeling
cycles
◦ Dose levels: 3 (half-maximal response,
optimal response, and a multiple greater
than mid-dose)
Status Draft guidance issued 1994,
withdrawn 2009
In effect since 2007, addendum on
secondary disease under consideration
1
“If a drug has been approved for the treatment of osteoporosis, bone mineral density (BMD) may serve as
an appropriate efficacy endpoint” (FDA, 1994)
2
“If bone quality is abnormal, but fracture efficacy is proven at year 3 of the phase 3 pivotal clinical study,
the pivotal clinical study may need to be continued to year 5 in the post-marketing period. However, if
fracture efficacy is not observed year 3 in the pivotal clinical study, and bone quality is observed to be
abnormal, the investigational PMO agent will not be approved (approval contingent on continuing the
controlled study for 5 years to determine safety and fracture efficacy)” (FDA, 1994)
50

2.2.5 Additional Guidelines of Interest
A few regulatory guidance documents related to the development of PMO agents exist apart from
those published by FDA and EMA. In general, these guidelines may be less well-known,
detailed, or accessible to English speaking readers (i.e., PMDA, CFDA). In general, regulators in
other constituencies tend to follow either the paradigm of FDA’s or EMA’s guidance, but
sometimes with added clinical bridging requirements specific to local populations, such as special
pharmacokinetic or clinical safety/efficacy evaluations in local populations as required by some
Asian DRAs (i.e., China, Japan, South Korea, Taiwan, etc.).
On one end of the spectrum, Australia’s Therapeutic Goods Authority (TGA) has simply adopted
the EMA Guideline on the evaluation of new medicinal products in the treatment of primary
osteoporosis in its entirety (effective: 07 August 2008). With the creation of a combined
Australia New Zealand Therapeutic Products Agency (ANZTPA) in July 2011, which is expected
to be fully operational within 5 years, it can be anticipated that requirements for PMO agents will
be aligned under a single trans-Tasman regulatory system (TGA, 2011). On the opposite end of
the spectrum, Health Canada, a DRA that commonly develops guidance documents of its own,
has not issued an equivalent guidance document for an indication of PMO, nor has it formally
adopted the use of either FDA’s or EMA’s guidance.
The Japanese Pharmaceuticals and Medical Devices Agency (PMDA) has issued its Guidelines
for Clinical Evaluation Methods of Drugs for Osteoporosis (PMSB, 1999). These guidelines
recommend double blind, placebo- or ‘standard’ drug-controlled studies. An assessment of
fracture incidence following at least three years treatment is ‘conceivably required’ as the
recommended primary endpoint for pivotal studies, and BMD and height are recommended as
secondary endpoints, although height is no longer a standard endpoint required by most DRAs for
51

pivotal, registration studies. The PMDA guidelines highlight that ‘racial factors’ be considered
when overseas clinical trial data are used for establishing drug efficacy since correlations
between BMD and fracture risk have not been evaluated “in plural races” (PMSB, 1999). The
content of this guideline is generally consistent with the basic tenants of the FDA and EMA
guidelines; however its syntax is often difficult to follow and is less specific in its
recommendations than the FDA and EMA standards. For this reason, it was not described in
detail in Chapter 2.4.4, Comparison of Key Guidance Documents.
Recently, the Chinese Center for Drug Evaluation issued its own guidance document for sponsors
seeking an indication of PMO in China (CFDA, 2011). Although an English version of this
document is not available, translation was accomplished through the use of the online Google
Translate tool (available at: http://translate.google.com). Similar to the PMDA Guidelines, it
identifies reduction of fracture risk as a key efficacy endpoint, and requires at least 3 years of
treatment data, preferably in a placebo-controlled trial. If overseas phase 3 studies demonstrating
efficacy and safety have been conducted, a confirmatory trial in China is required to corroborate
results adding additional time and costs to development. However, for the required bridging trial
in a Chinese population, BMD may be used as a surrogate endpoint of efficacy, rather than
fracture efficacy, if independent factors of fracture risk are also evaluated in the main multi-
regional clinical trial (CFDA, 2011). Required secondary endpoints include assessments of bone
formation and resorption markers, as is the current standard for PMO registration trials. In
addition, the CFDA guidelines require the applicant to reference the relevant guidelines on which
it relied to design the drug registration program (i.e., FDA and EMA guidelines). Although the
CFDA’s guidance appears to favor the expediency of placebo-controlled studies, similar to FDA
and EMA, it departs from these guidelines by allowing flexibility in trial duration, suggesting
instead that studies consider the unique properties and mechanism of action of the product under
52

evaluation when considering the appropriate study duration (CFDA, 2011). With this recent
example, one can anticipate that emerging economies will soon begin to assert their own policy
influence by issuing their own indication-specific guidelines, thus increasing overall regulatory
dissonance.
In 1995, the World Health Organization (WHO) and its Division of Drug Management and
Policies established a working party tasked with developing international “Guidelines for
Preclinical Evaluation and Clinical Trials in Osteoporosis” (WHO, 1998). The WHO is not a
DRA, and lacks the legal authority to mandate the use of their guidelines, although
representatives from DRAs and concerned scientists participated in its development. As such, the
WHO guidelines are written in a manner somewhat less prescriptive that the two guidelines from
the FDA and EMA, and read as “guiding principles for the design, implementation, and
interpretation of either preclinical or clinical trials in osteoporosis” (WHO, 1998) rather than
directives. These guidelines are interesting in that they advocate that the establishment of
therapeutic effect be based upon data from two well-designed and controlled studies (WHO,
1998), a position not advocated by either the FDA or EMA in PMO. In addition, a minimum
3 year study duration is recommended for phase 3 studies, but the WHO guidelines do not state
whether both recommended phase 3 studies must be of this duration. Since the 1998 WHO
guidelines are not regulatory authority guidelines per se, these guidelines are not further
considered herein.
2.3 Regulatory Dissonance in Postmenopausal Osteoporosis Drug Approval
Precedents
Regulatory approval precedents are similar to case law in that successful product approvals
clarify the requirements that are needed to gain regulatory approval for a particular product class
53

in a certain therapeutic area. Precedents establish an important baseline for the assessment of
subsequent products, by serving as static reference points that influence the framework for future
regulatory decisions. Thus, an approval of a novel NCE/NBE provides a developmental map for
future products in that indication, and when combined with the recommendations of available
guidance documents from DRAs (described in Chapter 2.2), provides a fairly comprehensive
understanding of the regulatory landscape in a particular indication. In addition, since review
decision precedents are established regionally, they can provide a useful window through which
to gain insight into the different views and decisions taken by specific DRAs that contribute to
regulatory dissonance.
Review precedence may be researched from regulatory authority websites and in the scientific
literature. Perhaps the best is publicly available summary basis of approval documentation (i.e.,
Summary Basis of Approvals – SBA (FDA), or European Public Assessment Reports – EPAR
(EMA), or Summary Basis of Decisions – SBD (Health Canada)). Scientific literature describing
clinical trials undertaken to support a marketing submission can augment the information
provided by regulators in summary basis of approval documentation (examples of which are
provided further in this chapter). However, when trying to gain insight into regulatory
dissonance, it is the negative decision by a DRA that points to the most problematic areas of
dissonance. As noted earlier, the details surrounding negative review outcomes are generally not
made public by DRAs, but are rich in relevant regulatory information; these examples provide the
“lessons learned” for industry. Some insight into the reasons why such products are delayed or
denied market access can be deduced when, for example, a product has been brought for
discussion to an Advisory Committee Meeting, which is a publically held meeting midway
through the review process that allows the FDA to gain input from informed experts on questions
of concern regarding the drug. Since the FDA, and not the Advisory Committee, is the final
54

arbiter of the regulatory decision, one must often “read the tea leaves” from such meetings though
Advisory Committee transcripts and briefing materials, scientific literature, scattered news
reports, and casual conversation with industry colleagues. However, not all drug review
decisions are brought to an advisory committee meeting. Such meetings are often reserved for
products whose novelty of performance is not straightforward and for which scientific/regulatory
controversy exists.
As described in Chapter 1, there are six main agents approved by the EMA (and other ex-US
DRAs), which were subsequently not approved by the FDA, and which serve as an indicator of
regulatory dissonance: cyclical disodium etidronate (Didronel), tibolone (Livial), strontium
ranelate (Protelos), intact parathyroid hormone or PTH 1-84 (Preotact), bazedoxifene (Conbriza),
and lasofoxifene tartrate (Fablyn).
The following brief summaries attempt to identify the reasons why these six products were not
approved in the US, and why an additional product was recently removed from the EU
marketplace, as a means to shed light on potential sources of regulatory dissonance that may have
contributed to these outcomes.
2.3.1 Cyclic Etidronate Disodium
Cyclical disodium etidronate (EU tradename: Didronel) was one of the first treatments
extensively evaluated for PMO (Greenspan, 2000). As one of the first drugs for PMO in the
1980s/1990s, it had few precedents from which to model its regulatory pathway. Thus, the
applicant would have relied on the sole guidance document available at the time, the FDA’s PMO
Guidance document of 1984. Although not subsequently approved for an indication of PMO by
the FDA, etidronate has been approved in the US for other bone diseases, including Paget’s
55

disease and heterotropic ossification. It is also approved outside of the US for PMO, but is not
considered a first line therapy for PMO.
The etidronate development program was supported by two main phase 3 studies. The first study,
a 3-year, randomized, double-blind, placebo-controlled study conducted in Denmark,
demonstrated a “significant increase in vertebral bone mineral content and, after approximately
one year of treatment, a significant decrease in the rate of new vertebral fractures” (Storm, 1990).
The second study, which was conducted in the U.S., appeared to corroborate the results of the
Danish study (Watts, 1990). However the FDA’s own analyses of this study “showed that in the
third year of the study the women who were taking etidronate had twice as many new spinal
fractures as the women who were not taking etidronate” (OTA, 1994). From these data, it was
clear that BMD was not acting as a positive predictor of fracture frequency. These data were the
subject of a March 1991 Endocrinologic and Metabolic Drugs Advisory Committee sponsored by
the FDA, which “voted that the data presented to them did not provide substantial evidence for
the efficacy of etidronate” (OTA, 1994). Although it was demonstrated that sustained treatment
with cyclical etidronate significantly decreased the rate of vertebral fracture, these effects were
only observed in patients at highest risk for fracture, thus substantial evidence of efficacy could
not be demonstrated (Greenspan, 2000). Interestingly, this review outcome did not constrain
regulators outside of the US to authorize the marketing of the product for PMO based upon the
same dataset (EU, Canada, Australia, Japan, etc.).
2.3.2 Tibolone
Tibolone (UK tradename: Livial) is a second-line treatment for PMO prevention available in 45
countries outside the US (Cummings, 2008). Livial is a synthetic hormone, which exerts a
combined androgenic, estrogenic and progestogenic effect (MHRA, 2007; Modelska &
Cummings, 2002), and thus is part of the class of PMO therapies termed hormone replacement
56

therapy (HRT). Livial is a legacy product first licensed nationally for PMO in 1987, and thus the
clinical development program did not benefit from either the common pan-EU osteoporosis
guidelines later developed by the EMEA, or the 1994 FDA guidance establishing fracture risk
reduction as the gold standard for confirming efficacy. It was only much later between 2001
through 2006 that a large (N=4534), randomized, placebo controlled, 3-year study evaluating the
primary efficacy variable of reduction of incident new vertebral fractures (ClinicalTrials.gov No.
NCT005199857) was conducted. Prior to this study, six relatively small trials (ranging between
an N of 21 and 519) of relatively short duration (ranging between 8 and 104 weeks) had been
undertaken by the sponsor which demonstrated significant effects of tibolone treatment on net
change in BMD (Modelska & Cummings, 2002). As described in the MHRAs Public
Assessment Report on the benefit:risk assessment of tibolone (MHRA, 2007), the LIFT study
(Long-term Intervention on Fractures with Tibolone) demonstrated a 50% reduction in the
incidence of new vertebral fractures in older patients, decreasing risk of both vertebral and
nonvertebral fractures (MHRA, 2007; Cummings, 2008). However, the study revealed
concerning signals such as a significant increase in the risk of stroke, a finding that resulted in
early closure of the study at the recommendation of the data and safety monitoring board
(Cummings, 2008), and a non-significant increased incidence of endometrial cancer. Thus,
although fracture efficacy was demonstrated, safety concerns appear to have contributed
significantly in the FDA’s decision not to grant approval (MHRA, 2007).
2.3.3 Strontium Ranelate
Strontium ranelate (EU trade name: Protelos/Osseor), a claimed dual action bone agent (DABA)
with bone ‘anabolic’ and ‘catabolic’ activity, is composed of two atoms of stable strontium and
organic ranelic acid, is another PMO agent that has been approved in the rest-of-world (ROW)
but not the U.S. The safety and efficacy of Protelos was evaluated in two randomized, placebo-
57

controlled MRCTs, entitled SOTI (Spinal Osteoporosis Therapeutic Intervention) and TROPOS
(Treatment Of Postmenopausal Osteoporosis) (Blake & Vogelman, 2006). In the SOTI trial,
comprised of 1649 subjects, a significant “49% lower risk of a new radiographic vertebral
fracture in the strontium ranelate group compared with the placebo group (p<0.001)” was
demonstrated, and at year 3, “the strontium ranelate group had a 41% lower risk of a new
radiographic vertebral fracture than the placebo group (p<0.001)” (Blake & Vogelman, 2006).
The TROPOS study had similar results, with a 45% and 39% reduction of vertebral fracture risk
reduction at years 1 and 3, respectively; however, vertebral fracture reduction was not a pre-
specified endpoint of the study and thus the pivotal study was not a fracture efficacy study per
FDA criteria (Blake & Vogelman, 2006). A pooled analysis of both studies was also prepared by
the sponsor, the data from which showed a reduction in vertebral fracture risk of 40% (Blake &
Vogelman, 2006). The action of strontium ranelate on BMD was equally impressive. However it
should be noted that strontium, an element which is taken up by bone to replace calcium lost,
makes a relatively larger contribution to BMD than calcium by means of its high molecular
weight. Blake and colleagues note that “at least 50% of the observed BMD increase is an artifact
caused by the high bone strontium content,” an effect which will “persist for many years after the
patient discontinues treatment” (Blake & Vogelman, 2006). Other researchers have reported that
“bone-biopsy specimens from patients treated with strontium ranelate show a reduction in bone
resorption but no evidence of increased bone formation” (Canalis, 2007). Thus claims of a dual
mechanism of action appear to be exaggerated.
In terms of safety, data from the two phase 3 studies showed that the annual incidence of venous
thromboembolism (VTE) including pulmonary embolism was increased” (Blake & Vogelman,
2006). Although negative review opinions from the FDA are not public information, it appears
probable that the FDA’s assessment may have weighted to a greater degree (1) the artefactual
58

BMD results (i.e., molecular weight of strontium vs. calcium) (Blake & Vogelman, 2006), (2) the
post-hoc evaluation of vertebral fracture efficacy, and (3) the increased VTE risk (Blake &
Vogelman, 2006), in its benefit:risk decision for Protelos. However, these concerns did not
prevent the authorization of strontium ranelate in the EU, and only recently have influenced the
terms of authorization within the EU. In March 2012, EMA issued the outcome of review
procedures (under Article 20 of Regulation (EC) No. 726/2004) following continued concerns
relating to increased risks of VTE and severe allergic skin reaction in a report entitled Questions
and answers on the review of Protelos and Osseor (strontium ranelate) (EMA, 2012a).
Moreover, this review has resulted in regulatory action (April 2013) in the form of a
recommendation to restrict the use of strontium ranelate in osteoporosis by the CHMP’s
Pharmacovigilance Risk Assessment Committee (PRAC). The specific recommendations by
CHMP include that strontium ranelate should only be used to treat a subset of postmenopausal
osteoporotic women at “high risk of fracture and severe osteoporosis” and men “at increased risk
of fracture,” and include additional measures such as a “restriction in patients with heart or
circulatory problems” (EMA, 2013). Despite the accumulating evidence impacting the safety of
this agent, PRAC recommended that the benefit:risk balance for strontium ranelate remain
positive, although it acknowledged that this evaluation is ongoing.
2.3.4 Intact PTH 1-84
Intact Parathyroid Hormone (EU trade name: Preotact; proposed US trade name: Preos), a
parathyroid hormone analogue (PTH 1-84;ALX1-11), was authorized for marketing in certain EU
member states in 2006. The pivotal registrational study (“TOP” trial) was a randomized, double-
blind phase 3 study (N=2600) with a primary endpoint to evaluate the incidence of new and/or
worsened thoracic and lumbar vertebral fractures at 18 months (ClinicalTrials.gov No.
NCT00172081) and secondary safety assessments. Although the study met its primary endpoint
59

of significantly reducing the risk for new or worsened vertebral fractures, “parathyroid hormone
treatment increased the percentage of participants with hypercalciuria, hypercalcemia, and nausea
by 24% (CI, 20% to 27%), 23% (CI, 21% to 26%), and 14% (CI, 11% to 16%), respectively,
compared to placebo” (Greenspan, 2007). In response to the same data package, the FDA issued
an approvable letter emphasizing their concerns with (1) imbalances of hypercalcemia in the
Preos study arm of the pivotal trial (media reports at the time describe an FDA request for an
additional clinical study to evaluate hypercalcemia) and (2) the reliability of the injection device.
Thus, it appears that differential benefit:risk assessments by the FDA and EMA is the cause of
these divergent regulatory review outcomes.
2.3.5 Bazedoxifene
Bazedoxifene acetate (EU trade name: Conbriza; TSE-424) is a another example of a PMO
product approved in ROW but not in the US. In this instance, an approval by the EMEA in April
2009, and in Japan (July 2010; Viviant is the tradename for Conbriza in Japan) for the treatment
of PMO did not prompt a reversal of an earlier decision by the FDA to block approval of the
agent in the US. Conbriza belongs to the class of PMO drugs called selective estrogen receptor
modulators (SERMs). Approved SERMs (e.g., Evista) were already marketed at the time of
regulatory review for bazedoxifene (Table 1). In general, safety concerns with this class of drugs
(e.g., VTE) and the reduced efficacy in preventing non-vertebral fractures versus that seen with
most first line anti-resorptive agents may be at issue with respect to the US non-approval.
The sponsor conducted two double-blind, randomized, placebo-controlled studies in support of
registration. In line with the FDA’s guidance for 2
nd
generation estrogen molecules, the first
study (N=1742) had a primary outcome that measured the percent change in lumbar spine BMD
after 2 years (ClinicalTrials.gov No. NCT00481169) (Sobieraj, 2011). The second study
exceeded the requirements for a 2
nd
generation estrogen molecule: it was a large (N=6847),
60

placebo and active-controlled study with a primary endpoint evaluating new vertebral fractures at
3 years and secondary endpoints of nonvertebral fractures, BMD, and BTM (Silverman, 2008a).
Up until this point in PMO drug development, no sponsor had conducted an active-controlled
pivotal study, let alone one with both placebo- and active-control arms.
The FDA issued approvable letters (Note: Per FDAAA, the approvable letter was succeeded in
2008 by the complete response letter) for a 2006 NDA filing for an indication of prevention of
PMO (April and December 2007, respectively), and later issued an additional approvable letter
for a 2007 NDA filing supporting an indication of treatment of PMO (May 2008). For both
actions, the FDA requested that it “receive and analyze, as part of its benefit-risk assessment,
final safety and efficacy data from a recently completed Phase III treatment evaluation of
bazedoxifene,” and requested further “analyses concerning the incidence of stroke and venous
thrombotic events” (Drugs, 2008) At the time of this writing, bazedoxifene has yet to be
approved by the FDA, no public information on the FDA’s assessment exists, and thus the
benefit:risk of this product remains in the balance.
2.3.6 Lasofoxifene
Lasofoxifene tartrate (EU tradename: Fablyn), another second generation SERM under the same
drug class as bazedoxifene, was approved by the EMEA in February 2009 for PMO prevention.
Of note, lasofoxifene was the last drug to be authorized in the EU with a labeled indication of
PMO prevention, an indication no longer supported by the EU Revision 2 guideline (EMEA,
2007). In the US, the regulatory challenges were numerous. Two separate NDA applications
supporting the indications of osteoporosis prevention and treatment of vaginal atrophy were
submitted in 2004 and rejected based on concerns that the product may cause cancer of the
uterine lining. Later in 2007, another NDA was submitted in support of a PMO treatment
indication based upon data from the PEARL study (Postmenopausal Evaluation and Risk-
61

Reduction with Lasofoxifene; ClinicalTrials.gov No. NCT00141323). This large (N=8556),
randomized 5 year phase 3 study evaluated not only 3-year vertebral fracture efficacy as a
primary endpoint and nonvertetral fractures, but also a novel safety endpoint of estrogen receptor
(ER)-positive breast cancer to address concerns that this product class is a cancer inducer
(Cummings, 2010). Secondary cardiovascular safety endpoints of major coronary heart disease
events and stroke (Cummings, 2010) were also evaluated given the anticipated safety concerns
associated with this drug class. Although the PEARL trial demonstrated reduced risk of
“nonvertebral and vertebral fractures, ER-positive breast cancer, coronary heart disease, and
stroke,” it also showed a significantly (3x) increased risk of VTE (Cummings, 2010).
In 2008, the FDA recommended that Fablyn be reviewed by its Reproductive Health Drugs
Advisory Committee (RHDAC), which resulted in a “committee recommendation that the
benefits of treatment with lasofoxifene outweighed the risks” (FDA, 2008a). Although the FDA
normally adopts the position of its Advisory Committee members, it did not in the case of Fablyn
and issued the sponsor (Pfizer) a complete response letter in 2009. Pfizer subsequently
announced (May 2010) that it was withdrawing its NDA for the treatment of PMO women at
increased risk of fracture.
At the time of this writing, lasofoxifene has yet to be approved by the FDA, and the benefit:risk
of this product and class remains in question. Since the FDA’s action, the marketing
authorization for Fablyn was allowed to lapse since it has not been marketed in the EU in the
three years following the granting of the authorisation, triggering a sunset clause (per article 14(4)
of Regulatory (EC) No. 726/2004) (EMA, 2012b). The circumstances surrounding this action are
unclear.
62

2.3.7 Calcitonin-containing Agents
Calcitonin-containing agents (Calcimar, Miacalcin, Fortical) provide a developing example of
possible regulatory dissonance between the FDA and EMA, as marketing authorization has been
withdrawn for this entire class of drugs by EMA. Recently, the EMA issued a Questions and
answers on the review of calcitonin-containing medicines – Outcome of a procedure under
Article 31 of Directive 2001/83/EC (19 July 2012; EMA/476001/2012;EMEA/H/A-31/1291)
(EMA, 2012c). The assessment concluded that the benefits of calcitonin-containing agents no
longer outweigh the risks in PMO. The conclusion was prompted by a review of data from an
“unlicensed oral calcitonin medicine” which suggested an increased incidence of prostate cancer.
This finding led to a full review of available clinical and post-marketing data from marketed
products made available from various drug sponsors of these agents and from the scientific
literature (EMA, 2012c). The EMA concluded that a significantly increased incidence of various
types of cancer appeared to be associated with long-term use of calcitonin-containing agents
(EMA, 2012c). Although the rate of new cancers was reported to be low, this risk combined with
lower relative levels of therapeutic benefit against vertebral fractures when compared to available
first-line therapies prompted the EU to revoke its marketing authorization (EMA, 2012c). The
US approval of this class of agents has not been withdrawn at the time of this writing; however an
FDA advisory panel advised in March of 2013 that calcitonin salmon no longer be indicated for
postmenopausal osteoporosis (FDA, 2013a). Whether FDA accepts the recommendations of its
advisory panel is yet to be seen, although this is widely expected. For this reason, this case may
or may not turn out to represent another case of regulatory dissonance, yet is summarized for
informational purposes.
63

2.3.8 Summary
What can be observed in these 6 cases is a differential pattern of benefit:risk assessments by
different DRAs, illustrating regulatory dissonance in the manner by which DRAs weigh certain
safety findings versus the clinical effects of investigational agents. In these cases, inherent
differences in DRA regulatory guidelines did not appear to have been a significant factor
contributing to the regulatory dissonance in these review outcomes. Differences in approaches in
drug review standards, such as the FDA’s reliance on patient-level data and non-acceptance of
post hoc subset analyses for efficacy data, may have played a role in these divergent benefit:risk
outcomes. Interestingly, most dissonance in regulatory review decisions was not observed with
the first line treatments such as the anti-resorptive therapies, but is noted primarily in the
regulatory treatment of second line therapies (Table 5). For second line treatments, assessments
of benefit:risk may be less unambiguous, leading to a higher probability of regulatory dissonance.
Also, presence of a significant safety concern appears to be weighted differently by EMA and
FDA. The EMA appears to stress the importance of treatment option availability, allowing
medical judgment to reside with the primary care physician; however the FDA appears to play a
more active role in prescribing medical judgment as it relates to use of specific treatments where
benefit:risk is less clear.
2.4 Ethical Controversies in Postmenopausal Osteoporosis Drug Approval
Standards
Regulatory standards are one way to shape the design of pivotal clinical studies and the content of
marketing authorization submissions. However it is recommended that sponsors also consider the
opinions of the medical community through the use of advisory boards made up of key opinion
leaders and study investigators since their opinions may deviate from the guidelines of the DRAs.
The opinions of ethics committee members must also be considered. Within the osteoporosis
64

medical community, there has been considerable debate regarding the optimal manner by which
to conduct clinical trials in osteoporosis. Specific aspects of the design of registration studies that
have been at the focal point of this dialogue include (1) whether pivotal, registration studies
should be placebo-controlled or be active-controlled, (2) whether the duration of study treatment
should be a minimum of three years for a novel agent versus a shorter duration, and to a much
lesser extent, (3) whether fracture efficacy is the most appropriate endpoint upon which to
evaluate the efficacy of all classes of PMO agents. The issues associated with the first two
aspects are particularly interesting because they have become contentious, and because they
illustrate the fact that not all sources of regulatory dissonance are between regional DRAs.
Regulators may only be observant onlookers to this ongoing debate, mindful that any heavy
handed application of their authority could be a lightning rod for further disagreement amongst
dissenting factions. Thus, even within the United States, where indication-specific guidance for
osteoporosis drug development has existed since 1979, agreement on the requirements for an
indication of PMO remains an elusive goal. This is underscored by the fact that the FDA’s Draft
1994 Draft Guidance document, which was never finalized, was formally withdrawn in 2009 per
the FDA’s regulatory guidance documents website (FDA, 2009b), although it is still reported to
be recognized by the FDA (Storm, personal communication). In this political context, a DRA
must be careful not to dictate medical opinion.
2.4.1 Placebo- versus Active-Controlled Studies
The US and EU guidelines for PMO clinical trials, as well as commercialized drug precedents,
have firmly established that placebo-controlled trial designs are the recognized standard for
establishing efficacy and safety. However, the opinions of regulators are only part of the
equation: investigators, institutional review boards (IRBs)/Ethics Committees (ECs) and research
ethicists have questioned the ethics of conducting placebo-controlled studies for over a decade
65

given the availability of effective interventions (EMEA, 2006). Although study investigators and
IRB/EC members are not government regulators per se, they are integral to the ongoing ethical
dialogue, and as such, are important drivers shaping the overall regulatory environment by their
oversight of the operational conduct of clinical studies. For instance, some concerned clinical
investigators have refused to participate in placebo-controlled studies in PMO, and some
IRBs/ECs have declined to approve such studies, although such studies meet approved regulatory
guidelines. These challenges to regulatory guidelines can complicate the operational conduct of
large scale pivotal PMO studies. The discourse in the medical community, involving key opinion
leaders, major foundations and advocacy groups representing patients, as well as regulatory
agencies themselves, is summarized in a series of publications, some of which call for a revision
of the available PMO guidelines in the US and EU.
In 2003, the Journal of Bone and Mineral Research (JBMR) featured a supplement devoted to the
debate surrounding the scientific validity and ethics of randomized, placebo-controlled phase 3
studies in the setting of PMO. The supplement was prefaced in Osteoporosis Trials: Ethics in
Study Design (Recker, 2003) and described the discordant views by investigators and
Investigational Review Boards (IRBs) developing in the late 1990s in relation to the continued
use of placebo-controlled studies. At that time, several anti-fracture agents (i.e., Fosamax, Evista,
Actonel) had been introduced under the regulatory framework created by the FDA and EMA
guidelines (Table 4). With this notable success came the quandary which this chapter addresses:
is it still ethical to randomize patients with documented PMO into a placebo-controlled study
when appropriate treatments can be obtained that would expose patients to less risk by treating
their underlying clinical problem (Stein, 2010)?
This concern amongst investigators, ethicists and drug manufacturers grew following the 2000
Revision V of the Declaration of Helsinki (DoH, World Medical Association), when Articles 5
66

and 29 appeared to question the use of placebo-controlled studies where a proven therapeutic
method exists (Kanis, 2003);
Article 5: “In medical research on human subjects, considerations related to the well being of the
human subject should take precedence over the interests of science and society.”
Article 29: “The benefits, risks, burdens and effectiveness of a new method should be tested
against those of the best current prophylactic, diagnostic, and therapeutic methods. This does not
exclude the use of placebo or no treatment in studies where no proven prophylactic, diagnostic or
therapeutic method exists.” (Kanis, 2003)
Although the DoH guidelines had broader implications beyond clinical trial work in PMO, their
implications could not be ignored. To address these nontrivial concerns, the American Society
for Bone and Mineral Research (ASBMR) organized two consecutive conferences (known as the
Natcher and Omaha meetings). The purpose of these early meetings was intentionally delimited
to allow an open discussion of the issues (“to inform and discuss, not create policy”). No clear
consensus was apparent amongst the diverse participants (e.g., FDA, NIH, OHSRP, ASBMR, key
opinion leaders from the scientific community, drug sponsors). Although it is difficult to do
justice to the positions represented by this supplement in this short overview, the main conceptual
arguments described below are extracted here from the following articles contained in the
supplement (Recker, 2003; Brody, 2003; Capron, 2003; Colman, 2003; Ellenberg, 2003; Fost,
2003; Heaney, 2003; Kanis, 2003; Khosla, 2003; Levine, 2003; Melton, 2003; Recker, 2003;
Rosenblatt, 2003; Temple, 2003; Weijer, 2003).
Brody and colleagues summarized the proceedings of the Omaha meeting, held at the Creighton
University Center for Health Policy and Ethics on 25 June 2001, in a paper titled Is the Use of
Placebo Controls Ethically Permissible in Clinical Trials of Agents to Reduce Fractures in
Osteoporosis? (Brody, 2003). This meeting was sponsored by the ASBMR and “introduced the
problem to ethicists, solicited their responses, and tried to formulate the proper questions to be
67

raised” for follow-up conferences (ASBMR, 2002). The conference examined the specific
question of whether placebo-controlled studies were still ethically acceptable given the
availability of approved osteoporosis agents. The participants, “a working group of ethicists,
clinical trials design experts and clinical investigators” all agreed that trials with placebo controls
were not ethically justifiable if they had substantial risk of serious outcomes (Brody, 2003).
However, participants considered whether placebo-controlled trials might be appropriate if
certain conditions were met: when appropriate informed consent is obtained; when enrollment is
limited to subjects who are refractory to approved therapies, and/or when such trials are designed
with provisions to transition subjects with significant bone loss or fractures onto an approved
osteoporosis agent (Brody, 2003). However, even under these stringent criteria, some ethicists
remained concerned that patients were still vulnerable to the serious adverse consequences of
their untreated disease (Brody, 2003).
Following the Omaha meeting, the World Medical Association issued a clarification of DoH
Paragraph 29 stating that placebo-controlled studies may be justified by “compelling and
scientifically sound methodological reasons” or in cases where the indication can be considered a
“minor condition and the patients who receive placebo will not be subject to any additional risk of
serious or irreversible harm” (Brody, 2003). Since both of these conditions assume a presumably
lower risk to subjects than that afforded subjects with severe PMO, neither condition diminished
the ethical concern and the debate continued.
Capron’s Osteoporosis Panel Summary summarized the proceedings of the second of these two
conferences, the Natcher conference co-sponsored by NIH and ASBMR in June 2002 in
Bethesda, Maryland. The session involved broad participation including “clinical investigators;
epidemiologists; representatives of government regulatory agencies, industry, and Institutional
Review Boards; government scientist; ethicists; and concerned lay-persons to address critical
68

issues in the development of new therapies for osteoporosis to prevent bone fractures” (Capron,
2003). Again at issue was the impact of the Revision to the DoH (Kanis, 2003), the outcome of
which implied that “placebo-controlled trials cannot be conducted with subjects who have
osteoporosis as defined by the World Health Organization (WHO), because there is a ‘current
therapeutic method’ for this disorder,” (Capron, 2003). Prospective patients, as of the late 1990s,
had multiple treatment options and could now decide between several approved and efficacious
anti-fracture efficacy agents (i.e., anti-resorptives and SERMs). This made it increasingly
questionable from an ethical standpoint to enroll patients at high risk for preventable osteoporotic
fractures into a trial where they might be randomized into a placebo group.
Capron highlights the dilemma in which drug sponsors find themselves: the FDA and EMA
guidelines generally recommend a single, randomized, pivotal, placebo-controlled study to
support product registration. However, as Capron and others point out, Institutional Review
Boards (IRBs) often refused to approve some clinical research studies that otherwise appeared to
meet the bar for scientific and regulatory acceptability. Although Capron summarized certain
alternative study design approaches proposed by the participants, these were acknowledged to
raise additional ethical issues. The outcome of the conference appeared only to succeed in
“framing the issues for future dialogue”.
At the same time, a paper published in Europe, entitled “Study Design in Osteoporosis: A
European Perspective” (Kanis, 2003), provided a different view. It was developed with the
participation of representatives from industry, national health authorities (NRAs) and other health
organizations (i.e., Belgium Ministry of Public Health, Medical Evaluation Board in The Hague,
French Regulatory Authority – AFFSAPS, the WHO Collaborating Centre for Metabolic Bone
Diseases, etc.). Recommendations included the continued use of the placebo-control design since
alternatives such as an active comparator methodology “demand prohibitive sample sizes and
69

carry a high risk of being invalid,” although the strength of an active comparator design “permits
unambiguous demonstration of efficacy” (Kanis, 2003).
Following these conferences, the FDA convened a public Advisory Meeting of the Division of
Endocrinologic and Metabolic Drugs (25 September 2002), entitled Standards of Evidence for
Approval of Drugs for Prevention and Treatment of Osteoporosis (FDA, 2002a). During this
meeting, whose primary goal was to consider whether anti-fracture efficacy should continue to be
required as an efficacy outcome, other aspects of clinical trial design were also considered. One
conclusion of that meeting was a stated position “that placebo-controlled trials, with drug or
placebo provided as an add-on therapy to calcium and vitamin D, could continue in women with
lower short-term fracture risk” (Kehoe, 2006). However, “to study high-risk subjects, active
control trials would be needed, with superiority, rather than noninferiority, as the endpoint of
choice” (Kehoe, 2006). Despite these recommendations, no significant changes to the FDA’s
published guidance resulted, and uncertainty remained.
In 2004, the FDA issued a notice in the Federal Register (FR), seeking specific comment from the
public on the appropriateness of overall “trial duration, trial design, use of intermediate endpoints
as outcomes, and use of placebo-controlled trials in high-risk patients” (Silverman, 2008b). The
specific questions in the 2004 FR were as follows:
(1) Is it appropriate to continue to use placebo controls in fracture end-point trials?
(2) Do fracture end-point trials need to be 3 years in duration, or could shorter studies provide
adequate evidence of a new osteoporosis drug’s effectiveness and safety? (Silverman, 2008b)
Although it was expected that a final FDA guidance would be issued following the appearance of
the FR notice, no such revision to the FDA guidance has appeared. However, the EMA did revise
their guidance (2006), providing fresh regulatory perspectives on both questions posed in the
70

2004 FR statement. The result of the current status quo has been the unintended consequence of
a gradually escalating challenge to the enrollment of higher-risk PMO subjects into placebo
controlled studies. It has also created a de facto requirement to conduct a separate active-
controlled study in higher-risk subjects should the sponsor wish to obtain broad product labeling
that includes both lower and higher risk study populations. Thus, although FDA and EMA
guidance and precedence had allowed a single adequate and well-controlled study to support
registration, the question of whether two studies would be needed to achieve approval for
prevention and treatment of PMO was now raised, and is still unanswered.
The most contemporary and influential of the consensus papers, published in 2008, was titled
Recommendations for the Clinical Evaluation of Agents for Treatment of Osteoporosis:
Consensus of an Expert Panel Representing the American Society for Bone and Mineral Research
(ASBMR), the International Society for Clinical Densitometry (ISCD), and the National
Osteoporosis Foundation (NOF) (Silverman, 2008b). Its recommendations were an outgrowth of
the work of a consensus panel comprised of “nine physicians and two PhDs…representing a
diversity of specialties (five endocrinologists [one a specialist in legal issues and one a specialists
in nutrition], two rheumatologists, two ethicists, one clinical trialist, and one biostatistician),”
identified by recommendation to a joint committee of the sponsoring organizations. The
consensus conference concluded that (1) placebo-controlled trials should include both vertebral
and non-vertebral endpoints, and (2) trials may be shortened to 18-24 months for demonstrating
efficacy (discussed below in Chapter 2.4.2). Importantly, the panel recommended that “placebo-
controlled trials with fracture endpoints” were appropriate for registration purposes and were
ethical if proper informed consent was obtained and if limited in use to a high-risk patient
population. The panel reaffirmed the FDA’s long-held position that “BMD or BTMs alone were
not appropriate for initial registration” for drugs with a new mechanism of action, but were less
71

certain whether fracture efficacy should continue to be required for “a new anti-catabolic agent
with a mechanism of action similar to approved drugs such as a bisphosphonate or selective
estrogen receptor modulator (SERM)” (Silverman, 2008b).
2.4.2 Treatment Duration for Pivotal Trials
A second concern of the FDA that contributed to its 2004 FR notice was also addressed by the
2008 Silverman consensus paper: defining the appropriate treatment duration for pivotal trials.
The Silverman paper answered this call by recommending that pivotal clinical efficacy trials be
shortened to 18-24 months (followed by a controlled study of overall safety and continued
reduction in fracture risk for at least 4-5 years). This recommendation is aligned with neither the
FDA’s 1994 Draft Guidance (which recommends 3 year treatment period) nor the 2007 EMA
Revision 2 Guidelines (which recommends a 2 year treatment period). Interestingly, the FDA
and EMA have in fact approved, under extenuating circumstances, a single osteoporosis agent
relying upon a phase 3 pivotal trial with a treatment period that is much shorter than 3 years
duration. This case, involving the bone anabolic (osteoanabolic) agent teriparatide, was an
exception to the rule at the time of marketing authorization in both the US and EMA since both
regions required a 3 year fracture efficacy trial at that time (it was only later that the EMA would
revise their guidelines, effectively reducing the recommended treatment duration from 3 to 2
years).
The marketing authorizations for teriparatide (also known by the tradename Forteo in the US, and
Forsteo in the EU) were granted based upon analyses of a clinical data package with included a
single long-term phase 3 study comprising only 19 months median exposure to treatment (ranging
18 to 23 months) (FDA, 2002b; EMEA, 2004b). This precedent is often cited as a key argument
against the FDA’s standard recommendation for 3 year treatment data for all new osteoporosis
products. However, the regulatory approval of Forteo/Forsteo was a unique circumstance. In this
72

case, an increased incidence of dose dependent osteosarcoma was observed in an ongoing 2-year
rat carcinogenicity study as phase 3 testing was in progress. The osteosarcomas were observed in
some cases “at exposures (AUC) equivalent to those commonly used in clinical studies of PTH in
treatment and/or prevention of osteoporosis” (FDA, 2000). These findings were considered
‘biologically plausible’ because of the known hormonal effects of Forteo on the parathyroid gland
(FDA, 2001). As a result, the sponsor (Eli Lilly & Co.) voluntarily halted all ongoing clinical
studies with the product as a precautionary measure, including its prospectively planned 3-year
pivotal study.
Following further review of the abbreviated clinical dataset, the treatment duration of which
reflected the time of the unplanned termination of the phase 3 program (FDA, 2001), marketing
authorization was granted in 2002 (Forteo in the US) and in 2003 (Forsteo in the EU). Positive
assessments by the FDA and EMA could be made based on a favorable benefit:risk assessment
that weighed the unique benefits of this new molecular class (an osteoanabolic agent with robust
and rapid bone forming activity) versus the potential and unknown risk of osteosarcomas in
humans. However, these marketing authorizations did not come without certain restrictions to the
labeled indications in both regions – responses which were notably aligned between the 2 regions.
Firstly, the approved indication of use was limited to postmenopausal women at high (US) or
increased (EU) risk for fracture, which was “defined as a history of osteoporotic fracture, multiple
risks factors for fracture, or patients who have failed or are intolerant to other osteoporosis
therapy” (FDA, 2009c). Secondly, the approved labeling limited the recommended treatment
duration to 2 years in the US and EU (FDA, 2002b). Additionally in the US, the FDA required
“Boxed Warnings” and a “Medication Guide” to highlight the potential risks relating to the
nonclinical finding, and required that a postmarketing surveillance study of the drug be conducted
(FDA, 2002b). Essential to the benefit:risk balance was the fact that no osteosarcoma had been
73

observed during clinical trial testing. Thus the risk to humans was theoretical and could “be
adequately managed in labeling,” and through “agreed upon commitments for post-marketing risk
management by the firm” (FDA, 2002b). The approved US package insert for Forteo contains a
boxed warning explicitly warning of the “uncertain relevance of the rat osteosarcoma finding to
humans,” and that Forteo should only be prescribed in “patients for whom potential benefits
outweigh potential risks” (Forteo, 2009). Recently, data through year 7 from an ongoing 15-year
US postmarketing surveillance study were published online in the Journal of Bone and Mineral
Research indicating that a positive association between teriparatide use and osteosarcoma
incidence in humans has yet to be established (Andrews, 2012).
The unique exception posed by PTH agents challenged the established US and EU regulatory
frameworks. When the EMA reduced its recommended treatment duration for pivotal PMO
clinical trials from 3 to 2 years in its 2007 Revision 2 guidelines, it is likely that the teriparatide
precedent was central to this shift in thinking. Shortly following the approval of Forteo in the
US, the FDA’s 2004 FR notice was published seeking expert advice from the osteoporosis
medical community on this question. This precedent has also paved the way for a potentially new
approval standard for PTH-like drugs with robust osteoanabolic effects, a paradigm that may
allow for pivotal PMO clinical studies of significantly shorter duration.
2.5 Recent Trends in a Changing Regulatory Landscape
In December 2009, the FDA’s 1994 Draft Guidance was abruptly withdrawn (FDA, 2009b), and
has not been followed by a final guidance document. Shortly thereafter, in 2010, the New
England Journal of Medicine published contrasting perspective pieces on topics nearly identical
to those in the 2003 JBMR supplement and the 2008 Silverman manuscript highlighting the
continuation of ongoing controversies surrounding the conduct of PMO clinical studies (Rosen,
2010; Stein, 2010). These events demonstrate minimal progress towards consensus in the
74

medical community. It might be speculated that dissonance in the community of osteoporosis
experts has been at least in part responsible for the delayed release of a revised US guidance
document. According to the FDA’s webpage for planned guidance documents in 2013, a revision
is not yet planned for publication during the remainder of calendar year 2013 (FDA, 2013b).
An examination of the two points of view represented in the recent NEJM publications can
illuminate the current state of the controversy over clinical trial design. Stein and Ray hold the
position that placebo-controlled studies with fracture end points in patients with osteoporosis will
nearly always be unethical, so alternative study designs to the standard active-comparator and
placebo-controlled designs must be considered. Suggested alternatives include clinical studies in
osteopenic patients (a condition resulting in bone loss, but less severe than osteoporosis itself)
rather than in higher risk osteoporosis subjects, “add-on study designs, in which a placebo or new
drug is added to the best available therapy,” or studies with “informed refusal, in which patients
have either refused currently available therapies, cannot tolerate them, or did not benefit from
them” (Stein & Ray, 2010). Stein and Ray proceed to reject these alternatives, arguing that (1) it
is invalid to “extrapolate findings in patients with osteopenia to those with osteoporosis,” that (2)
add-on designs provide only “limited clinical relevance” since they do not reflect how the product
will be used, and (3) informed-refusal designs have “an inherent conflict of interest” since “the
same clinician who advises the patient about the currently approved and available therapy may
recruit them for a trial that require refusal of such therapy” (Stein & Ray, 2010). The article
leaves the reader with little choice but to accept studies involving active comparator designs over
the alternatives (and placebo controls), despite the fact that such study designs “will be more
complex and expensive, usually require larger sample sizes, and have more methodological
complexities than placebo-controlled trials” (Stein & Ray, 2010).
75

In contrast, Rosen and Khosla argue that the ethical issue of utilizing placebo-controlled studies is
more complex than simply framing the question as one “of whether people who are highly
susceptible to an osteoporotic fracture should be involved in a clinical trials and thereby risk
random assignment to a placebo” (Rosen & Khosla, 2010). Rosen and Khosla offer the argument
that placebo-controlled studies are uniquely equipped to expose rare side effects. Moreover, in
the manner that placebo-controlled studies are typically conducted, they generally do not involve
women at high risk for fracture (defined as “women with a history of a fragility fracture of the hip
or spine, a very low bone mineral density (T score less than -2.5)”) (Rosen & Khosla, 2010). In
fact, it is argued that placebo-controlled trials, in their strictest sense, are seldom conducted on
osteoporotic trial participants because calcium and vitamin D are administered to both the control
and experimental groups (Rosen & Khosla, 2010). Also, they contend that noninferiority and
superiority study designs require greater numbers of patients to assess statistically marginal
treatment differences, and thus result “in more fractures over the duration of the trial than would
be observed in a placebo-controlled study and , thus, it would offer no ethical advantage” (Rosen
& Khosla, 2010).
The arguments of Rosen and Khosla (2010) make the point that placebo-controlled studies may
provide a less confounded evaluation of safety and may expose fewer subjects to the unknown
risks of an investigational therapy. However, the justification for such studies becomes tenuous
for studies with durations of more than two or three years during which the rate of vertebral and
non-vertebral fracture is expected to increase, resulting in an associated increase in morbidity and
premature mortality that is associated with low-trauma osteoporotic fractures (Bliuc, 2009).
Active-comparator studies may be more ethical from the standpoint of the individual prospective
patient, and allow a cleaner head-to-head comparison to the most appropriate standard of care.
However, there is no accepted approach to define non-inferiority margins, an estimation of which
76

relies on data of historical fracture rates in the intended population with the comparator. Thus, at
the time of this writing, the debate regarding whether the placebo-controlled trial design is the
standard for establishing efficacy and safety for PMO agents still continues. The clinical
programs for the next two PMO agents in line for approval give a mixed reading. Odanacatib,
which is scheduled to be submitted to the FDA and EMA in 2014, will be based on a single large
(>16,000 patients), placebo-controlled phase 3 study (ClinicalTrials.gov No. NCT00529373).
However, the applications submitted in 2012 to the FDA and EMA for Aprela (a combined
bazedoxifene and conjugated estrogen) are based on a phase 3 double blind, randomized,
placebo- and active-controlled efficacy and safety study, which includes a conjugated estrogen
control arm (ClinicalTrials.gov No. NCT00242710).
2.6 Status on Developing a Common Benefit:Risk Framework
Throughout a product’s lifecycle, including the post-marketing period, a product’s “benefit:risk”
profile (CIOMS, 1998) is refined further from data collected from broader populations exposed to
the drug. A benefit:risk decision is multifaceted, and must reflect not only the clinical experience
with the product, but also the drug’s intended use, the severity of the indication (for instance,
diseases that are “life-threatening and severely debilitating”), the availability of alternative
therapies (FDA, 2004) and the proposed place in therapy of an investigational product, such as
when a product addresses an unmet medical need (FDA, 2004; CIOMS, 1998). The benefit:risk
assessment may also consider a range of potential off-label uses (including abuse potential).
When the benefit:risk balance of an investigational agent is sufficiently positive (i.e., equipoise),
DRAs are mandated under their own legislative requirements to facilitate its development. Thus
benefit:risk not only influences ‘if’ a drug product is approved, but also the ‘when’ (how long the
approval process will take ) and ‘how’ (whether it will have access to priority rather than standard
approval mechanisms). However, the complexities of the decisions regarding benefit:risk, the
77

introduction of independent value judgments of experts, differing legal and regulatory
frameworks, different cultural norms and attitudes towards relative benefits and risks, and the
lack of a common benefit:risk review framework (or BRF) all contribute to divergent regulatory
review decisions.
As discussed in Chapter 2.3 (Regulatory Dissonance in PMO Drug Approval Precedents), it is
apparent that regional perceptions of benefit versus risk differ significantly between the FDA and
EMA for etidronate disodium, tibolone, strontium ranelate, intact PTH 1-84, bazedoxifene, and
lasofoxifene in postmenopausal osteoporosis. Initiatives by various regulators and industry
groups have recently initiated dialogue to inform a harmonized BRF as a means to facilitate
harmonized review outcomes. The assumption is that if the BRF and benefit:risk decision-
making procedures can be harmonized, regulatory decisions across regions may begin to
converge. Researchers at the Centre for Innovation in Regulatory Science (CIRS), and the
Consortium for Benefit-Risk Assessment (COBRA), an inter-agency initiative involving Health
Canada, Australian Therapeutic Goods Administration, Swissmedic and the Singapore Health
Science Authority, have piloted new methodologies for benefit:risk evaluation, with the
assumption that a more structured, quantitative and standardized approaches for benefit:risk
assessment may act to reduce the variability observed in some regulatory opinion outcomes
(Walker, 2011a; Walker, 2011b; Liberti, 2010). Pro forma templates are being developed for
benefit:risk assessments that utilize a stepwise, qualitative approach to decision criteria, using
weighted value judgments. For instance, the CIRS UMBRA (Unified Methodologies for Benefit-
Risk Assessment) initiative has proposed a basic eight-step benefit-risk assessment framework to
industry and regulatory stakeholders which has received initial endorsement by participating
stakeholders (UMBRA, 2013). These procedures are being tested on proprietary datasets
78

acquired from pharmaceutical companies, with the goal that this framework can be further
developed into more metric-based, quantitative methodologies (Liberti, 2010).
The FDA and EMA are initiating their own independent B-R methodology initiatives. These
frameworks appear to be “conceptually similar” despite being developed under different
legislative schemes (Walker, 2011b), thus giving hope that the various benefit:risk work streams
may at some point converge. However, Walker acknowledges that “cultural differences…in the
acceptability of risks among Europeans and US populations” and the level of data received by
each agency (i.e., “the availability of patient-level data to the FDA”) may be factors to consider
when considering divergent regulatory review outcomes. In addition, the constraints of
jurisdictional regulatory procedures and the need to address diverse clinical guidelines and
labeling requirements were noted as challenges to an approach over reliant on harmonized
benefit:risk assessment (Walker, 2011b). Clearly, it is not possible to align regulatory review
outcomes without considering the complexities introduced by different medical traditions,
different regulatory bodies and procedures, different review processes, and importantly, different
indication-specific guidance and review mechanisms.
As noted above, the FDA began its own reassessment of its benefit:risk procedures as part of the
PDUFA V (2013-2017) Implementation Plan in 2009. The result was a report published
February 2013 calling for changes in its benefit:risk decision-making paradigm, entitled
“Structured Approach to Benefit-Risk Assessment in Drug Regulatory Decision-Making (FDA,
2013c). In that report, the FDA rejects the notion of a fully quantitative approach to benefit:risk
methodologies, given that regulatory and medical decisions are often highly conditional, involve
subtle analysis and are rarely binary. FDA’s view is that the process of weighting factors in any
fully quantitative approach is itself inherently subjective, and may also reduce the transparency
into the assumptions that underlie a particular benefit risk decision. FDA’s proposed benefit:risk
79

review template is regulatory elegance defined: the template includes just 5 main review areas to
be considered. Two areas cover the therapeutic context for intended product (Analysis of
Condition and Unmet Medical Need), and the three following considerations address the specific
efficacy and safety data of the product, and the intended use of the product (Clinical Benefit, Risk,
and Risk Management) (FDA, 2013c). The result is that all the main areas considered by the
regulatory agency, including significant underlying assumptions, are clearly communicated to the
audience. A potential downside of this qualitative approach is that each regulatory agency, using
the same benefit:risk template, may notwithstanding reach distinct review conclusions. The FDA
benefit:risk methodology will increase the transparency of product review communications, and
as such may inform regulatory decisions by other agencies, but may not be a significant factor in
reducing regulatory dissonance.
2.7 Relative Benefit:Risk of Available Therapies for Postmenopausal
Osteoporosis
Any discussion of regulatory dissonance and the benefit:risk of available PMO therapies is
wholly inadequate without a discussion of the relative benefit:risk of available PMO therapies.
Table 5 summarizes the key adverse drug reactions observed from clinical trials and
postmarketing safety surveillance, as defined in the prescribing information for approved PMO
agents. Generally, safety events of interest may be observed at any stage of product life-cycle,
thus introducing further unpredictable factors that influence regulatory dissonance. In the thesis
survey described in Appendix I, a question introducing this concept refers to these factors as
“intrinsic factors,” which was designed to be an all-inclusive term that alludes to the unique
factors which make up the benefit:risk profile of a given product or class of products.
For instance, with antiresorptive agents, the mainstay of osteoporosis treatment, rare yet serious
events, such as osteonecrosis of the jaw, atypical femur fracture and esophageal cancer, have
80

been observed in the postmarketing setting (Whitaker, 2012). These have fundamentally shifted
the public’s perception of these products and the utilization of these products by primary care
physicians. The result is a continuing trend in PMO prescribing information towards proscribing
use to more narrowly defined sub-populations at relatively higher risk of fragility fracture, which
may include (1) placing “limitations of use” on use in osteopenic populations (i.e., younger
women without evidence of prior fracture), (2) requiring ‘drug holidays’ where intermittent
treatment breaks are mandated, or (3) placing restrictions on the duration of therapy (i.e.,
treatment discontinuation) (Whitaker, 2012) (Adachi, 2011). This trend towards greater
restriction on use is balanced by a paucity of robust clinical trial data, at least for some PMO
agents, demonstrating that therapeutic gains in bone mineral density are in fact maintained
following product withdrawal (Whitaker, 2012) (Adachi, 2011). It is also balanced by the fact
that osteoporosis, being a chronic treatment, requires treatment intervention. Thus, informed
decisions on treatment restrictions must be made carefully.
An FDA Advisory Committee was convened 09 September 2011 to discuss the optimal duration
of therapy with anti-resorptive agents in light of the rare yet serious safety risks associated with
increasing dose or cumulative long-term exposure to anti-resorptive therapies (Table 5) (FDA,
2011a). This committee cited emerging evidence suggesting an association between these
adverse events and dose level and/or long-term exposure (>3-5 years) to anti-resorptive therapy.
The committee also recommended that the current “Important Limitation of Use” statement on
bisphosphonate labeling (presented below) be updated to clarify (1) the optimal duration of use,
and (2) the frequency of patient re-evaluation due to changing individualized benefit:risk status
over time;
“The safety and effectiveness of [drug] for the treatment of osteoporosis are based on clinical
data of [xx] years duration. The optimal duration of use has not been determined. All patients on
81

bisphosphonate therapy should have the need for continued therapy re-evaluated on a periodic
basis.” (FDA, 2011a)
However, without robust fracture efficacy data that have specifically evaluated the benefits
(fracture efficacy) versus the risks (rare events) beyond >3-5 yrs., the committee could not
endorse further restrictions on duration of use (Whitaker, 2012) (FDA, 2011a). It is yet to be seen
whether concern over rare but serious adverse events will prompt regulatory authorities to require
longer and larger clinical registration studies for future investigational PMO agents, but
momentum in this direction may be gaining ground. In April 2013, the FDA took action by
updating class limitation of use wording affecting all anti-resorptive prescribing information. The
revised wording is remarkably consistent with prescient wording presented in the 2012 Whitaker
article:
“Patients at low-risk for fracture should be considered for drug discontinuation after 3 to 5 years
of use. Patients who discontinue therapy should have their risk for fracture re-evaluated
periodically.”
Theoretical or known safety signals of concern are not exclusive to the anti-resorptive class of
PMO agents, as can be seen in the precedents cited in Chapter 2.3. For approved PMO agents,
one can only review the Warnings and Precautions, Contraindications, and Adverse Drug
Reaction sections of current labeling to understand this point (Table 5). Thus, identifying patient
subgroups and treatment regimens that optimize the therapeutic benefit:risk for PMO products is
critical to guiding treatment decisions by the treating physician. Future clinical trial guidance
will likely be affected by this current gap in knowledge regarding long-term benefit:risk. Since
this question cannot be thoroughly informed through postmarketing surveillance or post-hoc
subgroup analysis alone, clinical trial guidance will likely evolve to inform these important
questions for future PMO agents.
82

The trend towards longer and larger clinical trials will also likely be fueled by regulators’
increasing interest in highlighting and communicating important risk information of rare events
upon long-term treatment in product labeling in the face of the current lack of robust, long-term
clinical data that can more fully characterize these risks. This trend towards lower tolerance of
product safety issues will also lead to greater product development attrition as the threshold for
acceptable safety is lowered, as R&D expenditures towards PMO product development increases
as a result of larger, longer trial work, and companies become less willing to accept additional
development risk in this setting. Moreover, this trend is expected to contribute towards continued
dissonance as DRAs grapple with these important concerns one by one.
Table 5 summarizes the main safety concerns with available PMO therapies, as derived from
current prescribing information posted on the web pages of the regional regulatory authorities
(FDA’s drugs@FDA webpage, the European Medicines Agency webpage, and the MHRA’s
electronic Medicines Compendium (eMC) webpage; accessed: July 2013).
83

Table 5. Key Adverse Drug Reactions from Clinical Trials and Postmarketing Safety
Reporting Therapeutic Product Class of Drugs for Postmenopausal Osteoporosis
Therapeutic Product
Class
Significant Safety Events and Limitations of Use Per
Approved Product Labeling
1

1
st
Line Postmenopausal Osteoporosis Treatments
Bisphosphonates
Atypical femoral fractures, ONJ, esophageal cancer,
musculoskeletal pain, hypocalcemia, renal impairment
Limitation of use statement: “The optimal duration of use has not
been determined. All patients on bisphosphonate therapy should be
have the need for continued therapy re-evaluated on a periodic basis”
Parathyroid Hormones
3
Osteosarcoma
2
, orthostatic hypotension, exacerbation of urolithiasis
RANK Ligand
Inhibitor
3

Atypical femoral fractures, ONJ, hypocalcemia, serious infections,
dermatologic adverse reactions, suppression of bone turnover
2nd Line Postmenopausal Osteoporosis Treatments
Selective estrogen-
receptor modulators
(SERM)
Cardiovascular disease, venous thromboembolism, death due to
stroke, hepatic impairment, hypertriglyceridemia, concomitant use
with systemic estrogens not recommended, use in premenopausal
women not recommended
Estrogen and Estrogen /
Progestin Combinations
Endometrial Cancer
2
, Cardiovascular Disorders
2
, Breast Cancer
2
and
Probable Dementia
2
, gallbladder disease, visual abnormalities,
anaphylactic reaction and angiodema, exacerbation of hereditary
angiodema
Strontium Ranelate
Venous thromboembolism, myocardial infarction, severe allergic skin
reaction
Tibolones
Endometrial hyperplasia and cancer, breast cancer, ovarian cancer,
venous thromboembolism, coronary heart disease (CAD), ischemic
stroke, may enhance effects of anticoagulants, drug-drug interactions
with CYP3A4 substrates
Salmon Calcitonins
EMA: Benefit-risk in PMO no longer positive, PMO indication
withdrawn (2012) due to increased incidence of cancers
FDA: A March 2013 FDA Advisory Committee was convened to
discuss the benefit-risk of salmon calcitonin in the PMO indication;
recommendations of the FDA panel consistent with EMA decision to
withdraw products from market. No further FDA action has resulted
to date
1

Safety summarized from publically sourced prescribing information on the FDA Drugs@FDA website
(http://www.accessdata.fda.gov/scripts/cder/drugsatfda/), European Medicines Agency website
(http://www.ema.europa.eu/ema/) or from the electronic Medicines Compendium (eMC) (www.medicines.org.uk)
2
Boxed Warnings in current US prescribing information
3
Products limited to an indication “for the treatment of postmenopausal women with osteoporosis at high risk for
fracture, defined as a history of osteoporotic fracture, or multiple risk factors for fracture; or patients who have failed or
are intolerant to other available osteoporosis therapy.” Other products listed are indicated for the treatment of
postmenopausal women with osteoporosis.
84

2.8 Status of Global Harmonization of Indication-specific Guidance
DRAs from the major ICH regions are beginning to converge on certain aspects of regulatory
review and to share information on certain elements of the review, such as safety. However the
substance of such interactions is generally not transparent to drug sponsors and little evidence
exists to suggest that product-specific reviews are being coordinated. That being said,
harmonization initiatives have been underway for the last two decades. The US FDA, European
Medicines Agency (EMA) and the Japanese Regulatory Agency (PMDA), including participation
from Health Canada and other DRAs, have since 1990 participated in a series of initiatives to
harmonize different aspects of requirements in the drug approval process through the
development of a series of working groups under a joint organization called the International
Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for
Human Use (ICH). Since its formation, ICH has successfully developed guidelines that have
harmonized many aspects of the drug development process, including for example clinical (GCP)
and manufacturing (GMP) standards, aspects of nonclinical testing (GLP), generalizable
requirements of clinical data packages, and dossier format (i.e., CTD format). The scope of these
activities to date has been focused primarily on broad, practical aspects of product registration.
For instance, an early and obvious success of ICH was the implementation in 2003 of the
Common Technical Document (CTD), a dossier format that has reduced the administrative
burden associated with multiregional regulatory filings for sponsors, and has facilitated faster
reviews and information exchange between regulators (ICH, 2010). Despite the adoption of ICH
guidelines, significant differences remain in the regional interpretation and implementation of
ICH guidance. Further, the ICH has been less active in the harmonization of technical standards
relating to specific disease indications or drug classes. In part, this may relate to the inclination
of sovereign nation states to retain governance over issues critical to the safety and efficacy of
85

their country’s drug supply. In order to align indication-specific guidelines across regions, an
unprecedented level of cooperation between DRAs would be needed, followed by an
unprecedented synchronization in the legal basis for the oversight of drug review/access/supply
across multiple regions. Other harmonization approaches, such as Mutual Recognition
Agreements (MRAs) between sovereign DRAs also offer potential mechanisms to streamline
access to respective markets, through a process of respecting prior authorizations of other DRAs
and encouraging information exchange. However to date, although such agreements now exist
between specific regions, they have not yet been extrapolated globally for purposes of global
harmonization of product-specific guidance and advice for individual products (Michor &
Ochalski, 2010). More recently, President Obama “issued an Executive Order, Promoting
International Regulatory Cooperation that increases the visibility of agencies’ interest in
international regulatory harmonization” (Beck, 2012). At the time of this writing, it is not clear
how this order will impact FDA’s activities as it relates to drug review and approval procedures.
The current state of regulatory dissonance amongst global regulators with respect to indication-
specific guidance can be addressed only through an unparalleled and unprecedented level of
cooperation between national/regional DRAs. Paradigms for this level of cooperation and
transparency have recently begun to emerge in the European Union (EU) through programs such
as the Voluntary Harmonisation Procedure (VHP) pilot program (HMA, 2010), which mitigates
the need for drug sponsors to meet separately with multiple competent authorities to obtain advice
on key pivotal clinical trials. The activities of the International Conference on Harmonization
(ICH), which support the harmonization of procedural guidelines across the tripartite region, offer
another paradigm for inter-agency cooperation. Together, these two examples could be used to
build a framework of transparent inter-agency review mechanisms, upon which harmonization of
indication-specific guidance and advice may be established. In another EU procedure, the
86

Community Referral, the “European Medicines Agency is requested to conduct, on behalf of the
European Community, a scientific assessment of a particular medicine or a class of medicines,
including cases of in which decisions diverge,” thus “leading to an opinion by the agency that
provides that basis upon which the European Commission issues a single, uniform decision”
(Michor & Ochalski, 2010). A similar system could be employed to resolve inter-agency issues
of regulatory dissonance beyond the EU.
There is evidence that DRAs may be moving in the direction of greater information sharing and
cooperation, at least between the US FDA and EMA. For example, in the 2011 “Pilot EMA-FDA
GCP Initiative” (EMA-FDA, 2011b), the EMA and FDA intend “to share information on
inspections and GCP-related documents of common interest and to conduct collaborative
inspections”. Other initiatives between the FDA and EMA to accept common submission
formats, such as the “Single Orphan Drug Designation Annual Report” (FDA, 2011b) and the
ICH E2F guidance "Development Safety Update Report” (FDA, 2011b) are also increasing as the
benefits of reduced cost, better information sharing, and more efficient approaches to submission
dossiers are experienced by regulators and sponsors alike. Moreover, the FDA has held a yearly
forum since 2006 entitled “CDER Forum for International Drug Regulatory Authorities” (FDA,
2012) to facilitate communication amongst regional regulators. Again, these efforts focus on
“low hanging fruit” such as initiatives to create single reporting formats, to share information via
joint inspections and compliance audits, and to harmonize relatively straightforward requirements
relating to nonclinical or chemistry, manufacturing and controls data. There even appears to be a
concerted effort to share administrative burden on a broad level, as evidenced by the
“Transatlantic Administrative Simplification Action Plan,” the key objective of which is “to
identify opportunities for administrative simplification through transatlantic cooperation at the
level of administrative practices and guidelines” (EMA-FDA, 2011). The activities described in
87

this report follow the workshop on ‘Medicines Regulation Transatlantic Administrative
Simplification Action Plan’ which was published June 2008 and identified “15 administrative
simplification projects to be taken forward (EMA-FDA, 2008). Two guiding principles for this
initiative were identified: (1) “No change to legislation should be required,” and, (2) “The
simplification should maintain or increase current levels of public health protection” (EMA-FDA,
2008). A few of the administrative simplification projects included in this initiative offer a
glimmer of hope for the harmonization of product-specific review, including projects such as the
proposed collaborations on risk management, parallel transatlantic scientific advice, and advance
therapy medicinal products.
As recently as 2009, the EMA and FDA joined efforts to develop a combined EMA-FDA parallel
scientific advice program. Although the program is being piloted in limited areas, termed
“clusters of interest” (i.e,, applied to limited areas such as advanced therapies, orphan drugs, new
technologies, etc.), the goal is promising: “to optimize product development and avoid
unnecessary testing replication or unnecessary diverse testing methodologies” (EMA-FDA,
2009). The program, as designed, however does not make inroads to curb the independent
decision-making powers of each Agency, as each retains the right to issue independent guidance.
Thus, a significant limitation of this program is that there is no provision of mandating alignment
of the advice provided, and as a result, the possibility exists that not just two, but three different
sets of recommendations could result (parallel advice plus two independent regional opinions). In
addition, drug sponsors are not entitled to receive parallel scientific advice if requested, as access
to such advice is currently at the discretion of the EMA and FDA.
Regulatory harmonization should not be limited to ICH regions or even the industrialized world,
because the industry trend is that MRCTs are increasingly global in conduct. In a recent RAPS
Regulatory Focus supplement on Global Harmonization Initiatives, Grignolo considered
88

harmonization efforts beyond the US and EU (Grignolo, 2010). He noted that the Global
Cooperation Group (GCG), under ICH, is working toward one useful goal of developing
partnerships, called Regional Harmonisation Initiatives (RHIs) that would include not only ICH
members but also specific nonmember countries (Grignolo, 2010). Two RHIs that now exist in
Asia and Latin America include the (1) ASEAN Consultative Committee for Standards and
Quality (ACCSQ) Pharmaceutical Product Working Group (PPWG) and the (2) Pan American
Network on Drug Regulatory Harmonisation (PANDRH). The activities of the PPWG support
the common economic goals of the nascent ASEAN Economic Community (AEC) by developing
common regulatory frameworks for drug registration, and by adopting a “wide array of ICH
safety, efficacy and multidisciplinary guidelines” (Grignolo, 2010). Aligned product-specific
guidance may be a long way off because the formation of a single economic community is
scheduled to take effect in 2015 at the earliest. The PANDRH has a similar goal “to improve
access to high-quality, safe and efficacious drugs within the region, and to promote technical
cooperation and sharing of knowledge and experience among more- and less-developed
regulatory agencies” (Grignolo, 2010). The impact of both organizations on regional
harmonization is yet to be observed.
2.9 Framing of Study of Regulatory Dissonance
Regulatory dissonance is a factor that regulatory professionals often consider when developing
integrated global strategies for drug development, but they may not be able to predict the extent
of factors contributing to such dissonance. Understanding the determinants of regulatory
dissonance and the points in development where it presents are important initial steps toward
developing regulatory standards and mechanisms to minimize its impact on product development.
At least three primary stakeholders in the drug development process appear to contribute to
divergent opinions in PMO drug development. These include (1) drug regulatory agencies, (2)
89

the medical community (including clinical investigators, ethics committee members, healthcare
organizations and patient advocates), and (3) drug sponsors. The concept of stakeholder analysis
is not well-developed in regulatory science literature; however it has been used elsewhere as an
organizing principle for understanding and decomposing dissonant opinions, and as a means of
exploring new perspectives and approaches to address policy issues (Brugha, 2000). Up until this
point in Chapter 2, I have primarily focused on the perspectives of the first two sets of
stakeholders which have been published in the form of regulatory guidance and scientific
perspective articles respectively. The perspectives of the third set of stakeholders, the drug
sponsors, are not as well-known outside of industry; nevertheless their real world experience with
the clinical development process is equally important to an understanding of the nature of
regulatory dissonance and its downstream consequences. Thus, the purpose of this exploratory
study is to assess the determinants of regulatory dissonance, and evaluate the consequences of
regulatory dissonance from the perspective of the drug sponsor, which is the stakeholder group
most directly affected by those challenges. It is the drug sponsor who must comply with the often
variable and non-aligned regulatory direction provided by the other two stakeholders.
Knowledge of their opinions and insights may be an important missing link in stakeholder views
which when understood fully might help to shape wider policy initiatives to reduce regional
regulatory dissonance.
90

2.10 Summary and Research Direction
In this exploratory study, several contributing elements to regulatory dissonance as it relates to
indication-specific regulatory requirements have been identified. We have also identified three
primary stakeholders of regulatory standards and policy, but have identified that the views of one
of these stakeholders, the drug sponsors, is not so well defined as the others.
This thesis aims to examine the views of industry concerning the importance of various domains
of contention or divergence that potentially affect PMO drug development: (1) dissonance in
technical guidance found in the US and the EU guidance documents, (2) divergent regulatory
authority feedback during development, (3) divergent drug approval precedents, (4) differing
drug review procedures and benefit:risk methodologies contributing to divergent drug approval
precedents, and (5) dissonance between regulators and a part of the medical community
advocating active- rather than placebo-controlled clinical trials, and advocating different
minimum treatment durations with investigational drug. The investigation will additionally
explore which areas of dissonance are most problematic and which in their opinion would be
fruitful areas on which future harmonization efforts should focus.
91

CHAPTER 3: METHODOLOGY
3.1 Introduction
This is an exploratory, qualitative research study evaluating the current perspectives of clinical
and regulatory professionals in industry who are involved in developing therapeutic agents for the
treatment of postmenopausal osteoporosis (PMO). A live interactive focus group and a self-
administered online survey instrument will be utilized to explore the sources and impact of inter-
regional dissonance on regulatory strategies/considerations for developing PMO drugs.
Methodology will reflect certain design features of stakeholder analyses as described by
Varvasovszky (2000) and Brugha (2000).
3.2 Development of Initial Survey
The online self-administered survey tool used in this case study explored the perceptions and
opinions of the study participants. The questionnaire included 35 questions, and the 6 topic areas
included the following (Table 6):
Table 6. Questionnaire Instrument: Breakdown of Areas of Inquiry
No. Areas of Inquiry
1 Professional profiles of individual respondents
2 Impact/implications of regulatory dissonance
3 General causes/factors contributing to regulatory dissonance
4 Aspects of regional indication-specific PMO guidance most/least aligned
5
Aspects of regulatory dissonance in PMO relating to drug review and approval
procedures
6 Potential policy solutions to reduce regulatory dissonance
92

The appropriateness, understandability and overall structure/organization of the questions were
guided by feedback obtained through a face-to-face, interactive focus group (N =16) convened to
provide input on the online survey instrument prior to implementation. The focus group included
participants from a diversity of backgrounds but who nonetheless had core knowledge regarding
the drug development process or survey methods in regulatory/policy literature (Table 7).
Faculty and doctoral students from the International Center for Regulatory Science at the
University of Southern California (USC) were among the participants; also included were two 2
osteoporosis clinical experts from industry for issue validation specific to the bone field. The
focus group session was co-chaired by Frances J. Richmond, BNSc, MSc, PhD, who is the
Director of the USC International Center for Regulatory Science.
The focus group participants (Table 7) were asked to consider the purpose of the thesis research
when commenting on the survey instrument. Participants were asked to provide input on the
overall organization of the survey and the relevance and clarity of each question, and feedback on
the proposed survey audience (i.e., inclusion/exclusion). The focus group was considered the
ideal method to validate the general ease-of-use of the survey, inform on potential gaps in topics,
and provide feedback on user experience in advance of survey deployment. Written feedback
was also welcomed.
93

Table 7. Description of Focus Group - Professional Backgrounds
No. Description of Participant Professional Backgrounds
1
Hesham Abdullah, MD, MSc, RAC, Vice President, Global Regulatory Affairs,
Oncology, AstraZeneca Pharmaceuticals, LP
2 Dana Bell, PhD, Faculty, USC Regulatory Science Program
3
Arkadi Chines, MD, Endocrinologist, Clinical Research Medical Director,
Amgen Inc.
4
Terry Church, MA, Doctoral Candidate, USC Regulatory Science Program,
Program Manager of the Women’s Cancer Program (WCP) at the USC Norris
Comprehensive Cancer Center
5 Ogo Egbuna, MD, MSc, FASN, Medical Sciences Medical Director, Amgen Inc.
6
Aimee Greco, MS, Doctoral Candidate, USC Regulatory Science Program,
Regulatory Affairs Professional 3M Unitek, a Division of 3M Healthcare
7 Michael Jamieson, DRSc, Adjunct Professor, USC Regulatory Science Program
8 Duane Mauzey, DRSc, RAC, Manager, CMC Scientific Affairs, Allergan Inc.
9
Cesar Medina, MBA, MS, RAC, Doctoral Candidate, USC Regulatory Science
Program
10
Caroline Mosessian, PhD, MHA, MSc, Executive Vice Chair/VP of Service Lines
Strategy and Operations at USC Health System, Executive Hospital Administrator
Orthopedics at USC University Hospital, Vice Chair and Associate Professor at
USC
11
Eunjoo Pacifici, PharmD, PhD, Assistant Professor of Clinical Pharmacy, USC
Regulatory Science Program
12
Ali Reza Rajaei, Director of Clinical Development at POM Wonderful, LLC, a
Division of Roll International Company
13
Valerie Ramsey, DRSc, MS, RAC, Director of Regulatory Affairs at Fleet
Laboratories
14
Session Co-Chair: Frances J. Richmond, BNSc, MSc, PhD, Director USC
Regulatory Science Program, and DIA Board of Directors Member
15
Nancy Smerkanich, Vice President, Regulatory Affairs Consulting, Octagon
Research Solutions
16
Loren Wagner, Senior Director of Pharmaceutical Development Compliance,
Allergan Inc.
94

Because five of the participants attended from remote sites, video conferencing with participant
consent was provided to facilitate the interactive experience of remote participants and the
proceedings were video captured.
Participants were emailed an electronic copy of draft thesis Chapters 1-3 (used in support of the
Qualifying Exam on 27 August 2012), a slide-based summary of the topic, a copy of the sample
online survey instrument, and a set of logistical instructions, 3 weeks in advance of the date of the
focus group. Immediately prior to the formal proceedings of the focus group coffee/snacks were
served, and participants were given the opportunity for introductions.
During the focus group, the Agenda as represented in Table 8 was followed. The moderator
delivered a brief overview describing what participants were to expect during the focus group, the
process to be followed, and the outcomes that were anticipated. Each of the proposed questions
was reviewed by giving a brief period for participant feedback. Once all survey questions were
reviewed, the moderator solicited general feedback on the format and content of the survey
instrument. The moderator then summarized the areas of feedback that were unexpected,
identified new themes which emerged during discussion, and also summarized specific actions to
be taken to refine the survey based on the feedback of the focus group.
95

Table 8. Focus Group Agenda
Agenda
Item
No.
Agenda Item Time Allotted
1
Brief introductions (name, affiliation, area of
regulatory / clinical expertise)
Participants / Moderator (Neal
Storm) – |30 Minutes
2 Brief background on topic
Participants / Moderator –
10 Minutes
3
Line review of survey instrument / discussion of
feedback
Participants / Moderator –
70 Minutes
4
Summary of key areas of feedback and follow-up
action items
Moderator (Neal Storm) –
10 Minutes
3.3 Survey Deployment and Analysis
The final survey was administered electronically to participants from 12 November 2012 through
14 March 2013. The target population for the survey was originally projected to include ≥ 40
industry professionals, and upon completion included 42 participants. Participants were selected
based on defined inclusion criteria: surveyees were required to be currently employed in the
pharmaceutical industry possessing drug development experience, be currently or previously
involved in PMO drug development, and be currently or previously acting as a regulatory or
clinical scientist. Representatives of DRAs and members of IRBs were excluded. Participants
were assured that no specific data regarding the identity of the companies or organizations would
be published should this work result in a publication, although that information would be known
to the principal investigator (Neal Storm). Participants were identified through professional
networks, conferences, referrals by colleagues, and general networking. All were screened for
their willingness to participate by phone/email. No remuneration was provided to encourage
participation, but the respondents were promised a summary of the results obtained after the
survey has been analyzed.
96

The Questionnaire Instrument was administered using the Qualtrics web-based survey product
Research Suite, and the collected data was not analyzed or viewed until all pre-specified
participants had completed the on-line survey. The Qualtrics software was employed to complete
the statistical analysis of the raw data. Given the qualitative nature of this study and its non-
experimental design, analysis was limited to basic statistical outputs, percentages, standard
deviations; variance and means of data were collected from responses tabulated using an ordinal
scale.
Following data analysis, 1:1 follow-up qualitative semi-structured interviews up to 1 hour in
length were held with selected study subjects to discuss key results and findings. These
interviews were conducted to obtain more granular feedback, obtain clarification and follow-up
on certain response areas of interest and new themes. This information was used to further
inform on the study conclusions. Neither the electronic survey responses nor the information
provided through the 1:1 interview procedures were attributed to the individual participants. The
names of the participants and their employers will remain anonymous.
97

CHAPTER 4: RESULTS
4.1 Focus Group
The purpose of the focus group was to maximize the quality and relevance of the information
obtained from the use of the survey instrument. For this reason, the focus group panel was
structured to be intentionally heterogeneous, by including USC Regulatory Science Program
faculty, doctoral students and graduates, and industry professionals with expertise in regulatory
science, clinical development and osteoporosis drug development.
4.1.1 Focus Group Outcome
The focus group was held at the USC International Center for Regulatory Science (ICRS) in Los
Angeles, CA on 21 September 2012. During the 90 minute meeting the focus group participants
raised several questions with respect to different aspects of the survey. In Section I pertaining to
the professional profiles of the respondents, the focus group participants recommended that the
choices for questions querying years of drug development experience (Question 1) and years of
experience developing PMO drugs (Question 2) be made consistent (accepted). The focus group
debated the minimum level of drug-development experience and minimum level of experience
with PMO drug development that might be used as inclusion criteria for study participation. The
argument was made that subpopulations with different experience levels might have different
views, and such differences could be explored through cross-tabulations. Concern was expressed,
however, that respondents with modest experience might lack the depth of knowledge to answer
questions that related to specific aspects of PMO regulatory guidance and precedence. The
participants came to the consensus that the criteria might be loosened to capture a wider range of
experience levels (accepted). One member of the focus group suggested that project managers be
considered as potential survey participants because some questions in the survey related to
98

timelines and costs. This suggestion was not supported by most, who felt that project managers
are not typically familiar with regulatory guidance and procedures. Another member suggested
that the question relating to regulatory dissonance be broken into two separate questions relating
to the impact of regulatory dissonance on delays to clinical development timelines, and regulatory
agency review timelines respectively (accepted). Focus group participants expressed doubts
regarding the proposed use of Likert scales to estimate the value judgments of respondents,
because the sliding scale was considered too subjective; the group favored a more limited series
of 3 options (accepted). Focus group participants also advised that the option of “joint scientific
advice meeting” should be captured under Question 32, and that the study conclusions should
explore the reasons why the new regulatory mechanism suggested in Question 32 has been
implemented in only a limited manner. Participants were interested in whether it was expected
that the results of the survey would be broadly applicable to other therapeutic areas, and this was
also deemed to be a good point to discuss in the Conclusions. The final survey, edited according
to focus group suggestions, is shown in Appendix I.
4.2 Analysis of Survey Results
4.2.1 Responses and Profiles of Respondents
Fifty links to the online questionnaire were disseminated between 12 November 2012 and
14 March 2013, from which 42 complete responses, equating to an 84% response rate, were
obtained. Follow-up emails to respondents through their personal rather than company email
accounts was required for several survey participants whose firms employed ‘Spamware’ that
filtered the emailed links sent automatically from the Qualtrics system. Five additional
respondents launched the survey, but completed only some of the questions (n=3), or none of the
questions (n=2). Four of these respondents were issued new electronic links to the survey (since
the prior link was deactivated as a result of disuse) and proceeded later to complete the survey;
99

one potential participant was removed from the survey pool because that individual lacked any
experience with PMO drug development and therefore did not fit the inclusion criterion for the
study. The eight members of the targeted participant pool who did not complete the survey
declined participation because they felt that they lacked sufficient drug development experience
in postmenopausal osteoporosis (n=4), or failed to respond to emails reminding participants to
complete the survey (n=4).
Respondents were typically well experienced professionally. Most (67%) had greater than
10 years of experience in the biopharmaceutical industry (Question 1). Of the others, 19% had 5-
10 years of experience, 7% had 2-5 years of experience, and 7% had less than 2 years of drug
development experience. This professional experience specifically included direct involvement
in the development of products that treat postmenopausal osteoporosis (PMO) (Question 2).
About one-quarter (24%) of respondents had more than 10 years of experience and a further
quarter (24%) had 5-10 years of experience. Respondents with 2-5 years of experience formed
the largest subgroup, accounting for about 38% of respondents, and only 14% of respondents had
less than 2 years of specific PMO experience.
Respondents were currently or previously employed as either “clinical scientists in the
biopharmaceutical industry” (52%) or “regulatory scientists in the biopharmaceutical industry”
(50%) (Question 3); all respondents identified with one or both of these two professions.
Percentages exceed 100% because some individuals reported two or more sets of responsibilities.
A small number (6%) of respondents also volunteered that they were currently or previously
employed in technically oriented professions including toxicology, clinical pharmacokinetics,
medical or scientific affairs, or drug safety.
100

Most survey participants (32 /42) were based in the United States at the time of survey, but a
majority identified prior industry experience in other countries/regions as well as the US. The
remaining 10 respondents were based in 6 other countries/regions, including Australia/New
Zealand, Canada, China, the European Union (Belgium and the United Kingdom), Japan, and the
United States (Question 4). However, some respondents self-associated with more than one
region as being a region whether they gained a majority of their regulatory and/or clinical
experience (Figure 1).
Figure 1. Question 4 – Please indicate the region in which you have acquired the majority
of your regulatory/clinical experience? Please check all that apply

In addition, five of the US-based respondents represented other regions at the time of the survey
(2 representing China, 1 representing Japan, 1 representing Latin America, and 1 representing
Emerging Market regions). As a result, a significant proportion of respondents were either
operating in another region at the time of the survey (24%, or 10/42), while others were US-based
101

but managed regulatory activities in other regions (12%, or 5/42). Thus, 36% (15-42) percent of
respondents were managing ex-US regulatory activities at the time they completed the survey
Twenty-seven respondents (64%) indicated that the US was the region where they acquired most
of their regulatory/clinical experience in PMO drug development. In the remaining group, 18
respondents (43%) acquired most of their regulatory/clinical experience in Europe,
14 respondents cited Canada (33%), 12 respondents cited Japan (27%), and 8 (19%) cited that
their experience in PMO drug development was not region specific, meaning their role was global
in nature. A small group was split evenly (6 respondents each, or 14%) between individuals who
obtained their experience in Australia/New Zealand, and “Other geographic region (please
specify)” where respondents cited Latin America, ASEAN, China, and Emerging Markets.
Respondents came from ten different pharmaceutical companies or consulting groups.
Most respondents suggested that they were more experienced with late than earlier phases of drug
development (Question 5). Seventy-six percent of respondents had phase 2 regulatory/clinical
experience and over 90% of respondents had phase 3 regulatory/clinical experience with
investigational PMO therapies (Figure 2).
102

Figure 2. Question 5 – What phases of drug development have you supported for
therapeutic products intended for the treatment/prevention of PMO?
Please check all that apply

Most respondents had experience with more than one therapeutic product for PMO (Question 6)
(Figure 3).
Figure 3. Question 6 – How many different therapeutic products for PMO have you
supported during your regulatory/clinical career?

103

The majority of respondents had interacted directly with drug regulatory authorities (88%),
clinical study scientific advisory boards (70%), or investigators (60%) (Question 7) (Figure 4).
However, only 43% of respondents identified institutional review boards or ethics committees as
bodies with which they have had direct interactions.
Figure 4. Question 7 – In your role, did your job responsibilities require you to interact
directly with any of the following? Please check all that apply

4.2.2 Impact/Implications of Regulatory Dissonance
Responses to questions regarding the effects of regulatory dissonance suggested that such
dissonance complicates and lengthens the development process. Most respondents (74%) agreed
with the view that regulatory dissonance has increased the challenges associated with developing
drug therapies in PMO (Question 8); 74% of the respondents agreed with this statement (“Yes”),
17% of the respondents did not agree (“No”), and 10% had no opinion (Figure 5).

104

Figure 5. Question 8 - In your experience, has regulatory dissonance increased the
challenges associated with developing drug therapies in PMO?

This exploratory study was relatively small in size. Nonetheless, important trends were noted
when cross-tabulations were conducted evaluating the responses of this question against factors
relating to the professional experience of the respondents (Section I, Professional profiles of
individual respondents). The conviction that regulatory dissonance increases the challenges
associated with developing PMO drugs appeared to be more strongly held by clinical scientists
than regulatory scientists in the biopharmaceutical industry (Question 3): 86% (19/22) clinical
scientists answered in the affirmative whereas only 62% (13/21) of the regulatory scientists had
affirmative responses (Appendix II, part i.).
In addition, further cross-tabulation of data showed a similar trend indicating that respondents
with longer drug development experience with PMO agents (Question 2) were proportionally
more likely to agree with the statement that regulatory dissonance across DRAs increased the
challenges associated with developing drugs in PMO. Amongst those who responded in the
affirmative, 50% (3/6) of those with 0 – 2 years of PMO drug development experience, 75% of
those with > 2-5 years of PMO drug development experience, 75% of those with > 5 – 10 years
105

of drug development experience, and 80% of those with > 10 years of PMO drug development
experience agreed with the statement that regulatory dissonance across DRAs increased the
challenges associated with developing drugs in PMO (Appendix II, part ii).
Most (70%) also agreed that regulatory dissonance had resulted in an increase in the timelines for
clinical development (Question 9). Most estimated the delays to be up to 12 months (34%,
n=14) or >12 to 24 months (32%, n=13), but a few respondents identified even longer delays of
>24 to 36 months or >36 to 60 months longer (1 and 1 respondent respectively) (Figure 6). A
minority indicated that regulatory dissonance did not increase clinical development timelines
(15%, or n=6).
Figure 6. Question 9 - In your experience, has regulatory dissonance increased the clinical
development timelines of PMO drugs (i.e., delays to study start, lengthier clinical studies)?
By how much time?

106

A majority of respondents (51%) agreed that regulatory dissonance had increased timelines
related to the reviews of market authorizations by regulatory authorities. When asked to estimate
the time added to standard application reviews (Question 10), most identified that timelines had
increased for up to 24 months. The modal answer was 0 to 6 months (22%; n=9), but 12% (n=5)
selected > 6 to 12 months, 10% (n=4) selected > 12 to 18 months, and 7% (n=3) selected > 18 to
24 months (Figure 7). However, 20% (n=8) of respondents indicated that regulatory dissonance
did not increase the duration of marketing application reviews, and 29% (n=12) could not answer
the question.
Figure 7. Question 10 - In your experience, has regulatory dissonance in drug registration
requirements increased the regulatory authority review times of marketing authorization
applications of PMO drugs? Please estimate time added to standard application reviews?

107

A majority of respondents (61%) agreed that regulatory dissonance increased the required size of
multi-regional, pivotal, phase 3 clinical studies. Only, 24% (n=10) of respondents identified that
extra study subjects were not required. Of the others, 5% (n=2) identified a minimal increase in
study subjects (+10s), 29% (n=12) identified a moderate increase in the number of study subjects
(+100s), and 27% (n=11) identified a significant increase in study subjects (+1000s)
(Question 11). A small number (15%; n=6) could not answer.
A majority of respondents (83%) agreed that regulatory dissonance had increased the complexity
of conducting phase 3 trial designs (Question 12), with answers ranging from an increase in
complexity characterized as minimal (20%, n=8), moderate (51%; n=21) or significant
(12%;(n=5). However, 10% (n=4) of respondents indicated that regulatory dissonance did not
increase trial/program complexity, and 7% (n=3) could not answer. Results relating to the effects
of dissonance on the costs of PMO drugs were more mixed (Question 13). A small majority
(54%) felt that it led to increased costs for patients/payers, characterizing the potential impact to
downstream costs as moderate (37%; n=15) or significant (17%; n=7). In contrast, 29% (n=12)
of respondents felt that regulatory dissonance had no impact on eventual drug costs and 29%
(n=12) could not answer.
A majority of respondents 27 (66%) selected “Yes” when asked whether regulatory dissonance
impacted patient access to important new treatments for PMO (Question 14) whereas 17% (n=7)
selected “No” to this question, and a further 20% (n=20) selected “No opinion” (Figure 8).
Respondents who responded in the affirmative were also given the opportunity to explain their
responses (Table 9).

108

Figure 8. Question 14 – In your experience, does regulatory dissonance impact patient
access to important, new treatments for PMO (i.e., products not
approved or delayed to market)?

109

Table 9. Question 14 – Open Responses from Respondents (In your experience, does
regulatory dissonance impact patient access to important,
new treatments for PMO)?
Question 14 – Open Responses from Respondents
Some HA place newer products as 2
nd
or 3
rd
line products when they have in fact demonstrated safety
and efficacy as first line
Definition of fracture risk and populations on label not necessarily clear or reflective of the populations
studied in Ph 3 trials
Delay, restricted and different labels, leading to differences in reimbursement
Regulatory dissonance may result in limitation of populations for which products will be approved.
However, patient access may be even more impacted by the fourth hurdle (reimbursement)
Delays in development timelines and approvals delay patient access to new drugs
Big problem in UK and other EU countries where regulators and payers have different agendas. Onerous
requirements affect label and therefore reimbursement and access
Delays in clinical development in response to country specific requirements extends the time before
filing takes place and ultimately time to market
Ethnic sensitivity analysis results in conducting local study and delay in approval from the western
counties
In China, for imported products, a phase 1 and a phase 3 study have to be completed prior to NDA/BLA
filing. The application for clinical trials and the conduct of the studies can add up to 4-5 years delay of
product availability
Based on the more limited indication statement approved for [REDACTED] in the US versus other PMO
prescription drugs previously approved by the FDA in the same clinical setting
Patient access is delayed
Delays to market
I believe it causes candidate abandonment, or even discontinuance of PMO programs
Revised indications versus predecessors make access more challenging
Divergent regional guidance leads to challenges in a global development program, therefore potential
delays to approval where global program is not "tailored" to each regional requirement
If the agency accepts the foreign data, we would market the drug much earlier
Delays in approval; indications that confuse some prescribers
Delay and restricted access due to labeling
If drugs are not registered in some regions, the patients cannot avail themselves of newer therapies
Takes longer to access and some therapies may not be taken forward due to complexity
Additional regulatory requirements can cause delays in study completion or additional studies to be
required
Many emerging market countries want special studies or have substantially longer regulatory review
timelines than in US, EU and Japan
Prolonged time to market
Delayed availability for patients in need

110

About half of the respondents (51%) felt that their company considers regulatory dissonance to be
an important factor when assessing the commercial POS or the net present value (NPV) of a
potential new PMO agent (Question 15). About one quarter (27%; n=11) held the view that it
was not an important factor, and a further 9 (22%) had no opinion (Figure 9). Respondents who
responded in the affirmative were also given the opportunity to explain their responses (Table
10).
Figure 9. Question 15 – In your experience, does your company consider regulatory
dissonance to be an important factor when assessing the commercial probability of success
or the net present value (NPV) of a new potential PMO agent?

111

Table 10: Question 15 - Open Responses from Respondents (In your experience, does your
company consider regulatory dissonance to be an important factor when assessing the
commercial probability of success or the net present value (NPV)
of a potential new PMO agent?)
Question 15 - Open Responses from Respondents
Extended timelines, more complex and larger trials all contribute to the NPV
Regulatory uncertainty in a key market will affect NPV
Regulatory dissonance is considered when assessing target population in various regions
Dissonance specific differences in label impact target populations potentially
The effects of country pricing in Europe in particular can make a product commercially unviable
Differing likelihood of regulatory success impacts income projections
In the case of [REDACTED], Regulatory Dissonance in terms of having to harmonize the global Phase 3
clinical study duration to accommodate the primary endpoint assessment at 3 years (as required by a
subset of regulatory authorities such as the FDA) clearly negatively impacted the NPV of this particular
agent
Impact on labeling and promotion
Assessments affected by regional differences
Regulatory probabilities and safety language affect commercial success and are incorporated into the
local forecast
Because it affects not only the timing of getting a regulatory approval but also the phase 3 program
(hence the cost of the drug development)
Include in assessment of phase 3 probability
Criteria for evaluating probability of technical and regulatory success takes into account divergent advice
from different agencies
The NPV depends on cost of development and delays due to RD between the similar approach taken in
the US and EU from that elsewhere in the world. Emerging Market countries with less scientifically
sophisticated Regulatory Agencies are the source of the Dissonance, not the FDA or EMA

112

Similarly, about half (53%, n=21) of respondents felt that regulatory dissonance decreased
industry innovation in drug research and development (Question 16). Only 5% (n=2) felt that it
increased industry innovation. A substantial minority (17, 43%) held the opinion that Regulatory
dissonance has not had an effect on industry innovation (Figure 10).
Figure 10. Question 16 - In your opinion, how does regulatory dissonance impact industry
innovation in drug research and development (as it relates to PMO)?

Finally, respondents were asked whether a certain amount of regulatory dissonance in
regional/requirements may be beneficial to sponsors of investigational agents (Question 17).
Twenty-two of the 41 respondents (54%) responded “No” and 4 had “No opinion”. Those
responding “Yes” (15, or 37%) were encouraged to explain their responses (Table 11).

113

Table 11. Question 17 - Open Responses from Respondents (Question 17 – Some have
suggested that a certain amount or regulatory dissonance in regional guidelines /
requirements for developing drug therapies may actually be beneficial to sponsors of
investigational agents. Do you agree?)
Question 17 - Open Responses from Respondents
Adaptation to specific populations becomes more lax in practice and hallo effects affecting
different regions
Provides a forum for innovation and focus on what matters most to the global community
It may challenge sponsors to develop better products, although it will drive up to development
costs and thereby limit the number of products a company can take forward in a given time
window
Increased complexity ends up answering more clinical questions
For Asia Pacific region, I do not agree the above statement. Major countries in the Asia Pacific
region, namely China, S. Korea, and Taiwan require the product to be approved in the source
country prior to their country's approval. Therefore, Regulatory Dissonance in these countries
does not form into any significant influences to the major regulatory bodies, i.e. US and EU. In
addition, these countries usually require local patient data for product approval which adds to
the cost of product development
Different agencies have different focus areas/areas of concern. Dissonance is a reflection of
that. It may also reflect different levels of expertise/deep insights into the underlying science.
Well, sometimes also the lack thereof
May allow more flexibility for approving drugs in different regions
Stimulates sponsor to consider broadest possible global application
By addressing unforeseen issues with protocols up front
Can encourage broader program
Information acquired due to special studies required by a country allows more targeted
promotion in that country, once the results are available
Each developed drug therapy is reviewed by the regulatory authority as a unique marketing
application; provides robustness in decision making of risk profile and benefit profile

4.2.3 General Causes/Factors Contributing to Regulatory Dissonance
Respondents were queried about their views regarding the leading causes of regulatory
dissonance for PMO agents (Question 18). The six leading causes of regulatory dissonance were
identified as regional regulatory authority advice (32 respondents, 78%), regional regulatory
authority guidelines (21 respondents, 51%), regulatory benefit:risk assessments (18 respondents,
44%), drug approval precedence (15 respondents, 37%), differing medical standards of care
(14 respondents , 34%), and Health Technology Assessments (HTA) (13 respondents (32%)
114

(Figure 11). Respondents were given the opportunity to provide more detailed text responses
(Table 12).
Figure 11. Question 18 - In your experience, which of the following lead to the most
regulatory dissonance in the setting of PMO? Please check the items that best apply

Table 12. Question 18 - Open Responses from Respondents (In your experience, which of
the following lead to the most regulatory dissonance in the setting of PMO?
Please check the items that best apply)
Question 18 - Open Responses from Respondents
Inconsistency within FDA
Changing ideas on what represents "safety" in a drug; ethical concerns raised in journals

115

4.2.4 Alignment of Regional Indication-specific Postmenopausal Osteoporosis Guidelines
Some survey questions explored in more detail the types of dissonance that were found to be
present in guidance regarding the development of drugs for PMO under the categories of 1) type
of controls, 2) duration of treatment, 3) duration of long-term follow-up, 4) primary and
secondary endpoints, 5) the timing of the primary endpoint, 6) the target population and
7) intrinsic factors related to the investigational agent (e.g., safety profile, mechanism of action,
novelty of product).
Control arm: Most respondents (83%) reported some level of regulatory dissonance in regulatory
advice or feedback regarding the type of control arm to be employed in pivotal, phase 3 PMO
clinical studies (Question 19). Of these respondents, 59% (n=24) felt that regulatory dissonance
existed, but was manageable, whereas 24% (n=10) acknowledged that the level of regulatory
dissonance was significant. Only 15% (n=6) expressed the view that no regulatory dissonance
existed; a single individual did not answer the question.
Duration of treatment: A majority of respondents (80%) reported some level of regulatory
dissonance with regard to regulatory advice/feedback on the required length of treatment required
for pivotal, phase 3 clinical studies (Question 20). Only 12% (n=5) of respondents expressed the
view that regulatory requirements/advice were aligned. Most (54%, n=22) viewed regulatory
dissonance as existing but manageable, and 27% (n=11) viewed the level of dissonance as
significant. A small number (7%, n=3) of respondents could not answer.
Duration of long-term safety follow-up: A majority of all respondents (80%) reported some level
of regulatory dissonance related to regulatory advice/feedback on the duration of safety follow-up
required following the assessment of the primary endpoint in pivotal, phase 3 clinical studies
(Question 21). A few (10%, n=4) did not identify that advice was dissonant. About half (53%;
116

n=21) felt that dissonance was present but manageable, and a smaller proportion (28%; n=11) felt
that the dissonance was significant. Ten percent (n=4) of respondents could not answer the
question.
Identification of endpoints and timing of assessments: A majority of respondents (70%) reported
some level of regulatory dissonance related to the regulatory advice/feedback with respect to the
required primary efficacy endpoint in pivotal, phase 3 clinical studies (Question 22). Of these,
61% (n=25) felt that the regulatory dissonance was present but manageable, and 10% (n=4)
regarded it as significant. Of the remaining 30%, 5% (n=2) of respondents could not answer, and
the remaining 24% (n=10) thought it present but manageable. Similar numbers were observed
when respondents were asked about the level of regulatory dissonance experienced with regard to
the regulatory advice/feedback on required secondary endpoints in pivotal PMO studies
(Question 23). In all, 59% (n=24) felt that the regulatory dissonance was present but
manageable. Only 10% (n=4) recognized a significant level of regulatory dissonance. A small
number answer (7%; n=3) could not answer, and the remaining respondents (24%; n=10) did not
report dissonance on this element.
Most (78%) also reported some level of regulatory dissonance with respect to recommendations
regarding the timing of the assessment of the primary endpoint (Question 24). Only 12% (n=5)
identified that no dissonance existed. In contrast, 54% (n=22) considered the regulatory
dissonance to be present but manageable and 24% (n=10) saw the dissonance as significant. A
small number (10%; n=4) again could not answer.
Target populations: Of all of the elements in this cluster of questions, the most common source
of dissonance was viewed to be the selection of the target population for the pivotal, phase 3
clinical studies (Question 25). Only 10% (n=4) did not feel that regulatory dissonance was
117

present. Of the 85% who identified the presence of dissonance, 61% (n=25) viewed it as
manageable and 24% (n=10) viewed it as significant. The remaining 5% (n=2) could not answer.
Intrinsic factors: The majority of respondents (81%) reported regulatory dissonance related to
factors intrinsic to the investigational agent under study (e.g., safety profile, mechanism of action,
novelty of product) (Question 26). A majority (59%; n=24) felt that intrinsic factors relating to
the product (i.e., safety profile) act to increase regulatory dissonance, and a nearly a quarter more
(22%; n=9) characterized this increase as significant (Figure 12).
Figure 12. Question 26 -In your view, do intrinsic factors relating to the investigational
agent under study (e.g., safety profile, mechanism of action, novelty of product)
impact the level of regulatory dissonance?

118

When asked, in the respondent’s experience, which of the areas identified above (Questions 19-
26) resulted in the greatest degree of regulatory dissonance when designing pivotal, phase 3
studies in PMO (Question 27), four areas received 89% of respondent selections. These four
included the type of control group (32%, n=13), the target population (27%; n=11), timing of the
primary endpoint assessment (15%, n=6 ), and intrinsic factors (15%, n=6 ) (Figure 13).
Figure 13. Question 27 - Amongst the areas identified above, which in your view results in
the greatest degree of regulatory dissonance when designing pivotal,
phase 3 studies in PMO?

119

4.2.5 Dissonance Relating to Drug Review and Approval Procedures
The survey explored the extent to which dissonance was seen to be present in regulatory
procedures and interactions, including 1) the assessment of benefit:risk, 2) feedback from
regulatory authorities and 3) allowable labeling claims. In all of these aspects, all respondents
recognized some degree of dissonance. Between 7 and 10% (3-4 respondents) could not answer
these questions.
Benefit:risk assessments: About half of the respondents (52%; n=21) felt that some regulatory
dissonance was present but manageable, and (37%;n=15) characterized the level as significant
(Question 28).
Regulatory Authority Feedback: More than half (56%; n=23) characterized the dissonance as
present but manageable, and about a third (34%;n=14) characterized it as significant
(Question 29).
Allowable indication statements: About half of respondents (52%; n=21) characterized the
regulatory dissonance on this aspect of regulatory approvals as present but manageable
(Question 30). The remaining 42% (n=17) characterized the degree of dissonance as significant.
Other sources of regulatory dissonance: Respondents were asked whether other potential sources
of dissonance arise during the development of PMO agents that should be considered by this
study (Question 31) (Table 13).
120

Table 13. Question 31 – Open Responses from Respondents (In your view, are there other
sources of regulatory dissonance that arise during the development of PMO agents that
should be considered by this study)
Question 31 – Open Responses from Respondents
Most regulatory dissonance is inherent in the interpretation of the managing officers from the different
regulatory agencies. Though the guidances are rather clear the “stamp” of the respective HA official is the
source of dissonance
Divergent directions for pediatric development. Divergent directions for additional indications
Understanding of antibody product and small molecule may not be consistent among the regulatory
agencies
I think there is a fundamental lack of knowledge regarding bone and osteoporosis by the regulatory
authorities who are currently at FDA. When I started dealing with the agency there were 3 or 4 well-
informed regulatory scientists in place who understood bone drug mechanisms, the limits of BMD, the
potential for micro-CT, finite element analysis and other sophisticated approaches to assessing bone
quality. Unfortunately, by virtue of the fact that most of the best bone people at FDA have since left the
agency, and the assessment of bone is now housed in Reproductive Medicine, industry faces a severe
problem of not having competent scientists to deal with, and the result is mixed messages, reneging on
advice given by previous FDA staff, and interference and micromanagement of study design in a most
incompetent manner
Internal dissention among regulatory agency staff scientists –Strong differences of opinion among various
national regulatory agencies –Legal and publicity concerns of regulatory agency staff –Conflicting and
even uninformed or mistaken ideas of outside members of FDA Advisory Committees –Limitations of the
science of drug development and evaluation
Regulatory boards requests are not consistent when people turnover
No global consensus on study design, study endpoints, control arms, use of novel endpoints measuring
bone quality, use of surrogate markers
The impact of the role of regional governments policy on health care (reimbursement) on the regulatory
guidance for development of PMO drugs in that country/region
Dissonance on inclusion/exclusion criteria, stopping rules for safety, pregnancy
Minor requests for changes in the details of the protocol (inclusion criteria etc) from various reg agencies
results in increased costs and delays due to amendments
Requirements for phase 1 studies. Requirements for sub-studies or additional safety assessments in
phase 3
Regulations in the EU and US require studies in pediatric populations, even when the only [adult]
indications are for postmenopausal osteoporosis and related bone diseases (male osteoporosis). Some
bone-active drugs are not appropriate for use in growing children. The US FDA takes a reasonable
approach and generally grants a waiver from the pediatric study requirement when the drug is not
appropriate for children. However the EU’s PDCO group (reviews pediatric studies requirements) may
request studies of drug in children that at a minimum might be considered unethical (because they would
never be used in children) or might even pose a small risk to the subject. And one cannot obtain approval
of a drug intended for use only in adults if PDCO has not granted a waiver
(Continued next page)

121

Question 31 - Open Responses from Respondents (continued)
Key sources of regulatory dissonance that impact on the regulator's review time line for PMO, or the
perception of the new therapeutic via publicly available information (impact on perception of risk profile
more broadly), can be experienced when: 1. The PMO therapy is a monoclonal antibody acting via a new
pharmacological pathway, and proposed for prescription by specialists AND general practitioners; the
paradigm shift is significant to the point of the local regulator seeking more evidence for an acceptable risk
profile and benefit profile (See next two points). 2. Additional indications are under investigation and the
regulator is of the opinion there is more data available to assist in decision making. While the local
regulatory authority may not have the mechanism to accept new data (due to statutory constraints) at the
time the current marketing application is under review, the regulatory authority may apply mechanisms to
stall time to making a decision, awaiting a decision from other reference regulatory authorities. 3. There is
concurrent submission with the proposed PMO indication, datasets and proposed indications for other bone
loss conditions comprising smaller studies or as perceived by the regulatory authority softer endpoints, and
then infers the safety profile in these other bone loss conditions has having a bearing on the interpretation
of risk in the PMO setting. The regulator can use this situation to influence the advisory committee and
their recommendations which are reflected in the regulatory information that is an output of PMO review
(for example the publicly available product assessment report)

4.2.6 Potential Policy Solutions to Reduce Regulatory Dissonance
Respondents were asked to identify potential mechanisms which would be most beneficial in
reducing Regulatory Dissonance, and were invited to select a single preference (“please check
one item”) (Question 32). Two of the potential mechanisms receiving the highest number of
respondent recommendations (56%; n=23) were identified as “Harmonized Regulatory Authority
Guidelines” (29%; n=12) and “Joint Scientific Advice Meetings” (27%; n=11). “Harmonized
Procedures for Assessing Benefit:Risk” received 15% (n=6) of respondent recommendations, and
“Mutual Recognition Agreements” (MRA) received 12% (n=5) of respondent recommendations
(Figure 14). Respondents were also given the opportunity to volunteer additional options (Table
14).
122

Figure 14. Question 32 - In your view, which of the following potential mechanisms would
be most beneficial in reducing regulatory dissonance? Please check one item.

Table 14. Question 32 - Open Responses from Respondents (In your view, which of the
following potential mechanisms would be most beneficial in reducing
regulatory dissonance?)
Question 32 - Open Responses from Respondents
Lobbying FDA to bolster its bone assessment staff with real bone scientists
MRAs would be most beneficial but I doubt those will happen for the big players

Respondents were asked to identify from a list of organizations the one that should take the lead
in implementing mechanisms to reduce Regulatory Dissonance, and were invited to select a
single preference (Question 33). Two main types of organizations - Regulatory Agencies (i.e.,
EMA, FDA, Health Canada, NIH, PMDA, etc.) (22 or 55%) and ICH (International Conference
on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use
123

(14 or 35%) - received 90% of respondent recommendations (Figure 15). Respondents were also
given the opportunity to volunteer additional options (Table 15).
Figure 15. Question 33 - In your view, which of the organizations listed below should take
the lead in implementing mechanisms to reduce regulatory dissonance?
Please check one item.

Table 15. Question 33 - Open Responses from Respondents (In your view, which of the
organizations listed below should take the lead in implementing mechanisms to reduce
regulatory dissonance? Please check one item)
Question 33 - Open Responses from Respondents
Bone societies, such as ASBMR
Now that the NIH is mentioned, the differences between the FDA requirements and the
Clinicaltrials.gov requirements are an administrative handicap. The NIH regulates reporting through
clinicaltrials.gov and those requirements may diverge from the standard FDA (and ICH) approach to
reporting. This just adds to cost, and doesn't really slow down development. Due to the requirements of
clinicaltrials.gov, some of what is posted on clinicaltrials.gov is not a full report of the study that the
FDA gets or what is in a published manuscript in a medical journal
In Question 34, the respondents were asked, given their drug development experience, whether
the mechanism they selected above to reduce regulatory dissonance in PMO would be helpful if
124

applied in other disease areas. Eighty-three percent (n=34) of the respondents agreed (Yes), 2%
(n=1) of the respondents did not agree (No), and 15% (n=6) had no opinion (Figure 16).
Figure 16. Question 34 - Given your drug development experience, do you think that the
mechanism you selected above to reduce regulatory dissonance in PMO would be helpful if
applied in other disease areas?

In Question 35, survey participants were thanked for their responses and time, and were invited
to elaborate on any of their responses in an open comment box (Table 16).
125

Table 16. Question 35 - Open Responses from Respondents (This is the end of the survey.
Please feel free to elaborate on any of your responses in this survey. Please also indicated
whether you would be open to discussing your feedback for the
purpose of informing this study)
Question 35 - Open Responses from Respondents
Regulatory dissonance is mostly due to the human factor and driving agreement among regulatory may
help align the differences. Open to discussing feedback
Open to discuss
Open
Yes, open to discussing feedback
Yes
I am happy to further discuss
I am happy to speak with you regarding my feedback. Thank you for the opportunity to participate in this
study
Yes
Certainly, I'd be pleased to do so
Yes
Suggest asking respondents about the level of their experience with different regulatory authorities and
weighting the responses to this survey accordingly
Standards seem to be changing constantly and the level of flexibility for novel therapies differs between
agencies and even within the same agency. We need more consistency in reviews and the type of
information acceptable for approval
Happy to discuss feedback
Thank you for giving me a chance to participate in this interesting survey. I am open to discussing any
opinions or comments for your study
Comprehensive survey - you captured it
Yes
Yes, I would be open to discussing my feedback for the purpose of informing this study
Yes
I would be happy to discuss
Would be willing to discuss feedback
Yes
The major dissonance is in between the regulatory guidelines vs what you need to do to develop a
marketable product and the ability to understand that new therapies with fundamentally different MOA
may need to be tested differently, because they may need to be use it different, (shorter duration,
sequential combo, etc)
4.2.7 Cross-Tabulations
Of interest to the researcher was whether factors relating to the professional experiences of
respondents might have affected their responses. These are explored throughout this chapter, and
the data from trends of interest are presented in Appendix II.
126

CHAPTER 5: DISCUSSION
5.1 Regulatory Dissonance Between DRAs
This exploratory study introduces regulatory dissonance, a concept which is intuitively
recognized amongst experienced industry regulatory professionals and clinicians and is readily
understood to be an important driver of the current, decentralized drug development process.
Despite the contribution of regulatory dissonance to the complexity of drug development,
relatively little rigorous study has been directed at the specific impact of such dissonance on
different aspects of drug development, particularly at the level of specific disease indications and
through the eyes of the drug developers themselves. Regional DRAs have been slow to recognize
and mitigate the important effects of regulatory dissonance on drug developers, perhaps because
enhanced harmonization of indication-specific requirements would minimize sovereign decision-
making authority in important areas relative to medical standard of care, or require a prohibitive
commitment of resources and time to address. Additionally, regulatory dissonance is not an
attribute that lends itself well to performance metrics because it has not been well-defined. Its
causes are multi-factorial and access to documentation that might illuminate potential areas of
dissonance as a result of DRA review is limited, particularly in situations in which regulatory
agencies have made negative decisions.
5.2 Methodological Considerations
Several limitations were anticipated in this study. Some were related to the challenges associated
with any social science project that uses survey methodology for data collection. For example,
quality surveys rely on the candor of the respondents, and this is often problematic if the survey
deals with topics that involve confidential materials or activities. Based on methodological
considerations discussed by Evans and Mathur (2005), an online, self-administered survey
127

method was selected on the premise that direct interactions with the respondents should be
reduced to encourage greater candor amongst participants (and reduce potential bias as a
consequence of experimenter expectancy). Nonetheless, Evans and Mathur also point out that an
online survey method may cause respondents to be concerned that their privacy may not be
protected or that the data may not be treated confidentially (Evans and Mathur, 2005). Thus,
special care was taken to (1) assure participants of the anonymity of their responses, and (2)
phrase questions in a manner that did not require respondents to speak directly about company-
specific activities. Even with these precautions, the investigator was concerned that respondents
might hesitate to share opinions different to those of their employers or respective DRAs, or
might feel constrained by the confidential and sensitive nature of their activities. The contrary
appears to have been true. Responses, particularly in the open comment fields, were surprisingly
candid. Furthermore, a few respondents sought out the investigator to discuss the survey by
telephone or email to share their experiences related to regulatory dissonance.
An online survey may be limited in its external validity if the sample of interest does not
represent a full range of individuals operating across geographic regions, in different regulatory
environments, and with divergent experience levels. An initial concern with this research was the
potential for geographic delimitation of the respondents primarily to the US, and the potential
bias introduced by an overreliance on respondents from a small number of US-based
biopharmaceutical companies. Thus, Section I of the survey (Appendix I) sought information on
the professional profiles of the respondents, to capture respondent characteristics such as
geographic diversity. Intensive efforts were made to recruit the most globally diverse group of
participants possible. This was achieved by requesting participation from (1) representatives
from company business partners in other regions, (2) company representatives who were
physically located in other regions, and (3) company representatives representing other regions
128

from their US-based offices. As a result, a significant proportion of respondents were either
operating in another region (24%), or were based in the US but managing regulatory activities in
other regions (12%). Thus, more than a third of respondents were managing ex-US regulatory
activities at the time that they completed the survey.
Another limitation anticipated under Chapter 1.5 was the fact that the survey pool was to be
composed of PMO regulatory and clinical scientists, and this grouping would be small,
specialized and difficult to recruit. Further, many participants would not be known to the
investigator at study outset. An effective solution to boost sample size in a specialized, hard-to-
reach group is to add additional participants through “referral” or “snowball” sampling
(Handcock & Gile, 2011). Snowball sampling is a non-random convenience sampling method
well-known to be vulnerable to bias because like-minded respondents with professional or
personal interrelationships might be recruited preferentially, making it difficult to generalize the
results to different populations (Faugier & Sargeant, 1997). Although only four study
participants were recruited in this manner to complete study enrollment, it is still acknowledged
that generalizations based on the results from a small, qualitative study may be problematic.
Even before the study commenced, the investigator was aware that the lessons learned from a
rather narrow case study of PMO might not generalize well to other indications/disease states.
Nonetheless, case studies may help us to understand the broad dimensions of regulatory
dissonance. Because so few studies have been conducted previously, these results may represent
the most up-to-date regulatory intelligence available. The simple process of organizing this kind
information in one place for comparative analysis can yield valuable insights that can be
extrapolated to other therapeutic areas. Not every lesson in science comes from highly structured
randomized trials (Gilgun, 2011). Nevertheless, the results here will need to be matched by
comparable research in other therapeutic areas to assess the degree to which the findings can be
129

generalized to other indications. Some research in other therapeutic areas is beginning to emerge.
Recently published work by Breckenridge (2011), for example, describes specific cases of
regulatory divergence. An analysis of rosiglitazone (Avandia) from clinical and observational
trials, and metaanalyses, demonstrated an increased risk of cardiac ischemic events following the
administration of rosiglitazone compared to other antidiabetic drugs (Breckenridge, 2011).
Breckenridge notes that this elevated risk was treated differently across regions. In 2010, a
benefit:risk assessment led by the FDA resulted in rosiglitazone remaining on the market, albeit
with additional safeguards in the form of a mandated Risk Evaluation and Minimization Strategy
(REMS). However, on the basis of a similar assessment during the same time period, the EMA
withdrew its marketing authorization. Thus it appears that the regulatory dissonance observed
with PMO agents might also be a factor of concern, albeit to a lesser or greater extent, in other
indications. The survey tool that has been developed here could be readily adapted to test
whether results of this case study can be generalized to other populations, and additional studies
are planned.
A final concern throughout the survey deployment period was the possibility that
contemporaneous changes in the regulatory environment might have an impact on study results.
Views of respondents may change, for example, when new regulations or guidance is introduced.
Fortunately, regulatory authorities did not significantly revise their guidance materials during the
4-month survey deployment period (12 November 2012 – 13 March 2013), and no other major
change was detected in the regulatory landscape (i.e., product approvals or withdrawals). On
19 October 2012, the EMA did release a notification that it was adopting a “Concept paper on the
need for revision of the guideline on the evaluation of medicinal products in the treatment of
primary osteoporosis” (EMA, 2012d). Initial concerns that this change might impact the survey
were allayed after inspection of the guideline showed it to have a narrow scope that focused not
130

on PMO in general, but rather on issues related to secondary osteoporosis, such as osteoporosis
caused by other factors such as glucocorticoid therapy (GIOP). Of course, the PMO regulatory
landscape will change over time, but the results collected here can serve as a benchmark against
which later surveys or other data-gathering exercises can be compared.
5.3 Potential Challenges of Regulatory Dissonance
The results of this study showed that regulatory dissonance is a problem familiar to regulatory
affairs professionals and clinical scientists, who are the industry representatives most likely to
interact directly with DRAs. Most survey participants, irrespective of length of experience or
geographic sphere of operation, concurred that regulatory dissonance across DRAs increased the
challenges associated with the development of drugs in PMO. This general finding may not come
as a surprise to industry representatives; however it may not be an aspect of regulatory science
that is fully appreciated by DRAs.
Impact of regulatory dissonance on clinical trial design and conduct
A majority of respondents agreed that regulatory dissonance had an unfavorable impact in all of
the domains that were evaluated with respect to clinical trial conduct (Survey Section II,
Impact/Implications of Regulatory Dissonance). However, its largest perceived effects appeared
to be an increase in the complexity of clinical trial design and conduct. To a somewhat lesser
degree, respondents also felt that regulatory dissonance had a detrimental impact on clinical
development timelines, patient access to newer treatments and the size of pivotal clinical trials.
The linkage between national regulatory divergence and increasing trial complexity, expanded
clinical development timelines, and delayed patient access to new treatments are not new
concerns, and have been highlighted in a recent report by the Organization for Economic
Cooperation and Development (OECD), a multi-national forum where countries cooperate on
131

policy and compare policy experiences. The OECD issued report, the Recommendation on the
Governance of Clinical Trials, calls for “improved consistency among national regulations and
their interpretations, and on streamlined procedures for the oversight and management of clinical
trials” (OECD, 2013). The increasing complexity of trial conduct in part can be attributed to an
increased reliance on multiregional trials, which are in part “more complex to perform than
national ones due in particular to the difficulties arising from the diversity of legal frameworks”
(OECD, 2013). The report of the OECD noted that national regulatory complexity is causing
clinical studies intended to address important public health problems to be cancelled or to be “so
delayed that their impact is dramatically reduced” (OECD, 2013).
A majority of respondents identified issues of regulatory dissonance related to DRA
recommendations concerning the use of placebo versus active controls, and the required duration
of pivotal studies. This is not surprising given the attention to these topics in the existing
literature, summarized in Chapter 2.4. However, respondents pointed to other aspects of clinical
guidance that differed between DRAs and is not so well-characterized in the literature. These
aspects included divergent criteria for the target population to be studied, the length of long-term
safety follow-up following the assessment of the primary endpoint, the timing of primary
endpoint assessments, and the type of required primary and secondary endpoints to be studied.
The fact that respondents viewed several of the domains of clinical trial design as the area in
which dissonance was greatest might account for a related observation that clinical scientists
appeared to have stronger convictions than regulatory scientists regarding the challenges posed by
regulatory dissonance on the development of PMO drugs (86% vs. 62%). The sample size for
this study was small, so the difference must be considered cautiously, but this may merit further
study.
132

Regulatory dissonance of key aspects of clinical trial design, those critical to eventual registration
of a new medicine, have been reported in at least one therapeutic area. Of note, Stella Augustus,
in her 2011 article “Drug Development in Oncology: A Regulatory Perspective,” identifies key
aspects of clinical trial design, conduct, and analysis where greater harmonization of regulatory
expectations may reduce uncertainties in demonstrating “proof of efficacy” (Augustus, 2011).
These areas, identified as critical to product registration in oncology, include:
(1) Appropriate end points in pivotal trials in advanced cancer-progression-free survival
(PFS) versus overall survival (OS), as the primary measure of efficacy;
(2) Acceptance of either the investigators’ assessment or the Independent Radiological
Facility (RF) determination of disease-related end points (progression or response) as
the primary data set for evaluating efficacy;
(3) The limits of concordance; discordance between the investigator-assessed data and IRF-
reviewed data as a measure of robustness or reliability of the investigator-assessed data;
(4) The differing expectation from study design and statistical analysis points of view in
different International Conference on Harmonization regions; and
(5) Acceptability of background or comparator regimens, albeit clinical established, but
frequently not formally registered, for demonstration of comparable or better efficacy of
new anticancer agents. (Augustus, 2011)
Several of these areas have similar coordinates to Section IV of this survey on osteoporosis
(Appendix I). This suggests that a central finding of this study, the fact that national guidelines
outlining the requirements for establishing efficacy are largely misaligned, may potentially hold
over other therapeutic areas.
A majority of respondents also felt that intrinsic factors related to the safety profile or underlying
mechanism of action of the investigational agent under study had an effect on regulatory
133

dissonance, particularly where the benefit:risk relationship is not clearly inclined toward the side
of benefit. This point is supported by the regulatory dissonance in PMO drug approval
precedents described in Chapter 2.3. Benefit:risk determinations are based on assessments of
multidimensional features, which may be weighted differently. Zafiropoulos summarized
findings of a project to evaluate current benefit:risk practice at EU national competent authorities
by stating that benefit:risk determinations are “done mainly intuitively based on expert judgment,
without the use of a systematic approach” (Zafiropoulos, 2012). In his study, that was based on
interviews of assessors at five EU national competent authorities, he observed significant
divergence in the way that the interviewed assessors defined the terms “benefit” and “risk”. The
definition of risk in particular was found to be prone to widely different interpretations by
assessors (EMA, 2011), including:
Absence of benefit; dangers/hazards for the patient, adverse events, direct or indirect harm to the
patient, frequency and severity of a side effect; harm to non-patients and to the general public;
unacceptable damage to the patient; what is lost compared to current therapy; the negative
aspects of a drug; the inverse of safety; pharmacokinetic interactions; insufficient duration;
probability of an adverse event or harm; negative impact on quality of life; failure to meet
endpoints; intolerability; uncertainty surrounding the risks; mortality; ‘a concept of gambling
which includes perception’. (EMA, 2011)
Interestingly, divergence was identified to go beyond the dissonant perspectives of the individual
assessors regarding the data with which they were presented. Even inherent organizational
differences at national authorities were found to influence the outcome of benefit:risk decision
making. Much effort is currently being expended internationally to produce more standardized
and quantitative benefit-risk tools (Liberti, 2010), however these efforts do not address what may
be the heart of the matter. Recommendations of Zafiropoulos stopped short of calling for a
common, international regulatory process for scientific advice and product review, a
recommendation suggested by the results of this survey of regulatory dissonance.
134

Respondents of this survey also identified, often in strong terms, specific areas of clinical trial
dissonance in regulatory recommendations, in areas such as feedback on novel endpoints for
measuring bone density or quality, the use of surrogate markers, acceptable inclusion/exclusion
criteria, and the use of stopping rules for safety. Further, some respondents expressed concern
that some DRAs lacked state-of-the-art scientific knowledge related to available techniques for
endpoint measurement. Thus the list of areas where challenges were observed included, for
example, knowledge related to newer mechanisms of action or more sophisticated approaches to
assessing bone density or quality (e.g., microtomography, finite element analysis, quantitative
computed tomography). These technical challenges may also act to delay decision making as
DRAs seek additional information from the company under scrutiny or require additional time to
reach internal consensus. This type of delay has been recognized in the literature, especially as it
pertains to innovative therapies, since innovative products “are often more difficult to review,
given the lack of [reviewer] experience with a new class of compounds, thereby contributing
to…regulatory-uncertainty” (Milne & Kaitin, 2012).
It seems clear from this work that dissonance is felt by industry professionals to have a strong
impact on clinical development timelines. This view is not surprising given the other research on
clinical trial conduct. For example, Pammolli and his colleagues concluded that “the average
time of development has increased from 9.7 years for products launched during the 1990s to 13.9
years for products launched from 2000 onwards,” an observation that they attributed to industry’s
drive to develop novel therapeutic targets that will command higher reimbursement premiums,
but which are characterized by high uncertainty and difficulties in development (Pammolli,
2011). The results of this thesis conclude that a significant portion of these difficulties in
development may result from the need by industry to address divergent assessments of the same
data package or study proposal contemporaneously.
135

The predominance of concerns regarding the effects of regulatory dissonance on pivotal clinical
study design, and on timelines and costs, would seem to define an area in which harmonization
initiatives by DRAs might prove to be particularly valuable. This harmonization would need to
consider not only written policies defining general considerations of study design but also the
phase-specific advice given to individual companies at specific points in development; more than
three-quarters of respondents felt that differences in DRA advice or recommendations contributed
moderately or significantly to the dissonance. This point is echoed by Breckenridge who adds
that although regulatory advice by different DRAs is based on the same results of clinical trials,
regulatory divergence often results when the same “evidence conflicts with [often divergent]
public health policy” (Breckenridge, 2011). Thus, it is clear that harmonization by way of static
national guidelines and or standardized benefit:risk modeling or tools may not be sufficient to
reduce regulatory dissonance. The actual apparatus for making risk-based decisions must be
shared cooperatively between regions, or alternatively, decisions made by one region must be
recognized by another.
Other areas of dissonance
A central observation in this work was the multi-factorial nature of regulatory dissonance. This
observation has been cited by others, which has led some to conclude that this may hinder
attempts to minimize it (Breckenridge, 2011). In addition to the concerns associated with clinical
trial design were other areas of concern noted by the respondents including 1) processes for
obtaining reviews and decisions, 2) labeling requirements, 3) issues related to reimbursement by
insurance carriers/governments and 4) regional specific differences in newly emerging DRAs or
economies.
136

1. Processes: Some respondents identified concerns with the way in which
information/opinion gathering occurred during product assessments by regulatory
officials. For example, respondents were critical of the way that DRAs rely on outside
scientific advisors, such as those associated with advisory committees, or on reviewers
who may be uninformed on certain scientific issues or may harbor bias with respect to
certain approaches, products or companies. This shortcoming might be a source of
dissonance in regulatory process that is difficult to change. Resources at most regulatory
agencies are limited and some DRAs must rely heavily on support from a network of
experts. For instance, the EMA draws upon “a list of over 3500 experts from different
European countries” to support the scientific work at that DRA (Ormarsdottir, 2008).
Respondents also indicated that some review divisions even within a single DRA may
offer divergent interpretation and direction. This observation has been corroborated in
the literature by Milne and Kaitin, who have described how differing and inconsistent
applications of policy between the FDA’s own reviewing divisions can result in internal
regulatory dissonance within a single DRA (Milne & Kaitin, 2012). Others noted that
further disruption may occur as the composition of a review team shifts over time with
the loss or transfer of influential personnel.
Some expressed concerns relating to the fact that the pace and predictability of regulatory
decision making was uneven across DRAs. One example of this delay was given by a
respondent who described how marketing approval for a monoclonal antibody acting
through a novel pharmacological pathway was delayed when a single local regulator
sought greater evidence and negotiation to assist in decision making. The review
decision in this case was delayed until a decision from another reference competent
authority was rendered. The inconsistencies with the processes of decision making are
137

currently a subject of concern beyond the confines of the PMO community, as discussed
in Chapter 2.6 and earlier in this chapter, so it is not unexpected that respondents see a
need for improvement in this specific domain.
2-3) Labeling and reimbursement: Particularly interesting were comments related to
1) onerous restrictions applied to some regional indication statements/labels (i.e., either
not reflecting clinical trial results or the population studied) and 2) decisions that limited
access and reimbursement, presumably based on largely the same dataset as that driving
regulatory decisions. This latter area is not always considered to be part of regulatory
purview but has been coming under scrutiny as the “fourth regulatory hurdle” (Paul &
Trueman, 2001). In fact these two areas may be linked; regional labels offering differing
interpretations on indicated populations affect the type of access that is permitted, can
confuse healthcare providers, and can affect the pricing assigned to a product for
reimbursement. As described in Dr. Robert Temple’s article Translating study results to
labeling, there are elements of medical judgment in deciding whether a particular “study
population entirely defines the labeled indication” (Temple, 2011). At issue in this article
was the criticism that some design elements for the JUPITER trial (rosuvastatin), such as
study population and particular endpoints, were not reflected equally in the prescribing
information in the US, Canada, and the European Union (Temple, 2011). Dr. Temple
opined that regulatory decisions can be made either to strictly limit an indication to the
exact population studied, or the indication statement can ultimately describe a population
that is broader than the population studied, based on medical judgment. Thus, an
indication statement is another point linked to benefit:risk where regulatory dissonance is
introduced across regions.
138

4) Regional differences: The regulatory capacity in some regions of the world is only
now developing, and varies in the level of regulatory sophistication (Morrison & Singh,
2012). Local populations for which treatments are intended are often not represented in
the pivotal clinical studies that supported approval in more mature markets. Thus ethnic
bridging data or sensitivity analyses may have to be carried out in local populations to
satisfy DRAs reluctant to approve product on the basis of ‘foreign’ clinical data. The
estimated number of patients needed to satisfy these regional requirements has often
“little scientific basis to it and can at times become a barrier for their inclusion in the
global development” (Morrison & Singh, 2012). The logistical challenges that this
presents can delay access to these markets, often by years, if there is no feasible path to
address these requirements within the main multi-regional clinical trial (which is often
the case due to prolonged clinical trial application review periods). For instance, one
participant noted that such requirements could add up to 4-5 years of delay to product
availability, particularly in non-western countries. These perspectives on the
extrapolation of global clinical trial data to a new region, although relevant to regulatory
dissonance, are not PMO specific and involve detailed analysis of current ICH and local
requirements to address (Chow & Hsiao, 2010). Thus, any apparatus for scientific advice
that seeks to diminish important dissonance of regulatory requirements must address the
regional needs of regulators from emerging economies, such as those of Asia where local
data requirements are most pronounced.
The qualitative insights in regulatory dissonance identified by the respondent comments are
consistent with research by Huckle, who examined divergence in regulatory decisions between
the FDA and EMA between 2005 and 2008 across all indications. Huckle estimated a “64%
concordance in approval of submissions for the same products and 36% divergence,” where “the
139

EU rejected 31 applications approved by US and the US rejected 24 applications that the EU had
approved” (Huckle, 2010). The reasons most cited for these divergent decisions included
requirements related to the number of required pivotal studies, the need for placebo versus
comparator studies, and the acceptability of ‘foreign data’.
Interestingly, these data suggest that the EMA and FDA reject applications subsequently
approved by the other agency at approximately the same rate. However, a different picture
emerges when the numbers of approvals versus rejections are examined at the level of the
individual indication. PMO applications in contrast to applications in general had a
preponderance of negative review decisions that were mostly given by US rather than EU DRAs.
Over the last decade (2003-2013), only 3 novel PMO agents were approved in the US (excluding
products with the same active ingredient as an approved product). These were ibandronate
(Boniva), zoledronic acid (Reclast) and (denosumab) Prolia (CenterWatch, 2013). By
comparison during the same period, the EMA authorized 8 novel agents: teriparatide (Forsteo),
strontium ranelate (Protelos, Osseor), ibandronate (Bonviva), zoledronic acid (Aclasta),
parathyroid hormone (rDNA) (Preotact), lasofoxifene tartrate (Fablyn), bazedoxifene (Conbriza)
and denosumab (Prolia) (EMA, 2013c). When one excludes Forsteo from this analysis for the
reason that it was approved by the FDA as Forteo in 2002, the results are nevertheless
remarkable. Novel first-in-class products for this indication were approved for marketing only
43% of the time by the FDA when compared to approvals by the EMA. The complete lists of
PMO products approved by the FDA and EMA since 1995 are provided in Appendix III and
Appendix IV, respectively. This observation underscores the need to understand regulatory
dissonance at the level of the individual product class and disease state.
In sum, result of this study show that divergent guidelines and treatment of benefit:risk contribute
to regulatory dissonance, but these issues appeared to rank somewhat lower in significance by
140

respondents when compared to the dissonance resulting from regulatory advice received during
premarket meetings and regulatory reviews. However, policy initiatives directed at
harmonization often focus quite narrowly or exclusively on one or another of the areas studied
here, or are limited in their scope. A more holistic view is to develop and implement a cluster of
policy solutions that address different aspects regulatory dissonance, targeted at different points
in product development, in order to increase the efficacy of policy initiatives directed towards
harmonizing regional requirements. It can also be said that potentially effective mechanisms to
reduce regulatory dissonance are currently underutilized by DRAs, such as the alignment of key
regulatory guidance documents (i.e., FDA and EMA osteoporosis guidelines) or the
implementation of parallel or cross-regional advice procedures that lead to binding agreements.
5.4 Potential Advantages of Regulatory Dissonance
It is important to recognize that respondents did not have a completely negative view of the
impact of regulatory dissonance. Some respondents felt that regulatory dissonance is integral to
scientific debate, and thus is an essential aspect of scientific and drug development processes that
should not necessarily be discounted. For instance, respondents highlighted that different
scientific or regulatory views may prompt additional clinical questions to be explored within the
main registration study. Regulatory dissonance may also offer companies added advantage of
greater access to a diversity of expertise, ideas and concerns. In this sense, the scientific and
regulatory challenges that are presented to sponsors may push them to develop better products
and answer additional scientific questions, so that the resulting prescribing information is more
robust and informative. The fact that one respondent opined that divergent DRA expertise might
uncover unforeseen issues in protocol design suggests a positive feedback loop that may
ultimately prevents expensive trial mistakes. Regulatory dissonance may also combat the
psychological aspects of groupthink and silos that pervade any insular organization. If many of
141

the respondents found the regulatory review process to be an important driver that improves
product quality, one must question whether efforts to assure greater regulatory convergence and
efficiency may have untoward effects on product quality overall. The positive aspects of
regulatory dissonance should be considered in future policy as well. For example, some level of
regulatory dissonance might encourage broader clinical development programs that would
ultimately provide a broader data package to support global marketing authorizations, and could
encourage a more robust and thoughtful level of decision making relative to benefit:risk. Of
course, such dissonance can inadvertently create an uneven playing field by allowing access to
inferior or riskier products in certain markets. On balance, it is difficult to escape the conclusion
that greater convergence of product-specific guidance could in a positive way act to raise the bar
on regulatory standards globally, and thus needs to be recognized as an important element of
global public health policy.
5.5 Regulatory Dissonance and Implications for Business Strategy
Implications for business strategy
Many factors can operate when strategic decisions are made regarding product investment
portfolios. Research by Pammolli and colleagues have noted a significant industry shift in R&D
expenditure towards riskier, pharmaceutical targets (Pammolli, 2011). Their research regarding
the fate of 28,000 compounds since 1990 showed that the decline of R&D productivity,
associated with increasing product attrition rates, was associated with an “increasing
concentration of R&D investment in areas in which the risk of failure is high, which correspond
to unmet therapeutic needs and unexploited biological mechanisms” (Pammolli, 2011). Pammolli
reasoned that visions of potentially high future revenue appear to offset less palatable estimates of
high development costs. The findings of this current thesis may suggest that the impact of
regulatory dissonance on development costs and potential revenue may be underestimated by
142

firms developing new therapies in some disease indications, such as PMO. This may mean that
research candidates optimistically introduced to fill product pipelines may fall out of clinical
development as companies assume less risk in the face of regulatory challenges. In support of
this interpretation, about half of the respondents felt that regional regulatory dissonance increased
the downstream costs of PMO drugs to patients/payers, decreased the level of industry innovation
in drug research and development in PMO, increased the regulatory authority review timelines,
and lowered the commercial probability of success (POS) or the net present value (NPV) of a
potential new PMO agent. A case in point may very well be Aprela (conjugated
estrogen/bazedoxifene), shown in Figure 2: the primary completion date of its phase 3 study was
September 2008, yet marketing authorization applications were only recent resubmitted to the
FDA and EMA in 2012 and regulatory action is expected Q4 of 2013. Such delays to launch and
regulatory uncertainties can be absorbed only by the largest firms, such as the sponsor for this
product (Pfizer). Armed with the knowledge of the impact of regulatory dissonance, few smaller
firms will find this particular therapeutic area attractive for future investment given the challenges
that such delays and uncertainties impose. Future research should be undertaken to estimate these
effects more systematically across other disease areas.
It is not yet clear whether the lessons regarding regulatory dissonance identified here would be
externally valid for other drug classes, such as oncology or diabetes, where equivalent research
has yet to be undertaken and long-standing guidance documents exist. Unlike oncology, a
condition that is generally accepted to be serious and life-threatening, PMO is a chronic
asymptomatic condition until a fragility fracture presents (Levine, 2011). Further, clinical
sequelae are often considered as an inevitable consequence of aging, rather than symptoms of
underlying disease requiring treatment. Even following a hip fracture, patients are often neither
diagnosed nor treated for the disease although such patients are at high risk for a repeat fracture
143

(Lewiecki, 2009). Thus from the vantage of patients, the potential benefits of any particular PMO
therapy may not be seen to outweigh the risks and inconvenience associated with a particular
treatment regimen, until of course it is too late. These perceptions may be in part responsible for
the relatively poor adherence to therapeutic regimens amongst patients; 20-30% of patients
discontinue therapy 6-12 months of initiating therapy (Papaioannou, 2011) (Warriner & Curtis,
2009). Also, primary healthcare providers who treat osteoporosis patients are generally
complacent about initiating PMO therapy until a fracture presents or when a patient is deemed to
possess multiple risk factors, such as advancing age or a BMD T score lower than -2.5.
Regulatory agencies, too, may be less willing to accept risk for products where treatment benefit
versus risk is not as clearly defined, and may increasingly be taking a cue from PMO patients.
The FDA’s recent emphasis on patient-centered evaluation of benefit:risk is a case in point, since
the patients themselves are now seen as “the best judges of the risks that patients are willing to
bear” (Milne & Kaitin, 2012). We might speculate that in disease states such as PMO, which are
not widely perceived as serious by the patient, a greater tendency towards regulatory dissonance
prevails, caused by the general apathy towards the benefits of treatment in the face of the
potential risks. It is possible that public outreach and education programs, that are non-
promotional or therapy-specific in nature, may do much to influence the views and actions of
patients and regulatory authorities. That being said, the current situation makes PMO a
particularly interesting target in which to study regulatory dissonance. However, more research
will be needed to understand if the types of regulatory dissonance identified in this study are a
common feature across other therapeutic areas, including those where the disease state is clearly
recognized to be serious and for which therapy is essential.
The combined challenges of all of these competing factors may eventually drive pharmaceutical
companies to focus unduly on therapeutic areas where unmet medical need is greatest, where
144

regulatory pathways may be inherently more aligned, where industry-DRA cooperation is
greatest, and where return on investment is greatest due to premium pricing of such products.
Examples of such areas are rare disease or oncology indications, where consensus on unmet
medical need is clear, and as a result, products may be fast-tracked to market for patients in
serious need (Arrowsmith, 2012). Some evidence of this trend towards R&D consolidation in
specific disease areas has appeared in recent research. In a division-by-division comparison of
drug approvals by the FDA between the years 2011 and 2013, a time when 82 NMEs were
approved, the rate of approvals across reviewing divisions was highly uneven, with the largest
proportion of approvals occurring in oncology (Prevision Policy, 2013). Approval metrics
researched by Prevision Policy showed that “three review divisions that make up the Office of
Hematology and Oncology Products (OHOP) accounted for 35% (29 of 82) of all the NMEs
approved over the past two-and-a-half years.” By comparison during the same timeframe, only
4% (3 of 82) of all the NMEs approved by the FDA occurred in the Division of Bone,
Reproductive and Urologic Products (Office of Drug Evaluation III) (Prevision Policy, 2013). Of
note, based on research in Chapter 2, none of these appear to be products for PMO. Thus the
subtle message may be that although PMO is an underdiagnosed condition where a vast
serviceable population exists, treatment rates are continuing to decline and biopharmaceutical
companies may be steering R&D investment away from this and other indications with similar
benefit:risk profiles due perhaps in part to regulatory challenges.
Amongst respondents for whom regulatory dissonance was seen to be an important factor when
assessing the monetary value of a PMO program, respondents noted that the value of a program is
affected in large part by two factors: the cost of development and the delays due to regulatory
dissonance. The former may simply be a function related to developing treatments for chronic
and degenerative diseases, which as DiMasi points out, are conditions “generally more costly to
145

test, as they typically require more complex patient care and monitoring, longer periods for
effects to be observed, or larger trial sizes to establish their efficacy” (DiMasi, 2003). However,
the second point is a function of unexpected and divergent advice pertaining to clinical trial or
registration requirements. For instance, regulatory dissonance from a DRA in a key market that
acts to increase study size, duration, or complexity, may create uncertainties not only with respect
to the future registration status in that market, but in all other regions, since a single dataset is
generally required to support global registrations. There is also the purely ethical concern related
to exposing patients longer than is necessary to potentially inefficacious treatments and to
unnecessary procedures in a needlessly complex study. One specific case cited by a survey
respondent was one in which a “subset of regulatory authorities” required that the primary
endpoint be evaluated at 3 rather than 2 years, and this had a negative impact on the valuation of
this particular agent. Others highlighted the significance of regulatory dissonance from DRAs in
non key markets, where DRAs with less scientific sophistication may be significant contributors
to regulatory dissonance. Although the criteria for evaluating the probability of technical and
regulatory success often take divergent advice from different agencies into account, the
potentially negative impact on financial returns are often underestimated, and the probability of
regulatory success (PORS) is often overestimated.
Implications on program valuation
The implications of this study do not end with a call to regional DRAs to align guidance, provide
parallel product-specific recommendations and agreement on a set of benefit:risk assessments
tools. They also may be instructive for the companies as well, as those companies attempt to
predict costs and product value. Firms must shield their organizations and investors from
unexpected risk by assessing the regulatory risk posed by dissonance, which might appear from
this survey to be underestimated.
146

The financial risks associated with drug development are generally reflected by sponsors in
discounted cash flows, and are commonly given value using methods such as risk-adjusted net
present value, or rNPV, as proposed by Stewart and colleagues in Nature Biotechnology (Stewart,
2001). This method was originally developed by Stewart to assess more generally the future cash
flows of products in high-risk industries, such as the biotechnology industry. In such industries,
developmental cycles can be lengthy and unpredictable, thus the standard NPV model used to
predict cash flows in other industries was considered inadequate (Stewart, 2001). Studies have
since shown that techniques, such as rNPV, have better performance in estimating the valuation
of R&D projects (Liu and Wen, 2006). Nonetheless, these methods still appear to be limited in
their ability to predict future outcomes of therapeutic development programs, and one reason may
be the relative lack of attention to future regulatory risk.
An inspection of the methodology to calculate NPV gives some insight into the potential
challenge of calculating value quantitatively. The rNPV formula, shown below in Figure 17,
discounts intrinsic risk in the development of novel biotechnology products, on the basis of
anticipated revenue, cost, risk, and time (Stewart, 2001).
147

Figure 17. The Risk-Adjusted New Present Value Formula
1

rNPV =
NPV
PR
0
- ∑ C
i
R
0
/ R
i

i=0

rNPV (risk-adjusted net present value): The current value of a biotechnology when
revenue, risk, costs, and time are all taken into account; the fair selling price of a
biotechnology
P: Payoff
C
i
: Each associated cost
R
0
: Current risk mediated; the likelihood that a biotechnology will reach the market

R
i
: Risk mediated after 1 years have passed with success
R
0
/R
i
(risk adjustment factor): The likelihood that a cost, revenue, milestone, payment,
or investment will actually materialize
1
Stewart et al, 2001
However, the assumptions that are made in this model are often unduly narrow. More
particularly, the risk adjustment factor, R
0
/R
i,
often does not consider regulatory risk, such as
regulatory dissonance (RD), and is often disregarded either because risk projections based on
regulatory intelligence may lead to projections that are unreasonably pessimistic and are
dismissed, or because the firm’s own regulatory staff are not proactively engaged in the rNPV
assessment process. Because regulatory dissonance can have such a strong effect on ultimate
product timelines and costs, RD must be assessed more rigorously as part of R
0
/R
i.
evaluation,
early in the development process.
Using the formula in Figure 18 one can also factor in the risks associated with regulatory
dissonance described in this thesis, by adding an additional independent variable that reflects
regulatory dissonance observed in a specific therapeutic area or drug class, RD
obs
, as observed
from currently available regulatory intelligence. The term, RD
obs,
may be estimated from a
knowledge of dissonance in approval precedence, regulatory guidelines, and
advice/recommendations of similar products, for example. Moreover, one might postulate that a
148

quantitative representation of projected regulatory benefit:risk at the projected time of approval,
or B/R
proj
, may also be assigned to particular disease states to further factor for the likelihood of
future regulatory success. For instance, for disease states that are life-threatening and represent
unmet medical needs such as pancreatic or liver cancer (e.g., where regulatory dissonance is
expected to be lower, and where fast-track approval pathways are more likely to be made
available), a B/R
proj
factor of >1.0 may be assigned. For conditions such as postmenopausal
osteoporosis where many available therapies exist or are under development by competitors, and
where treatment options are underutilized due to lack of disease state awareness, a downwardly
weighted B/R
proj
, or <1.0, may be projected to reflect the lower chances of market authorization.
Figure 18: Modified Regulatory Risk-Adjusted New Present Value Formula
ⁿ
rNPV =
NPV
PR
0
- ∑ C
i
R
0
/ R
i
RD
obs
(R
0
/Ri) B/R
proj
(R
0
/Ri)

i=0

Although RD
obs
and B/R
proj
are proportional, one is an estimate of regulatory dissonance in the
current regulatory environment based on regulatory intelligence for a product within its disease
state, and the other is an estimate of the regulatory environment at the time of projected approval.
It cannot be known with certainty whether a qualitative technique of this kind could have foretold
with enough precision the risk that was associated with regulatory dissonance (risk) for the PMO
approval precedents described in Chapter 2.3. However, the regulatory dissonance observed in
these cases had financial consequences that cannot be ignored and rigorous regulatory assessment
is needed. For companies that focus their R&D efforts on one or a few disease states, accurate
estimates of RD
obs
and B/R
proj
for a particular disease state is critical so that risk is appropriately
discounted by corporate managers and is accurately communicated to investors.
149

5.6 Regulatory Dissonance and Implications for Policy Development
Where are the areas of potentially high priority for policy makers?
In the literature review, two areas of PMO drug approval standards were described as ones where
continued scientific/ethical controversy exists (Chapter 2.4, Ethical Controversies in PMO Drug
Approval Standards). These included (1) whether pivotal, registration studies should be placebo-
controlled or active-controlled, and (2) whether the duration of study treatment should be a
minimum of three years for a novel agent versus a shorter duration. However, it appears from
this analysis that the list of areas of some priority is broader.
Nearly all of the evaluated aspects of clinical trial design in phase 3 registration studies in PMO
could benefit from a convergence of DRA advice, recommendations and guidelines. The results
strongly suggest a need for greater alignment on multiple domains of trial conduct. However,
four areas of concern may be prioritized based on the responses received: type of control group,
target population, timing of the primary assessment, and intrinsic factors related to benefit:risk.
These findings are striking; guidance documents for the development of PMO agents have been
available in the US since 1979 and in the EU since 1997, yet organic convergence of regulatory
requirements has not yet occurred. The results underscore the fact that policy aimed at decreasing
regulatory dissonance should not stop with the alignment of regional guidance documents alone.
Individual policy measures alone, such as the harmonization of regional guidance documents
communicating clinical trial expectations, will not act to reduce regulatory dissonance, and must
be coupled with other processes to optimize their efficacy.
When asked which regulatory mechanisms might be most beneficial to reduce regulatory
dissonance in drug development, the two mechanisms most commonly identified were
(1) harmonized regulatory authority guidelines (29%) and (2) joint scientific advice procedures
150

(27%). Relatively less emphasis was placed on harmonized procedures for the assessmentof
benefit:risk (15%) and establishing mutual recognition agreements (12%). This finding is
interesting because the area that seems to be currently receiving the most recent attention from
regulators with the endpoint of harmonization are those directed at developing a common
benefit:risk framework and methodology (see Chapter 2.6, Status on Developing a Common
Benefit:Risk Framework). Perhaps the reason that regulators are focused on this issue relate to
the fact that the concept of benefit:risk analysis is neither product specific or clinical trial specific,
but rather simply cuts across most product classes. Clearly, developing a common benefit:risk
framework is important, but data from this survey indicate that it may not be the only important
area of policy that must be synchronized if regulatory dissonance is to be effectively reduced.
The implications of these results suggest that harmonized guidance documents alone might not be
particularly efficacious. Guidance harmonization may need to be coupled with other
mechanisms, such as comprehensive parallel advice mechanisms amongst regulatory authorities
that produce binding agreements (optimally including key emerging market DRAs), and the
optimization of benefit:risk procedures and or mutual recognition procedures (although these two
solutions for reducing regulatory dissonance were ranked somewhat lower).
If greater indication/disease specific harmonization is needed, how do we get there? Respondents
clearly felt that two main types of regulatory bodies should take the lead in implementing
mechanisms to reduce regulatory dissonance. When asked to make a single selection,
respondents appeared to favor regulatory agencies or ICH as the most appropriate institutions to
lead the initiatives. It was surprising that only 5% of respondents favored pharmaceutical
industry associations, such as EFPIA or PhRMA, given that it is exactly these industry
associations that have recently been most active in calling for change. These agencies have
lobbied the US Federal Office of Information and Regulatory Affairs (OIRA) and the EU
151

Directorate-General for Enterprise and Industry (DG ENTR) respectively to move in the direction
of greater “US EC Regulatory Compatibility” (PhRMA-EFPIA, 2012). The open letter from
PhRMA/EFPIA dated 31 October 2012 gives strong support for initiatives that can reduce
regulatory differences and duplicative requirements that can impede efficiency in global drug
development, review and evaluation. Several broad ranging “regulatory compatibility” proposals
were outlined as a means to reduce transatlantic regulatory dissonance. These included,
1) Greater Coordination to Reduce Regulatory Burden for Sponsors, that contained a proposal for
a more accessible parallel scientific advice process with binding outcomes, 2) Increased
Collaboration under the Auspices of the ICH to Secure Greater Regulatory Compatibility (which
contained a proposal to develop a harmonized structural framework and methodology for benefit-
risk assessment), and 3) Miscellaneous Proposals (which included a proposal for a collaborative
process for developing therapeutic area guidelines) (PhRMA-EFPIA, 2012). The need for such a
multi-pronged approach is reinforced by this study, but the respondents clearly recognized that
only regional DRAs and accepted regulatory harmonization bodies such as ICH have the ability
to bring about substantive changes with respect to these regulatory policy issues.
Given that large, multi-national pharmaceutical companies have large global ‘footprints’, it would
thus seem important that drug makers, as key stakeholders, advocate strongly for harmonization
of indication-specific expectations through organizations such as PhRMA (Pharmaceutical
Research and Manufacturers of America) and EFPIA (European Federation of Pharmaceutical
Industries and Associations), as well as their respective governments. Divergent indication-
specific guidance and product-specific advice is a barrier in the global fight against disease, and is
therefore a global public policy issue that requires a much greater degree of attention by regional
policymakers than it currently receives. Thus, much advantage can be gained if data sharing,
collaborative review mechanisms, consensus on regulatory guidance documents and review
152

standards could be established, especially in regions whose regulatory structures are only
developing, and before country-specific approaches begin to diverge at an increasing rate.
Nevertheless, policy change is difficult and often contentious. The challenges of harmonizing the
shifting regulatory standards for PMO agents are underscored by the FDA’s recent withdrawal in
December 2009 of the 1994 Draft Postmenopausal Osteoporosis Guidance (FDA, 2009b) and
EMA’s 2007 revision to their osteoporosis guidelines. In part, the withdrawal of the US guidance
may relate to the inherent problems of satisfying the concerns of varying US constituencies such
as the medical community, patient advocacy and industry, or the difficulty of its application for
differing types of products. As Chapter 2 suggests, the medical community is divided on its
views regarding some aspects of PMO drug development. If reasoned consensus cannot be
reached within the medical community or a single DRA, the obstacles may be too great for a
transatlantic or, even more challenging, global consensus on regulatory guidelines for registering
new osteoporosis agents. Currently, the role and responsibility of regional DRAs and ICH in
reducing regulatory dissonance has not always been set as a high priority. However, an
overarching finding of this study is that greater international cooperation and coordination
amongst regulatory DRAs is needed, and this will require broad leadership and cooperative action
from regional authorities.
5.7 How Best to Effect Change in the Area of Regulatory Dissonance of
Indication-specific Regulatory Requirements on a Global Level?
At one end of a spectrum of possibilities are narrow initiatives to reduce regulatory dissonance
one issue at a time. Of course, this may have greater likelihood of success from a short term
implementation standpoint or a political point of view, but may fail to have the impact needed to
deal with a growing multifaceted global problem. At the other end of the policy spectrum are
omnibus efforts to change a range of barriers to harmonization concurrently. Such an approach
has the obvious challenges relating to the difficulty of gaining consensus on particular elements.
153

It may be that a more intermediate and targeted model to reduce regulatory dissonance is needed
in which different initiatives are directed at particular phases of product development and/or life-
cycle. Harmonized guidance documents may be most helpful in earlier phases of development,
prior to the initiation of pivotal trials when the clinical program is being designed and basic
regulatory boundaries are needed to aid in planning. Parallel scientific advice may be most
helpful at key points during the clinical study phase and into product review to ensure inter-
agency alignment on key issues pre-registration (i.e., at end of phase 2, pre-filing, and meetings
during regulatory review). Harmonized benefit:risk procedures at the registration phase can act to
minimize divergent review outcomes and divergent product labeling.
5.8 Conclusions and Future Directions
The results of the study prompt concerns that regulatory dissonance cannot be reduced by a single
mechanism of harmonization. Interregional collaboration on guidance documents and
convergence with respect to decision making by various DRAs must be implemented in tandem
to facilitate harmonization. However, this case study of a single disease state should be
supplemented by the assessment of regulatory dissonance across other therapeutic areas and
disease states to establish the generality of these conclusions. It is hoped that this exploratory
research will prompt follow-on study of regulatory dissonance in other product sectors, to see if it
is vulnerable to class-related effects. Moreover, although results of this study reflect views
derived from only one stakeholder, that of industry, it does not mean that this work need stop
with the industry input received to date. There may be value in sharing the results of this study
with regulators, with the goal of developing a consensus opinion, so that the research reflects the
broadest of views.
For future study, it may be useful to question whether industry respondents can identify a trend
toward increasing or decreasing regulatory dissonance over time. The world may be at an
154

inflection point where a finite window of opportunity for reducing regulatory dissonance exists.
As the world becomes economically ‘flatter’, and emerging nations become increasingly more
influential by developing autonomous regulatory frameworks for regulatory decision making, we
might expect that they will develop regulatory requirements consistent with their own cultural
and political needs, yet ones as demanding as those in the main ICH regions. An increase in
regulatory dissonance may be an inevitable feature of global regulatory affairs for the coming
decades.
In the area of osteoporosis, this is already being felt with the dissemination of divergent technical
guidance in emerging markets and with the requirements for ethnic bridging studies by many
DRAs. Thus, the optimal time for DRAs and industry to collaborate on solutions for greater
alignment of regulatory decision making may be now.
155

LITERATURE REFERENCES
Adachi JD, Josse RG, Russell RG. If you don’t take it – it can’t work: the consequences of not
being treated or nonadherance to osteoporosis therapy. Therapeutics and Clinical Risk
Management. 2011; 7:181-198.
Andrews EB, Gilsenan AW, Midkiff K, et al. The US Postmarketing Surveillance Study of Adult
Osteosarcoma and Teriparatide: Study Design and Findings From the First 7 Years. J Bone
Miner Res. 2012;27(12):2429-2437.
Arrowsmith J. From the Analyst’s Couch: A Decade of Change. Nature Reviews – Drug
Discovery. January 2012. 2012;11:17-18.
ASBMR, 2002. Osteoporosis Trials: Ethical Considerations in Study Design (conference held
14 – 15 June 2002, Bethesda, MD). The American Society for Bone and Mineral Research
(ASBMR). Accessed on 14 May 2012 at:
http://www.asbmr.org/TopicalMeetings/PastTopicalMeetings.aspx.
Augustus, S. Drug Development in Oncology: A Regulatory Perspective. American Journal of
Therapeutics. 2011;18:323-331.
Beck LE. President Obama’s International Regulatory Harmonization Initiative – An Opinion.
Federal Regulations Advisor. Accessed on 03 July 2012 at:
http://www.fedregsadvisor.com/2012/05/02/president-obamas-international-regulatory-
harmonization-initiative-an-opinion/.
Blake GM, Fogelman I. Strontium ranelate: a novel treatment for postmenopausal osteoporosis:
a review of safety and efficacy (Review). Clinical Intervention in Aging. 2006;1(4):367-375.
Bliuc D, Nguyen ND, Milch VE. Mortality Risk Associated With Low-Trauma Osteoporotic
Fracture and Subsequent Fracture in Men and Women. JAMA. 2009;301(5):513-521.
Breckenridge A. Regulatory divergence (Clinical Trials Commentary). Clinical Trials.
2011;8:425-426.
156

Brody BA; Dickey N, Ellenberg SS, et al. Is the Use of Placebo Controls Ethically Permissible in
Clinical Trials of Agents Intended to Reduce Fractures in Osteoporosis. J Bone Miner Res.
2003;18(6):1105-1109.
Brugha R, Varvasovszky Z. Stakeholder analysis: a review (Review article). Health Policy and
Planning. Oxford University Press. 2000;15(3):239-246.
Canalis E, Giustina A, Bilezikian JP. Medical Progress: Mechanisms of Anabolic Therapies for
Osteoporosis (Review Article). N Engl J Med. 2007;357(9):905-916.
Capron A, Donaldson S, Raisz LG, et al. Osteoporosis Panel Summary. Journal of Bone and
Mineral Research. 2003;18(6):1160-1162.
CenterWatch, 2013. FDA Approved Drugs by Medical Condition. CenterWatch website.
Accessed on 04 August 2013 at: http://www.centerwatch.com/drug-information/fda-
approvals/drug-categories.aspx?View=O.
CFDA, 2011. Considerations for clinical trials evaluating the treatment of postmenopausal
women with osteoporosis drugs. 06 July 2011. State Food and Drug Administration Center for
Drug Evaluation. Zhang J, Wang, H. Record Number: Beijing ICP 09013725. Accessed on 04
April 2012 at: http://www.cde.org.cn/dzkw.do?method=largePage&id=312265.
Chen JS, Sambrook PN. Antiresportive therapies for osteoporosis: a clinical overview. Nature
Reviews Endocrinology. 2012;8:81-91.
Chow S-C, Hsiao CF. Bridging Diversity: Extrapolating Foreign Data to a New Region.
Pharmaceutical Medicine. 2010;24(6):349-362.
CIOMS, 1998. Benefit-Risk Balance for Marketed Drugs: Evaluating Safety Signals. Reports of
CIOMS Working Group IV. 1998 (Reprinted 2000, 2004). CIOMS. Accessed on 14 March 2012
at: http://www.cioms.ch/publications/g4-benefit-risk.pdf.
Colman EG, Hedin R, Swann J, et al. A brief history of calcitonin (Viewpoint). Lancet. 2002.
359:885-86.
157

Colman EG. The Food and Drug Administration’s Osteoporosis Guidance Document: Past,
Present, and Future. J Bone Miner Res. 2003;18(6):1125-1128.
Crandall CJ, Newberry SJ, Diamant A, et al. Comparative Effectiveness Review No. 53 –
Treatment to Prevent Fractures in Men and Women With Low Bone Density or Osteoporosis:
Update of a 2007 Report (AHRQ Publication No. 12-EHC023-EF). Prepared by Southern
California Evidence-based Practice Center – AQ RAND Corporation (under Contract No. HHSA-
290—2007-10062-I) for the Agency for Healthcare Research and Quality, U.S. Department of
Health and Human Services, Rockville, MD, USA. March 2012. Accessed 23 June 2013 at
www.effectivehealthcare.ahrq.gov/reports/final.cfm.
Cummings SR, Ettinger B, Delmas PD, et al. The Effects of Tibolone in Older Postmenopausal
Women. N Engl J Med. 2008;359;7:697-708.
Cummings SR, Ensrud K, Delmas, PD, et al. Lasofoxifene in Postmenopausal Women with
Osteoporosis. N Engl J Med. 2010;362:686-96.
Dent RE, Storm N, Steiner RS, et al. The Tactics of Modern-Day Regulatory Trials. Journal of
Bone and Joint Surgery. 2012;94Suppl1(E):39-44.
DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug
development costs. Journal of Health Economics. 2003;22:151-185.
Drugs, 2008. Bazedoxifene: bazedoxifene acetate, TSE 424, TSE-424, WAY 140424. Drugs in
R & D. 2008;9(3):191-6.
Ellenberg SS. Scientific and Ethical Issues in the Use of Placebo and Active Controls in Clinical
Trials. J Bone Miner Res. 2003;18(6):1121-1124.
EMEA, 1997. Note for Guidance on Involutional Osteoporosis in Women (CPMP/EWP/552/95).
EMEA. 25 September 1997. Available at the International Drug Regulatory Affairs
Compendium (IDRAC), No. 14473. Accessed on 26 June 2012 at:
http://www.idrac.com/idrac/DocInformation.aspx?ID=00014473®ion1=EU.
158

EMEA, 2001. Note for Guidance on Involutional Osteoporosis in Women (CPMP/EWP/552/95
rev 1). EMEA. 25 September 1997. Available at the International Drug Regulatory Affairs
Compendium (IDRAC), No. 27617. Accessed on 26 June 2012 at:
http://www.idrac.com/idrac/DocInformation.aspx?ID=00027617®ion1=EU.
EMEA, 2004a. Tenth annual report of the European Medicines Agency. EMEA. 2005.
EMEA/61492/2005/EN/Final. Accessed on 03 November 2012 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Annual_report/2009/12/WC50001669
6.pdf.
EMEA, 2004b. Forsteo, INN-teriparatide Scientific Discussion. EMEA. 2004. Accessed on 26
June 2012 at: http://www.emea.europa.eu/docs/en_GB/document_library/EPAR_-
_Scientific_Discussion/human/000425/WC500027992.pdf.
EMEA, 2005. Recommendation on the Need for Revision of CHMP Note for Guidance on Post-
menopausal Osteoporosis in Women (CPMP/EWP/552/95 rev 2). EMEA. 21 April 2005.
Available at the International Drug Regulatory Affairs Compendium (IDRAC), No. 50357.
Accessed on 26 June 2012 at:
http://www.idrac.com/idrac/DocInformation.aspx?ID=00050357®ion1=EU.
EMEA, 2006. Overview of Comments Received on Draft Guideline on the Evaluation of
Products in the Treatment of Primary Osteoporosis (Doc. Ref. CPMP/EWP/433354/2006).
EMEA. 16 November 2006. Available at the International Drug Regulatory Affairs
Compendium (IDRAC), No. 62834. Accessed on 26 June 2012 at:
http://www.idrac.com/idrac/DocInformation.aspx?ID=00062834®ion1=EU.
EMEA, 2007. Guideline on the evaluation of new medicinal products in the treatment of primary
osteoporosis (Doc. Ref. CPMP/EWP/552/95 Rev. 2). EMEA. 16 November 2006. Accessed on
08 March 2012 at: http://www.emea.europa.eu/pdfs/human/ewp/55295enfin.pdf.
EMA, 2010. Reflection paper on the need for active control in therapeutic areas where use of
placebo is deemed ethical and one or more established medicines are available (Draft). EMA.
November 2010. EMA/759784/2010/CHMP. Accessed on 07 August 2012 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/01/WC5001
00710.pdf.
159

EMA, 2011. Benefit-risk methodology project – Work package 1 report: description of the
current practice of benefit-risk assessment for centralised procedure products in the EU regulatory
network. EMA (Human Medicines Development and Evaluation). 25 May 2011.
EMA/227124/2011. Accessed on 17 August 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/07/WC500109478.pdf.
EMA, 2012a. Questions and answers on the review of Protelos and Osseor (strontium ranelate).
EMA. 15 March 2012. EMA/183047/2012. Accessed on 16 March 2012 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Medicine_QA/2012/03/WC500124208
.pdf.
EMA, 2012b. Fablyn (lasofoxifene): Cessation of validity of the marketing authorisation in the
European Union (Public Statement). EMA. 7 May 2012. EMA/296759/2012. Accessed on 17
August 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Public_statement/2012/06/WC500129
057.pdf.
EMA, 2012c. Questions and answers on the review of calcitonin-containing medicines. EMA. 19
July 2012. EMA/476001/2012. Accessed on 03 August 2012 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Referrals_document/Calcitonin_31/W
C500130149.pdf.
EMA, 2012d. Concept paper on the need for revision of the guideline on the evaluation of
medicinal products in the treatment of primary osteoporosis. EMA. 19 October 2012.
EMA/CHMP/520786/2012. Accessed on 26 October 2012 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Other/2012/10/WC500134108.pdf.
EMA, 2013a. Recommendation to restrict the use of Protelos / Osseor (strontium ranelate) –
CHMP confirms recommendation from the PRAC (Press Release). EMA. 26 April 2013.
Accessed on 16 June 2013 at:
http://www.ema.europa.eu/ema/index.jsp?curl=pages/news_and_events/news/2013/04/news_deta
il_001774.jsp&mid=WC0b01ac058004d5c1.
160

EMA, 2013b. Concept paper on the need for revision of the guideline on the evaluation of
medicinal products in the treatment of primary osteoporosis (EMA/CHMP/520786/2012). EMA.
16 October 2012. Accessed on 16 August 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/10/WC5001
34467.pdf.
EMA, 2013c. European public assessment reports (search criteria for osteoporosis therapies).
EMA. 16 October 2012. Accessed on 16 August 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/10/WC5001
34467.pdf.
EMA-FDA, 2008. Medicines Regulation: Transatlantic Administrative Simplification Action
Plan. EMA. 17 June 2008. EMA/440869/2011. Accessed on 16 March 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/07/WC500109672.pdf.
EMA-FDA, 2009. General Principles – EMA-FDA Parallel Scientific Advice. 22 July 2009.
EMEA/24517/2009. Accessed on 16 March 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Other/2009/11/WC500014868.pdf.
EMA-FDA, 2011. Transatlantic Administrative Simplification Action Plan – Final Report on
Implementation. 26 July 2011. EMA. EMA/440869/2011. Accessed on 16 March 2013 at:
http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/07/WC500109672.pdf.
Evans JR, Mathur A. The value of online surveys. Internet Research. 2005;15(2):195-219.
Faugier J, Sargeant M. Sampling hard to reach populations. Journal of Advanced Nursing.
1997;26:790-797.
FDA, 1994. Guidelines for Preclinical and Clinical Evaluation of Agents Used in the Prevention
or Treatment of Postmenopausal Osteoporosis (Draft, Withdrawn December 2009). FDA. 1994.
Division of Metabolism and Endocrine Drug Products, FDA, Bethesda, MD, USA.
161

FDA, 1998. Providing Clinical Evidence of Effectiveness for Human Drug and Biological
Products. 1998. FDA. Division of Metabolism and Endocrine Drug Products, FDA, Bethesda,
MD, USA. U.S. Department of Health and Human Services, Food and Drug Administration,
Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research
(CBER), FDA, Rockville, MD, USA.
FDA, 2000. Guidance for Industry – Development of Parathyroid Hormone for the Prevention
and Treatment of Osteoporosis (Draft Guidance). FDA. 2000. U.S. Department of Health and
Human Services, Food and Drug Administration, Center for Drug Evaluation and Research
(CDER), FDA, Rockville, MD, USA.
FDA, 2001. Endocrinologic and Metabolic Advisory Committee (DMEP) Proceedings
(Presentation slides from Bruce S. Schneider, MD, CDER, FDA). FDA. 27 July 2001. Accessed
09 March 2013 at: www.fda.gov/ohrms/dockets/ac/01/slides/3761s2_03_Schneider.ppt.
FDA, 2002a. Proceedings of a Metabolic and Endocrine Advisory Committee. FDA. September
2002. Accessed on 05 March 2012 at:
http://www.fda.gov/ohrms/dockets/ac/cder02.htm#EndocrinologicMetabolicDrugs.
FDA, 2002b. Forteo [teriparatide (rDNA origin)] Injection Drug Approval Package (Application
No.: 021318; Approval Date: 11/26/2002). FDA. Accessed on 16 March 2012 at:
http://www.accessdata.fda.gov/drugsatfda_docs/nda/2002/21-318_Forteo.cfm.
FDA, 2004. Guidance for Industry – Available Therapy. FDA. 1994. U.S. Department of Health
and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research
(CDER), Center for Biologics Evaluation and Research (CBER), FDA, Rockville, MD, USA
FDA, 2008a. Final Summary Minutes – Advisory Committee for Reproductive Health Drugs
Meeting on 08 September 2008. FDA. 2008. Accessed on 03 March 2012 at:
http://www.fda.gov/ohrms/dockets/ac/08/minutes/2008-4381m1-final.pdf.
FDA, 2009a. The Critical Path Initiative – Report on Key Achievements in 2009. FDA. 2009.
Accessed on 02 July 2012 at:
http://www.fda.gov/downloads/ScienceResearch/SpecialTopics/CriticalPathInitiative/UCM22165
1.pdf.
162

FDA, 2009b. CDER Guidance New/Revised/Withdrawn 1/1/2009 – 12/31/2009. Accessed on
03 March 2012 at:
http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/uc
m079659.pdf.
FDA, 2009c. Summary Minutes of the Advisory Committee for Reproductive Health Drugs.
FDA. 09 August 2009. Accessed on 20 July 2013 at:
http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/Repro
ductiveHealthDrugsAdvisoryCommittee/UCM249267.pdf.
FDA, 2010a. Guidance for Industry – Non-Inferiority Clinical Trials. FDA. March 2010. U.S.
Department of Health and Human Services, Food and Drug Administration, Center for Drug
Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER), FDA,
Rockville, MD, USA.
FDA, 2010b. U.S. Food and Drug Administration New Release: International Collaboration:
FDA and European Medicines Agency Agree to Accept a Single Orphan Drug Designation
Annual Report. FDA. 26 February 2010. Accessed on 08 March 2012 at:
http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm202300.htm.
FDA, 2011a. Joint Meeting of the Reproductive Health Drugs Advisory Committee and the Drug
Safety and Risk Management Advisory Committee Meeting Proceedings. FDA. 09 September
2013. Accessed on 20 July 2013 at:
http://www.fda.gov/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/DrugSafetyandRi
skManagementAdvisoryCommittee/ucm250295.htm.
FDA, 2011b. Guidance for Industry – E2F Development Safety Update Report. FDA. August
2011, ICH. U.S. Department of Health and Human Services, Food and Drug Administration,
Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research
(CBER), FDA, Rockville, MD, USA.
FDA, 2012. CDER Forum for International Drug Regulatory Authorities website. FDA. 2012.
Accessed on 08 March 2012 at: http://www.fda.gov/Drugs/NewsEvents/ucm167011.htm.
163

FDA, 2013a. Joint Meeting of the Advisory Committee for Reproductive Health Drugs and the
Drug Safety and Risk Management Meeting Proceedings. FDA. 05 March 2013. Accessed on
09 March 2013 at: http://www.fda.gov/AdvisoryCommittees/Calendar/ucm335406.htm.
FDA, 2013b. Guidance Agenda – New & Revised Draft Guidances CDER is Planning to Publish
During Calendar Year 2013. FDA. 2013. U.S. Food and Drug Administration website.
Accessed on 12 July 2013 at:
http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/U
CM314767.pdf .
FDA, 2013c. Structured Approach to Benefit-Risk Assessment in Drug Regulatory Decision-
Making – Draft PDUFA V Implementation Plan – February 2013 - Fiscal Years 2013-2017.
FDA. 2013. Accessed on 09 March 2013 at:
http://www.fda.gov/downloads/ForIndustry/UserFees/PrescriptionDrugUserFee/UCM329758.pdf.
Fizazi K, Carducci M, Smith M, et al. Denosumab versus zoledronic acid for treatment of bone
metastases in men with castration-resistant prostate cancer: a randomized, double-blind study.
Lancet. 2011;377(9768):813-822.
Forteo. Forteo
®
(teriparatide [rDNA origin] injection) US Package Insert. 2012. Eli Lilly & Co.
Accessed on 20 July 2013 at: http://pi.lilly.com/us/forteo-pi.pdf.
Fost N. Ethical Issues in Clinical Research on Fracture Prevention in Patients With Osteoporosis.
J Bone Miner Res. 2003;18(6):1110-1115.
FRAX, 2013. Kanis JA. FRAX WHO Fracture Risk Assessment Tool. World Health
Organization Collaborating Centre for Metabolic Bone Diseases, University of Sheffield, UK.
Sheffield UK. Accessed on 13 August 2013 at: http://www.shef.ac.uk/FRAX/index.aspx.
GAO, 2012. Prescription Drugs - FDA Has Met Most Performance Goals for Reviewing
Applications: Report to Congressional Requesters. United States Government Accountability
Office. March 2012. GAO-12-500. Accessed on 13 August 2013 at:
http://purl.fdlp.gov/GPO/gpo25071.
Gilgun, JF. Yes They Are: The Generalizability of Case Studies. Current Issues in Qualitative
Research. 2011;2(4).
164

Greenspan SL, et al. Bisphosphonates: Safety and Efficacy in the Treatment and Prevention of
Osteoporosis. Am Fam Physician. 2000. 1;61(9):2731-2736.
Greenspan SL, Bone HG, Ettinger MP, et al. Effect of Recombinant Human Parathyroid
Hormone (1-84) on Vertebral Fracture and Bone Mineral Density in Postmenopausal Women
with Osteoporosis: A Randomized Trial. Annals of Internal Medicine. 2007;146(5):326-39.
Grignolo A. Expanding Global Pharmaceutical Harmonization: A Look at ASEAN’s PPWG and
The America’s PANDRH. RAPS Regulatory Focus. March 2010;15(3):8-14.
Hamowy, R. Medical Disasters and the Growth of the FDA. Independent Policy Report –
Enlightening ideas for public policy. The Independent Institute. February 2010. Accessed on 23
June 2012 at: http://www.independent.org/pdf/policy_reports/2010-02-10-fda.pdf.
Heaney RP. Ethical Issues in the Design of Osteoporosis Clinical Trials: The State of the
Question. J Bone Miner Res. 2003;18(6):1117-1120.
Handcock MS, Gile KJ. On the Concept of Snowball Sampling. Sociological Methodology.
2011;41(1)367-371.
Hill S, Johnson K. Emerging Challenges and Opportunities in Drug Registration and Regulation
in Developing Countries (Issues Paper – Access to Medicine). DFID Health Systems Resource
Centre. 2004. Accessed on 08 July 2012 at:
http://www.hlsp.org/LinkClick.aspx?fileticket=0TXMdaAk5KA%3D&tabid=1643.
HMA (Heads of Medicines Agencies). Clinical Trials Facilitation Groups – Guidance document
for a Voluntary Harmonisation Procedure (VHP) for the assessment of multinational Clinical
Trials Applications – Version 2. March 2010. Doc. Ref.: CTFG//VHP/2010/Rev 1. Accessed on
24 June 2012 at: http://www.hma.eu/fileadmin/dateien/Human_Medicines/01-
About_HMA/Working_Groups/CTFG/2010_03_VHP_Guidance_v2.pdf.
Huckle P. Are today’s regulatory submissions flawed? An industry perspective: Workshop
Report, 2010. CMR International Institute for Regulatory Science. London, UK, September
2008.
165

ICH, 2010. The Value and Benefits of ICH to Drug Regulatory Authorities – Advancing
Harmonization for Better Health. International Conference on Harmonisation of Technical
Requirements for Registration of Pharmaceuticals for Human Use. 2010. Accessed on 08 March
2012 at:
http://www.ich.org/fileadmin/Public_Web_Site/News_room/C_Publications/ICH_20_anniversary
_Value_Benefits_of_ICH_for_Regulators.pdf.
Kanis JA, Alexandre J-M, Bone HG, et. Al. Study Design in Osteoporosis: A European
Perspective. Journal of Bone and Mineral Research. 2003;18(6):1133-1138.
Kanis JA, McCloskey EV, Johansson H, et al. European guidance for the diagnosis and
management of osteoporosis in postmenopausal women. Osteoporos Int. 2013. 24:23-57.
Kehoe T. Bone Quality: A Perspective from the Food and Drug Administration. Current
Osteoporosis Reports. 2006;4:76-79.
Khosla S. Surrogates for Fracture Endpoints in Clinical Trials. J Bone Miner Res.
2003;18(6):1146-1149.
Kumar M, Mao Y. Meeting With FDA Can Increase the Probability of Product Approval. RAPS
Regulatory Affairs Focus. October 2010. Accessed on 07 August 2012 at:
http://www.amarexcro.com/articles/docs/RAPS_Focus_Meeting_with_FDA_Oct2010.pdf.
Lewiecki EM. Managing osteoporosis: Challenges and strategies. Cleveland Clinic Journal of
Medicine. 2009;76(8):457-466.
Lewiecki EM, Bilezikian, EM. Denosumab for the Treatment of Osteoporosis and Cancer-
Related Conditions. Clinical Pharmacology & Therapeutics. 2012;91(1):123-133.
Levine RJ. Placebo Controls in Clinical Trials of New Therapies for Osteoporosis. J Bone Miner
Res. 2003;18(6):1154-1159.
Levine JP. Identification, Diagnosis, and Prevention of Osteoporosis. Am J Manag Care.
2011;17S:S170-S176. Accessed on 23 June 2013 at:
http://www.ajmc.com/publications/supplement/2011/a357_may11/A347_11may_Levine_S170to
176.
166

Liberti L, McAuslane N, Walker SRW. Progress on the development of a benefit/risk framework
for evaluating medicines. RAPS Regulatory Focus. March 2010; 32-27.
Liu T-L, Wen S-Y. Evaluations of Drug Development Status in Biopharmaceutical Industry.
The XVI ACME (Association for Chinese Management Educators) International Conference on
Pacific Rim Management. 27 – 29 July 2006.
Mazziotti G, Canalis, E, Giustina, A. Drug-induced osteoporosis: mechanisms and clinical
implications. Am J Med. 2010;123(10):877-84.
Melton III LJ. Adverse Outcomes of Osteoporotic Fractures in the General Population. J Bone
Miner Res. 2003;18(6):1139-1141.
Michor S, Diaz S. Mutual Recognition Agreements in the EU, Canada and Other Countries.
RAPS Regulatory Focus. March 2010; 16-20.
Michor S, Ochalski S. Community Referral Procedures. RAPS Regulatory Focus. March 2010;
28-31.
Milne C-P, Kaitin KI. FDA Review Divisions: Performance Levels and the Impact on Drug
Sponsors. Clinical Pharmacology & Therapeutics. 2012;91(3): 393-404.
MHRA, 2007. Medicines and Healthcare Products Regulatory Agency. UK Public Assessment
Report Tibolone (Livial): benefit-risk evaluation. Accessed on 30 July 2013 at:
http://www.mhra.gov.uk/home/groups/pl-p/documents/websiteresources/con2032229.pdf.
Modelska K, Cummings S. Clinical Review 140 – Tibolone for Postmenopausal Women:
Systematic Review of Randomized Trials. The Journal of Clinical Endocrinology & Metabolism.
2002;87(1):16-23.
Morrison A, Singh R. Rethinking emerging markets for regulatory affairs with the creation of a
Regulatory Authority Development Index. Regulatory Rapporteur. 2012;9(4):5-7.
OECD, 2013. OECD Recommendation on the Governance of Clinical Trials. OECD. Accessed
on 16 August 2013 at: http://www.oecd.org/sti/sci-tech/oecd-recommendation-governance-of-
clinical-trials.pdf.
167

Olson S, Claiborne AB. Forum on Drug Discovery, Development, and Translation -
Strengthening a Workforce for Innovative Regulatory Science in Therapeutics Development
(Workshop Summary – 20 -21 September 2011). Sponsored by the Institute of Medicine of the
National Academies. The National Academies Press. 2012.
Ormarsdottir S, Reginster J-Y, Abadie E. European regulatory perspectives for innovative
therapies. Osteoporos Int. 2008;19(6):725-731.
OTA, 1994. Office of Technology Assessment, U.S. Congress. Public Information About
Osteoporosis: What’s Available, What’s Needed? – Background Paper. U.S. Congress, Office of
Technology Assessment. OTA-BP-H-131. Washington, DC: U.S. Government Printing Office,
July 1994.
Pammolli F, Magazzini L, Riccaboni M. The productivity crisis in pharmaceutical R&D
(Analysis). Nature Reviews. June 2011;10(428-438).
Papaioannou A, Kennedy C, Dolovich L, et al. Patient Adherence to Osteoporosis Medication:
Problems, Consequences and Management Strategies. Drugs Aging. 2007;24(1):37-55.
Paul JE, Trueman P. ‘Fourth hurdle review’, NICE, and database applications.
Pharmacoepidemiol Drug Saf. 2001;10(5):429-38.
Pharmaceutical Medicine, 2012. Drugs in Clinical Development for Osteoporosis.
Pharmaceutical Development – Summary and Table (Adis Profile Summary). 2012;26(4):249-
252.
PhRMA-EFPIA, 2012. Re: Request for Public Comments on Promoting U.S. EC Regulatory
Compatibility. Filed electronically 31 October 2012. Accessed on 24 June 2013 at:
http://ec.europa.eu/enterprise/policies/international/cooperating-governments/usa/jobs-
growth/files/consultation/regulation/24-efpia-phrma_en.pdf.
PMSB, 1999. Guidelines for Clinical Evaluation Methods of Drugs for Osteoporosis. Evaluation
and Medical Safety Bureau, Pharmaceutical and Medical Safety Bureau, Ministry of Health and
Welfare, Japan (PMSB/ELB). 15 April 1999. PMSB/ELD No. 742.
168

Prevision Policy, 2013. FDA Approvals, Division-by-Division: Oncology Outpaces All Others
by Wide Margin In Approvals (and Post-Market Burden) (Regulatory and Policy Research
Notes). Prevision Policy Health Policy + Business Forecasts. 2013 (02 July 2013). Accessed
05 August 2013 at http://www.previsionpolicy.com/research/regulatory-policy/2013/07/08/fda-
approvals-division-by-division-oncology-outpaces-all-others-by-wide-margin-in-approvals-and-
post-market-burden/.
Prolia 2013. Prolia
®
(denosumab) US Package Insert. 2013. Amgen Inc. Accessed on 20 July
2013 at: http://pi.amgen.com/united_states/prolia/prolia_pi.pdf.
Rampell, C. ‘Great Recession’: A Brief Etymology. New York Times, Economics – Explaining
the Science of Everyday Life. 11 March 2009. Accessed 14 April 2012 at:
http://economix.blogs.nytimes.com/2009/03/11/great-recession-a-brief-etymology/.
Recker, RR. Osteoporosis Trials: Ethics in Study Design: Proceedings of Two Conferences.
J Bone Miner Res. 2003;18(6):1116.
Reginster JY, Abadie E, Delmas P, et al. Recommendations for an update of the current (2001)
regulatory requirements for registration of drugs to be used in the treatment of osteoporosis in
postmenopausal women and in men. Osteoporos Int. 2006;17:1-7.
Riggs L, Hodgson SF, O’Fallon M, et. al. Effect of Fluoride Treatment on the Fracture Rate in
Postmenopausal Women with Osteoporosis. N Engl J Med. 1990;322:802-809.
Rising K, Bacchetti P, Bero L. Reporting Bias in Drug Trials Submitted to the Food and Drug
Administration: Review of Publication and Presentation. PLOS Medicine. 2008;5(11):1561-
1570.
Rosen CJ, Khosla S. Placebo-controlled Trials in Osteoporosis—Proceeding with Caution.
NEJM. 2010;363(14):1365-1367.
Rosenblatt M. Is It Ethical to Conduct Placebo-Controlled Clinical Trials in the Development of
New Agents for Osteoporosis? An Industry Perspective. J Bone Miner Res. 2003;18(6):1142-
1145.
169

Servier, 2004. Osteoporosis – primary prevention: Servier response to the NICE Statement of
Reasons and clarification on Servier response. NICE (National Institute for Health and Care
Excellence) Website. Accessed on 18 June 2012 at:
http://guidance.nice.org.uk/TA/Wave6/39/FAD3/ManufacturerResponse.
Silverman, 2008a. Silverman SL, Christiansen C, Genant HK, et al. Efficacy of Bazedoxifene in
Reducing New Vertebral Fracture Risk in Postmenopausal Women With Osteoporosis: Results
From a 3-Year, Randomized, Placebo-, and Active-Controlled Clinical Trial. Journal of Bone
and Mineral Research. 2008;23(12):1923-1934.
Silverman, 2008b. Silverman SL, Cummings SR, Watts NB, et al. Recommendations for the
Clinical Evaluation of Agents for Treatment of Osteoporosis: Consensus of an Expert Panel
Representing the American Society for Bone and Mineral Research (ASBMR) the International
Society for Clinical Densitometry (ISCD), and the National Osteoporosis Foundation (NOF).
Journal of Bone and Mineral Research. 2008;23(1):159-165.
Sobieraj DM. Bazedoxifene: An investigational selective estrogen receptor modulator for the
treatment and prevention of osteoporosis in postmenopausal women (Abstract). Formulary.
2011;46:159-176.
Solberg M, Richmond FJ. Transparency in Drug Submission Processes of 3 Asian Countries: A
Survey of Industry Views. Drug Information Journal. 2012;45(2):216-225.
Stein CM, Ray WA. The Ethics of Placebo in Studies with Fracture End Points in Osteoporosis.
NEJM. 2010;363(14):1367-1370.
Stewart KK, Allison PN, and Johnson RS. Putting a price on biotechnology. Nature
Biotechnology. 2001;19:813-817.
Stopeck AT, Lipton A, Body J-J, et al, on behalf of the 20050136 Breast Cancer SRE Study
Investigators. Denosumab compared with zoledronic acid for the treatment of bone metastases in
patients with advanced breast cancer: a randomized, double-blind study. J Clin Oncol.
2010;28:5132-5139.
170

Storm T, Thamsborg G, Steiniche T, et al. Effect of Intermittent cyclical Etidronate Therapy on
Bone Mass and Fracture Rate in Women with Postmenopausal Osteoporosis. NEJM.
1990;33:18:1265-1271.
Temple RJ. Study Designs in Osteoporosis. J Bone Miner Res. 2003;18(6):1129-1132.
Temple RJ. Translating study results to labeling. Clinical Trials. 2011;8:427-429.
TGA, 2011. Australia New Zealand Therapeutic Products Agency (ANZTPA) fact sheet.
Australian Government, Department of Health and Ageing, Therapeutic Goods Administration
Website. Accessed on 14 August 2012 at: http://www.tga.gov.au/about/international-anztpa-
factsheet.htm#what.
Thomas R, Henshaw R, Skubitz K, et al. Denosumab in patients with giant-cell tumour of bone:
an open-label, phase 2 study. The Lancet Oncology. 2010;11:3:275-280.
UMBRA, 2013. Benefit-Risk Assessment Support System (BRASS) Update. UMBRA (Unified
Methodologies for Benefit-Risk Assessment) Update Issue 2. February 2013. Accessed on 14
February at:
http://cirsci.org/sites/default/files/CIRS%20UMBRA%20Issue%202%20Feb%202013.pdf.
Varvasovszky Z, Brugha R. How to do (or not to do)…A stakeholder analysis. Health Policy
and Planning. Oxford University Press. 2000;15(3):338-345.
Von Eschenbach A. Andrew von Eschenbach: Toward a 21
st
-Century FDA (Opinion). The Wall
Street Journal. 15 April 2012. Accessed 07 August 2012 at:
http://online.wsj.com/article/SB10001424052702303815404577331673917964962.html.
Walker, 2011a. Walker, S; McAuslane, N; Liberti, L. Refining the benefit-risk framework for
assessing medicines: valuing and weighting the parameters. Reg Affairs J. 2011;March:14-17.
Accessed on 18 June 2012 at the Centre for Innovation in Regulatory Science (CIRS) website:
http://213.120.141.158/.
171

Walker, 2011b. Walker, S; McAuslane, N; Liberti, L. Developing a common benefit-risk
assessment methodology for medicines – a progress report (Regulatory Features). Scrip
Regulatory Affairs. December 2011 Issue. Accessed on 18 June 2012 at the Centre for
Innovation in Regulatory Science (CIRS) website: http://213.120.141.158/.
Warriner AH, Curtis, JR. Adherence to osteoporosis treatments: room for improvement. Current
Opinion in Rheumatology. 2009;21:356-362.
Watts NB, Harris ST, Genant HK, et al. Intermittent Cyclical Etidronate Treatment of
Postmenopausal Osteoporosis. NEJM. 1990;323(2):73-79.
Weijer C. The Ethics of Placebo-Controlled Trials. J Bone Miner Res. 2003;18(6):1150-1153.
Whitaker M, Guo J, Kehoe T, Benson G. Bisphosphonates for Osteoporosis – Where Do We Go
from Here? N Engl J Med. 2012. 366;22:2048-2051.
WHO (World Health Organization), 1998. Guidelines for preclinical evaluation and clinical trials
in osteoporosis. World Health Organization. Geneva, Switzerland. 1998. Accessed on 05
August 2012 at: http://whqlibdoc.who.int/publications/1998/9241545224_eng.pdf.
Zafiropoulos N, Phillips L, Pignatti F, Xavier L. Evaluating benefit:risk: An Agency
perspective. Regulatory Rapporteur. 2012;9(6):5-8.
172

APPENDIX I: SURVEY ON REGULATORY DISSONANCE IN THE
GLOBAL DEVELOPMENT OF PHARMACOTHERAPIES FOR
POSTMENOPAUSAL OSTEOPOROSIS
I. Professional profiles of respondents
(1) How many total years of drug development experience do you have in the
biopharmaceutical industry?
(2) How many years of experience do you have specifically developing therapeutic products
intended for the treatment/prevention of postmenopausal osteoporosis (PMO)?
(3) What has been your role within your organization? Please check all that apply.
(4) Please indicate the region in which you have acquired the majority of your
regulatory/clinical experience? Please check all that apply.
(5) What phases of drug development have you supported for therapeutic products intended for
the treatment/prevention of PMO? Please check all that apply.
(6) How many different therapeutic products for PMO have you supported during your
regulatory/clinical career?
(7) In your role, did your job responsibilities require you to interact directly with any of the
following? Please check all that apply.
II. Impact/implications of regulatory dissonance
(8) In your experience, has Regulatory Dissonance increased the CHALLENGES associated
with developing drug therapies in PMO?
(9) In your experience, has Regulatory Dissonance increased the CLINICAL
DEVELOPMENT TIMELINES of PMO drugs (i.e., delays to study start, lengthier clinical
studies)? By how much time?
173

(10) In your experience, has Regulatory Dissonance in drug registration requirements increased
the REGULATORY AUTHORITY REVIEW TIMES of marketing authorization
applications of PMO drugs? Please estimate time added to standard application reviews.
(11) In your experience, has Regulatory Dissonance increased the SIZE of multi-regional,
pivotal, phase 3 clinical studies in PMO? By how many subjects?
(12) In your opinion, has Regulatory Dissonance increased the COMPLEXITY of phase 3 trial
designs or registration programs in PMO?
(13) In your experience, does Regulatory Dissonance impact the eventual COSTS of PMO
drugs for patients/payers?
(14) In your experience, does Regulatory Dissonance impact PATIENT ACCESS to important,
new treatments for PMO (i.e., products not approved or delayed to market)?
(15) In your experience, does your company consider Regulatory Dissonance to be an important
factor when assessing the COMMERCIAL PROBABILITY OF SUCCESS OR THE NET
PRESENT VALUE (NPV) of a potential new PMO agent?
(16) In your opinion, how does Regulatory Dissonance impact INDUSTRY INNOVATION in
drug research and development (as it relates to PMO)?
(17) Some have suggested that a certain amount of Regulatory Dissonance in regional
guidelines/requirements for developing drug therapies may actually be beneficial
to sponsors of investigational agents. Do you agree?
III. General causes/factors contributing to regulatory dissonance
(18) In your experience, which of the following lead to the most Regulatory Dissonance in the
setting of PMO? Please check the items that best apply.
174

IV. Alignment of regional indication-specific PMO guidance
(19) In your experience, how significant is the level of Regulatory Dissonance in the
recommendations for the TYPE OF CONTROL ARM (e.g., placebo or active control) in
pivotal, phase 3 clinical studies in PMO?
(20) In your experience, how significant is the level of Regulatory Dissonance in the
recommendations for the DURATION OF TREATMENT in pivotal, phase 3 clinical
studies in PMO?
(21) In your experience, how significant is the level of Regulatory Dissonance in the
recommendations for the DURATION OF LONG-TERM SAFETY FOLLOW-UP
following the assessment of the primary endpoint?
(22) In your experience, how significant is the level of Regulatory Dissonance in
the recommendations for the PRIMARY EFFICACY ENDPOINT in pivotal, phase 3
clinical studies in PMO?
(23) In your experience, how significant is the level of Regulatory Dissonance in
the recommendations for the SECONDARY ENDPOINTS in pivotal, phase 3 clinical
studies in PMO?
(24) In your experience, how significant is the level of Regulatory Dissonance in the
recommendations for the TIMING OF THE PRIMARY ENDPOINT ASSESSMENT (e.g.,
24 vs. 36 months) in pivotal, phase 3 clinical studies in PMO?
(25) In your experience, how significant is the level of Regulatory Dissonance
in the recommended TARGET POPULATION to be studied in the pivotal, phase 3 clinical
studies of investigational agents for PMO?
(26) In your view, do INTRINSIC FACTORS relating to the investigational agent under study
(e.g., safety profile, mechanism of action, novelty of product) impact the level of
Regulatory Dissonance?
(27) Amongst the areas identified above, which in your view results in the greatest degree
of Regulatory Dissonance when designing pivotal, phase 3 studies in PMO?
175

(28) Please assess the level of Regulatory Dissonance that you have observed between drug
regulatory authorities in their assessments of BENEFIT:RISK for the same PMO agent.
IV. Dissonance relating to drug review and approval procedures
(29) Please assess the level of Regulatory Dissonance resulting from divergent DRUG
REGULATORY AUTHORITY FEEDBACK received during product reviews, meetings or
negotiations.
(30) Please assess the level of Regulatory Dissonance between drug regulatory authorities in
terms of ALLOWABLE INDICATION STATEMENTS for product labeling.
(31) In your view, are there other sources of Regulatory Dissonance that arise during the
development of PMO agents that should be considered by this study?
If you do not believe that there are other sources of Regulatory Dissonance, please indicate
"none".
V. Potential policy solutions to reduce regulatory dissonance
(32) In your view, which of the following potential mechanisms would be most beneficial in
reducing Regulatory Dissonance? Please check one item.
(33) In your view, which of the organizations listed below should take the lead in implementing
mechanisms to reduce Regulatory Dissonance? Please check one item.
(34) Given your drug development experience, do you think that the mechanism you selected
above to reduce Regulatory Dissonance in PMO would be helpful if applied in other
disease areas?
(35) This is the end of the survey. Thank you for your responses and time. Please feel free to
elaborate on any of your responses in this survey. Please also indicate whether you would
be open to discussing your feedback for the purpose of informing this study.
176

APPENDIX II: CROSS TABULATION DATA ANALYSIS
This exploratory study was relatively small in size. Nonetheless, important trends are noted in
Chapter 4, for which the raw data is provided.
i.

ii.

APPENDIX III: COMPLETE LIST OF POSTMENOPAUSAL OSTEOPOROSIS THERAPIES APPROVED BY EMA SINCE 1995
7 Westferry Circus. Canary Wharf. London. E14 4HB
Tel +44 (0)20 7418 8400. Fax +44 (0)20 7418 8416
E-mail info@ema.europa.eu Website www.ema.europa.eu
An Agency of the European Union
Adapted from: http://www.ema.europa.eu/ema/index.jsp?curl=pages%2Fmedicines%2Flanding%2Fepar_search.jsp&mid=WC0b01ac058001d124&searchTab=&alreadyLoaded=true
Medicine Name Product Number
Active Substance
Marketing Authorisation
Holder Status Authorisation date
Aclasta EMEA/H/C/000595 zoledronic acid Novartis Europharm Limited Authorised 15/04/2005
Adrovance EMEA/H/C/000759
alendronate sodium trihydrate /
colecalciferol Merck Sharp & Dohme Ltd. Authorised 04/01/2007
Bondenza (previously Ibandronic
Acid Roche) EMEA/H/C/000502 ibandronic acid Roche Registration Ltd. Withdrawn 23/02/2004
Bonviva EMEA/H/C/000501 ibandronic acid Roche Registration Ltd. Authorised 23/02/2004
Conbriza EMEA/H/C/000913 bazedoxifene Pfizer Ltd. Authorised 17/04/2009
Ei EMEA/H/C/000184 lif hd hl id Dii hiS k E G bH Ah i d 05/08/1998 Evista EMEA/H/C/000184 raloxifene hydrochloride Daiichi Sankyo Europe GmbH Authorised 05/08/1998
Fablyn EMEA/H/C/000977 lasofoxifene tartrate
Dr. Friedrich Eberth
Arzneimittel GmbH
Withdrawn 24/02/2009
Forsteo EMEA/H/C/000425 teriparatide Eli Lilly Nederland B.V. Authorised 10/06/2003
Fosavance EMEA/H/C/000619 alendronic acid / colecalciferol Merck Sharp & Dohme Ltd. Authorised 24/08/2005
Ibandronic Acid Teva EMEA/H/C/001195 ibandronic acid Teva Pharma B.V. Authorised 17/09/2010
Optruma EMEA/H/C/000185 raloxifene hydrochloride Eli Lilly Nederland B.V. Authorised 05/08/1998
Osseor EMEA/H/C/000561 strontium ranelate Les Laboratoires Servier Authorised 21/09/2004
Preotact EMEA/H/C/000659 parathyroid hormone (rDNA) Nycomed Danmark ApS Authorised 24/04/2006
Prolia EMEA/H/C/001120 denosumab Amgen Europe B.V. Authorised 26/05/2010
Protelos EMEA/H/C/000560 strontium ranelate Les Laboratoires Servier Authorised 21/09/2004
Raloxifene Teva EMEA/H/C/001075 raloxifene hydrochloride Teva Pharma B.V. Authorised 29/04/2010
Vantavo (previously Alendronate
Sodium And Colecalciferol,
MSD) EMEA/H/C/001180 alendronic acid / colecalciferol Merck Sharp & Dohme Ltd. Authorised 16/10/2009
Zoledronic acid Teva Pharma
EMEA/H/C/002437
zoledronic acid Teva Pharma B.V. Authorised 16/08/2012
177
APPENDIX IV: COMPLETE LIST OF POSTMENOPAUSAL OSTEOPOROSIS THERAPIES APPROVED BY FDA SINCE 1995
Adapted from: http://www.centerwatch.com/drug-information/fda-approvals/drug-categories.aspx?View=O
Medicine Name
Active Substance Indication License Application Holder Status Authorisation date
Activella Estradiol/norethindrone acetate Prevention of postmenopausal osteoporosis Novo Nordisk Approved 18 April 2000
Alora Estradiol transdermal patch Prevention of postmenopausal osteoporosis Watson Pharmaceuticals Approved 15 April 2002
Boniva Ibandronic acid
Treatment and prevention of postmenopausal
osteoporosis Roche / GlaxoSmithKline Approved 16 May 2003
Estradiol tablets Estradiol Prevention of postmenopausal osteoporosis Barr Laboratories Approved 28 October 1997
Estratab Esterified estrogen Prevention of postmenopausal osteoporosis Solvay Pharmaceuticals Discontinued 11 March 1998
Evista Raloxifene hydrochloride Prevention of postmenopausal osteoporosis Eli Lilly & Co. Approved 09 December 1997
Evista Raloxifene hydrochloride Treatment of postmenopausal osteoporosis Eli Lilly & Co. Approved 30 September 1999
Femhrt Progestin-estrogent combination Prevention of postmenopausal osteoporosis Parke-Davis Approved 15 October 1999
Forteo Teriparatide
Treatment of postmenopausal women with
osteoporosis at high risk for fracture Eli Lilly & Co. Approved 26 November 2002
Fosamax Alendronate
Treatment and prevention of postmenopausal
osteoporosis Merck Approved 29 September 1995
Miacalcin
Synthetic calcitonin-salmon nasal
spray Treatment of postmenopausal osteoporosis Sandoz Pharmaceuticals Approved 17 August 1995
Ortho-Prefest Progestin-estrogent combination Prevention of postmenopausal osteoporosis
Ortho-Mc Neil Pharmaceutical,
R.W. Johnson Approved 22 October 1999
Premarin Conjuated estrogen tablets Prevention of postmenopausal osteoporosis Wyeth Approved 16 July 2003
Prempro & Premphase
Conjugated
estrogens/medroxyprogesterone acetate Prevention of postmenopausal osteoporosis Wyeth Approved 17 November 1995
Prolia Denosumab
Treatment of postmenopausal women with
osteoporosis at high risk for fracture Amgen Approved 01 June 2010
Reclast Zoledronic acid Treatment of postmenopausal osteoporosis Novartis 17 August 2007
Vivelle Estradiol transdermal system (ETS) Prevention of postmenopausal osteoporosis Novartis Approved 23 August 2000
178

Abstract (if available)

Abstract The purpose of regulatory indication-specific guidance documents, standardized regulatory review procedures and benefit:risk assessments are to ensure the safety and effectiveness of drug products and to set a common regulatory bar for industry. However, nationally-developed regulatory standards for clinical trial conduct and drug approval become potentially problematic when companies must operate across differing regional constituencies and address the requirements of various stakeholders to attain market authorizations globally. When this occurs, companies face dissonance in regulatory review and approval requirements and standards. The absence of a clear global regulatory pathway complicates the development of novel therapies since it is assumed to result in added uncertainty, time, costs and delays to the development of important treatments for patients. The findings indicate that regulatory dissonance is multifaceted, with perhaps greatest impact on application review outcomes, clinical trial design and timelines, but also on areas as diverse as allowable labeling claims and reimbursement. The degree of downward pressure on drug development and innovation from regulatory dissonance may be underestimated by industry and may be disease state dependent. The results also suggest that dissonance cannot be reduced by a single mechanism of harmonization. Integration of advisement and decision-making procedures across Drug Regulatory Authorities (DRAs) and harmonization of national regulatory guidelines may be most efficacious in reducing regulatory dissonance. Additional benefit may result if these are implemented in parallel with other recognizable methods of harmonization, such as standardization of benefit:risk methodology and through expanded use of mutual recognition pacts.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Benefits-risk frameworks: implementation by industry

PDF

Regulatory CMC strategies for gene and cell therapies during mergers and acquisitions: a survey of industry views

PDF

Regulatory harmonization in a resource-limited setting: the World Health Organization Collaborative Procedure for Accelerated Registration

PDF

Views on global harmonization of pharmacopeial standards: a survey of key stakeholders

PDF

Regulatory agreements for drug development collaborations: practices in the medical products industry

PDF

Challenges in the implementation of Risk Evaluation Mitigation Strategies (REMS): a survey of industry views

PDF

Evaluation of FDA-sponsor formal meetings on the development of cell and gene therapies: a survey of industry views

PDF

Experience with breakthrough therapy designation: an industry survey

PDF

Regulation of pediatric cancer drug development: an industry perspective

PDF

Examining the regulatory framework for drug compounding: industry views and experiences

PDF

The impact of FDA Patient-Focused Drug Development (PFDD) on US drug development strategies: a survey of views from pharmaceutical product companies

PDF

Incentivizing quality in the manufacture of pharmaceuticals: manufacturers' views on quality ratings

PDF

Risk approaches and standards used in hospitals: a survey of industry views

PDF

Regulating cosmeceuticals in the United States: a cosmetic industry view

PDF

Current practices in biocompatibility assessment of medical devices

PDF

Design control for software medical devices: an industry survey of views and experiences

PDF

U.S. veterinary drug shortages: industry views on potential changes to regulatory policy

PDF

Regulatory team development in post-merger integration: a survey of views from medical product companies

PDF

Contract research organizations: a survey of industry views and outsourcing practices

PDF

Implementation of tobacco regulatory science competencies in the tobacco centers of regulatory science (TCORS): stakeholder views

Asset Metadata

Creator Storm, Neal E. (author)

Core Title Regulatory dissonance in the global development of drug therapies: a case study of drug development in postmenopausal osteoporosis

School School of Pharmacy

Degree Doctor of Regulatory Science

Degree Program Regulatory Science

Publication Date 11/08/2013

Defense Date 08/29/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag benefit:risk,dissonance,divergence,drug development,drug therapy,global,global development,guidance document,guideline,OAI-PMH Harvest,osteoporosis,postmenopausal osteoporosis,regulation,regulatory,regulatory affairs,regulatory agency,regulatory authority,regulatory dissonance,regulatory divergence

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Richmond, Frances J. (committee chair), Cosenza, Mary Ellen (committee member), Pacifici, Eunjoo (committee member), Rodgers, Kathleen E. (committee member)

Creator Email nstorm@amgen.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-345661

Unique identifier UC11295434

Identifier etd-StormNealE-2142.pdf (filename),usctheses-c3-345661 (legacy record id)

Legacy Identifier etd-StormNealE-2142.pdf

Dmrecord 345661

Document Type Dissertation

Format application/pdf (imt)

Rights Storm, Neal E.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA