University of Southern California Dissertations and Theses

Bilateral testicular germ cell tumors: distribution of histologic types

(USC Thesis Other)

Bilateral testicular germ cell tumors: distribution of histologic types

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content BILATERAL TESTICULAR GERM CELL TUMORS: DISTRIBUTION OF
HISTOLOGIC TYPES
by

Jennifer Phay Thomas

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)

August 2010

Copyright 2010 Jennifer Phay Thomas

ii

Acknowledgements
I would like to acknowledge and deeply thank my mentor Dr. Victoria Cortessis
for her encouragement and support in both the academic and personal aspects
of this paper, my career, and my life.

I would like to thank my thesis committee members, Dr. Ann Hamilton and Dr.
Kim Siegmund for their insight and valuable advice on this thesis and working
with me to understand the findings more fully. I would like to thank as well as my
friends for their support, especially Carol Davis-Dao for never letting me quit.
Thank you to my husband Alex and my father Allen for their support and love.

Lastly, I would like to acknowledge three people who have driven my work yet
are no longer here to enjoy the celebration; Mr. Morgan Hobbs, who taught me
girls can excel in science, Bo Johnson, who taught me never to stop fighting nor
give up, and to Teresa who taught me to strive to be kind to others no matter how
hard my life is.
iii

Table of Contents

Acknowledgments ................................................................................................. ii
List of Tables ........................................................................................................ iv
Abstract ................................................................................................................ vi
Chapter One: Background .................................................................................... 1
Chapter Two: Methods ......................................................................................... 6
Study subjects and methods ........................................................................... 6
Definitions of variables .................................................................................. 11
Estimates of percentage of seminomatous and non-seminomatous TGCT .. 12
Statistical Analyses ....................................................................................... 13
Literature Review .......................................................................................... 15
Chapter Three: Results ...................................................................................... 16
Estimation of distribution of TGCT by age groups ......................................... 16
Predictive value of the histologic type of a first TGCT on the histologic type
of a second TGCT ......................................................................................... 16
Comparison of men with bilateral TGCT to others with TGCT ...................... 25
Figure 1: Comparison of mean age at first/only diagnosis of unilateral and
bilateral cases by broad histologic type ........................................................ 26
Literature Review .......................................................................................... 31

Chapter Four: Discussion ................................................................................... 35
References ......................................................................................................... 38
Appendices ......................................................................................................... 41
Appendix A-1:Specific Histologic Type and Age at Diagnosis of First and
Second Tumors Among 550 Men with History of Two Primary Testicular
Germ Cell Tumors ........................................................................................ 41
Appendix A-2: Comparison of Unilateral and Bilateral cases by Race and
Ethnicity ........................................................................................................ 44
Appendix A-3: Comparison of Unilateral and Bilateral Cases by year of
SEER registry origin ..................................................................................... 47

iv

List of Tables
Table 1: Specific Histologic Types of Germ cell Tumors Belonging to the
Broadly Defined Categories, Seminoma and Non-Seminoma (ICD-0 code) ........ 3

Table 2: 550 Men with History of Two Primary Testicular Germ Cell Tumors, by
Method of Identification, Broadly Categorized Tumor Histology, and Age at
Diagnosis of each TGCT .................................................................................... 18

Table 3: Broad Histologic Type and Age at Diagnosis of First and Second
Tumors Among 550 Men with History of Two Primary Testicular Germ Cell
Tumors ............................................................................................................... 20

Table 4a: 2x2 table for unadjusted Odds Ratio estimate of seminomatous first
diagnosis predicting a seminomatous second diagnosis ................................... 22

Table 4b: Estimated Odds Ratios for a seminomatous first diagnosis predicting
a seminatous second diagnosis by unconditional logistic regression among all
bilateral cases. ................................................................................................... 22

Table 4c: Estimated Odds Ratios for a seminomatous first diagnosis predicting
a seminatous second diagnosis by unconditional logistic regression among
population based bilateral cases ........................................................................ 23

Table 4d: Estimated Odds Ratios for a seminomatous first diagnosis predicting
a seminatous second diagnosis by unconditional logistic regression among
synchronous cases ............................................................................................. 24

Table 4e: Estimated Odds Ratios for a seminomatous first diagnosis predicting
a seminatous second diagnosis by unconditional logistic regression among
asynchronous cases ........................................................................................... 24

Table 5a: Analysis of Unilateral cases and Bilateral cases matched by year of
birth and registry of diagnosis, estimated Odds Ratio of a seminomatous
first/only diagnosis predicting an outcome of bilateral TGCT .............................. 28

Table 5b: Analysis of Unilateral cases and Bilateral cases matched by year of
birth and registry of diagnosis, estimated Odds Ratio of a seminomatous
first/only diagnosis predicting an outcome of bilateral TGCT diagnosed after
1980 ................................................................................................................... 28

v

Table 5c: Analysis of Unilateral cases and Bilateral cases matched by year of
birth and registry of diagnosis, estimated Odds Ratio of a seminomatous
first/only diagnosis predicting an outcome of bilateral TGCT diagnosed after
1990 ................................................................................................................... 28
Table 6a: Comparison of terms addressing heterogeneity within broadly
defined seminomatous TGCT as a risk factor for development of bilateral
TGCT, comparing pure seminoma, 9061, dysgerminoma 9060, and
germinoma 9064 to other seminoma .................................................................. 30

Table 6b: Comparison of terms addressing heterogeneity within broadly
defined seminomatous TGCT as a risk factor for development of bilateral
TGCT, comparing pure seminoma, 9061, to other seminomas .......................... 31

Table 6c: Comparison of terms addressing heterogeneity within broadly
defined seminomatous TGCT as a risk factor for development of bilateral
TGCT, comparing anaplastic seminoma 9062, to other seminoma .................... 31

Table 7: Previously published studies addressing histologic types in bilateral
TGCT ................................................................................................................. 33

Appendix A-1:Specific Histologic Type and Age at Diagnosis of First and
Second Tumors Among 550 Men with History of Two Primary Testicular Germ
Cell Tumors ....................................................................................................... 41

Appendix A-2: Comparison of Unilateral and Bilateral cases by Race and
Ethnicity ............................................................................................................. 44

Appendix A-3 Comparison of Unilateral and Bilateral Cases by year of SEER
registry origin ..................................................................................................... 47

vi

Abstract

Testicular germ cell tumors (TGCT) are the most commonly diagnosed neoplasm
of young men 15-45 years of age, and incidence of these tumors has been
increasing in recent decades for reasons currently unknown. Histologic type of
TGCT can be divided into the broad categories of seminomas and non-
seminomas. It currently remains to be elucidated if these histologic types share
a common etiology. In this study we analyzed the largest set of bilateral cases of
TGCT ever assembled, 550, to determine if the histologic type of the first tumor
predicts that of the second. Presuming an individual with a history of two primary
tumors has the same risk factors for both tumors, a concordance of histologic
types in the same individual would suggest that the risk factors for seminomas
and non-seminomas differ. The data show no association between the histologic
type of the first and second tumors when the analysis is adjusted for the age at
first diagnosis (odds ratio (OR)=0.95) This finding suggests that the two broad
histologic types share a common etiology, but some age related factor impacts
differentiation between the two. We also compared 16,204 presumed unilateral
cases of TGCT to 440 bilateral cases matched on year of birth and region of
diagnosis to determine if histologic type of the first/only TGCT confers risk of
developing a second TCGT. This analysis suggests that a first diagnosis of
seminoma is a risk factor for developing testicular cancer after adjustment for
age at first or only diagnosis (OR=1.401; 95% CI=1.135-1.730). This finding
vii

suggests that seminomatous first tumors should be considered as a risk factor for
the development of bilateral TCGT despite the current view that non-
seminomatous TGCT are considered to be more aggressive.
1

Chapter One: Background

Testicular Germ Cell Tumors (TGCT) are the most commonly diagnosed
neoplasm of young men in the United States and Europe. Based on the most
recent national statistics available, (2003-2007), 5.4 per 100,000 men per year
will be diagnosed in the United States,
(http://seer.cancer.gov/statfacts/html/testis.html). With the introduction of
platinum-based chemotherapy in the 1970s, long term survival rates improved
dramatically and now exceed 90%. As survival has increased it has become
apparent that TGCT survivors may experience important sequelae including
infertility (Cortessis 2003), sexual dysfunction (Walsh 2009), and elevated risk of
risk of additional cancers (Cortessis 2003).

For decades, incidence of TGCT have increased steadily in most western
countries (Garner 2005), including the United states where age-adjusted
incidence rates rose from 1.46 per 100,000 for the period 1935-39 to 5.01 per
100,000 during the period 1990-92 (Zheng 1996). Since this trend is not
explained by changes in known risk factors, it is postulated to result from one or
more unknown but increasingly common environmental exposures. New insights
into TGCT etiology are therefore highly desired, not only to understand why
TGCT have become more common, but also as the basis of new preventive
measures.
2

TGCT occurs in three characteristic age groups; infants, young men, and elderly
men. However, the great majority of cases (approximately 85%) occur among
young men 14-45 years of age (www.seer.cancer.gov, see full citations SEER
website), among whom TGCT can occur as any of 16 specific histologic types, or
as several combinations thereof. These histologic types are broadly categorized
into two groups, seminomas and non-seminomas, as shown in Table 1. Differing
clinical and epidemiological properties are observed for these broadly defined
histologic types. For example, seminomas tend to be diagnosed at slightly older
ages (Cortessis 2003) and to have better prognosis, particularly in the context of
disseminated disease (Nichols 2003). Mixed germ cell tumors, which have both
seminomatous and non-seminomatous elements, more strongly resemble non-
seminomas in these respects and have traditionally been classified with non-
seminomas.

3

Table 1: Specific Histologic Types of Germ cell Tumors Belonging
to the Broadly Defined Categories, Seminoma and Non-Seminoma
(ICD-0 code)

Seminomas
Seminoma, NOS (9061)
Seminoma, anaplastic (9062)
Spermatocytic seminoma/Spermatocytoma (9063)
Dysgerminoma (9060)
Germinoma/Germ Cell Tumor, NOS (9064)

Non Seminomas
Germ Cell Tumor, non-seminomatous (9065)
Embryonal Carcinoma, NOS (9070)
Yolk Sack (9071)
Teratoma, benign/Teratoma, NOS (9080)
Teratocarcinoma/Mixed embryonal carcinoma and teratoma (9081)
Malignant teratoma, undifferentiated (9082)
Malignant teratoma, intermediate (9083)
Dermoid cyst, NOS/teratoma with malignant transformation (9084)
Mixed germ cell Tumor (9085)
Choriocarcinoma, NOS (9100)
Choriocarcinoma combined with other germ cell elements (9101)

Whether seminomas and non-seminomas share common etiology is not known.
Differences in clinical behavior and peak age at occurrence have led to
speculation that their etiologies may differ. However, there are also
commonalities: the most established risk factors - a personal history of
undescended testis and a family history of testicular cancer- are associated with
both histologic types, and the two types have overlapping age-incidence curves.
Tumors of both types display some chromosomal losses and gains at common
locations (e.g., gain of X, 7, 8,12p and 21; loss of Y 4, 5, 1p, 11,13 and 18) and
some differences between seminomas and non-seminomas at chromosomes 7,
4

15,17, 19, and 22 (Oosterhuis 2005) and some recent evidence based on secular
trends point to a shared etiology (Bray 2006).

Men with a history of TGCT are reportedly 12 to 124 times more likely to develop
a second primary TGCT in the contralateral testicle than are unaffected men of
like race and age to develop a first TGCT (Fossa 2005). This remarkably
increased risk of a second primary tumor suggests that in a given man, the same
set of precipitating factors likely acts on both testes. Therefore, men with a
history of two primary TGCTs provide a unique opportunity to better understand
whether seminomas and non-seminomas are likely to share a common causality.
With this objective we studied men with a history of bilateral TGCT, examining
the distribution of histologic type between the two primary tumors diagnosed in
each man. We reasoned that if seminoma and non-seminoma have separate
etiologies, then the histology of a man’s first and second tumor would tend to be
the same – either both seminoma or both non-seminoma -- in accordance with
his history of either seminoma-related or non-seminoma-related risk factors. We
further reasoned that if, instead, common factors lead to both seminoma and
non-seminoma, histology of a man’s first TGCT would not predict histology of his
second TGCT.
Although previous studies have addressed other aspects of bilateral TGCT,
concordance of histologic types has not been investigated thoroughly, possibly
because of the limited numbers of bilateral cases enrolled in individual studies.
5

In the research presented here we assessed concordance of histologic type in a
large series of 550 men with a history of two primary TGCTs. We also compared
demographic characteristics of men diagnosed with bilateral TGCT with those of
men with a history of only one TGCT.

6

Chapter Two: Methods

Study subjects and methods

We defined bilateral testis cancer as a history of two primary TGCTs in a single
man. Subjects were men in the United States with bilateral TGCT, with first
diagnosis occurring as early as 1972 and second diagnosis occurring as late as
2006. We required that both diagnoses meet each of the following criteria: 1)
malignant primary tumor, 2) located in testis, and 3) scored as one of the
following histologic types, ICDO-3 Code,: seminoma, not otherwise specified
(NOS),9061; seminoma, anaplastic, 9062; spermatocytic seminoma, 9063; germ
cell tumor , non-seminomatous, 9065; embryonal carcinoma NOS, 9070; yolk sac
tumor, 9071; teratoma, malignant NOS,9080; teratocarcinoma, 9081; malignant
teratoma, undifferentiated, 9082; malignant teratoma, intermediate, 9083;
teratoma with malignant transformation,9084; mixed germ cell tumor, 9085;
choriocarcinoma, 9100; choriocarcinoma combined with other germ cell
elements, 9101. We also included testicular germ cell tumors scored as
dysgerminoma, 9060 or germinoma, 9064 although these designations were
devised for germ cell tumors of seminoma-like histologic appearance occurring at
extra-testicular sites. Specifically excluded from this definition were carcinoma in
situ (CIS), invasive testicular tumors of non-germ cell origin, and extra-gonadal
germ cell tumors.
7

Data on bilateral cases were obtained from three sources: 1) case listings of
bilateral TGCT provided by the California Cancer Registry (CCR), supplemented
by instances of bilateral TGCT self-reported by men with one TGCT identified by
the CCR, 2) data on men with bilateral TGCT identified by the Surveillance
Epidemiology and End Results (SEER) programs outside of California, and 3)
additional men with bilateral TGCT referred by clinicians and survivor groups.

The CCR is a network of 10 regional registries located throughout California
(year founded): Santa Clara Region (1988), Central California (1988),
Sacramento (1986), Tri-Counties (1988), Desert Sierra (1988), Northern
California (1987), Orange County (1983), San Francisco Bay Area Region
(1973), Los Angeles County (1972), and San Diego and Imperial Counties
(1988). By record matching, the CCR identified 177 men with a history of bilateral
TGCT. The CCR had also provided listings of men with one occurrence of TGCT
diagnosed in California. As part of an ongoing population-based study of TGCT,
202 of these men self-reported a history of two primary TGCTs. By matching on
name and date of birth we learned that 91 of these men had not been identified
by the record matching effort of the CCR, due to one TGCT diagnosis occurring
either outside of California or prior to inception of the reporting regional registry.

8

The SEER program is a network of 17 population-based cancer registries located
in various sites across the United States (year founded): Alaska Natives (1984),
Atlanta Metropolitan (1975), California excluding San Francisco/ San Jose
Metropolitan/ Los Angeles (1988), Connecticut (1973), Detroit Metropolitan
(1973), Hawaii (1973), Iowa (1973), Kentucky (1995), Los Angeles (1988),
Louisiana (1995), New Jersey (1979), New Mexico (1973), San Francisco-
Oakland SMSA (1973), San Jose-Monterey (1988), Seattle/ Puget Sound (1974),
and Utah (1973). By record matching within the SEER 17 data, 443 men with a
history of two primary TGCTs were identified. We excluded from this series all
132 cases identified by SEER registries within the state of California, thereby
avoiding duplicate inclusion of cases identified through the CCR, leaving 311 for
analysis.

A third set of men with a history of bilateral TGCT diagnosed outside California
were referred to the study by clinicians or in response to notices posted on
websites of survivor groups. For this referred group of 21 men, histology and
date of each of the two TGCT primary diagnoses was confirmed by medical
records including pathology reports. To avoid duplicate inclusion of men in this
set and SEER data, we eliminated from this set the one man who may have also
been identified by SEER, based on year of birth and years of diagnosis. Thus, a
set of 20 men referred to the study were included in the analysis.

9

A total of 550 men with a history of bilateral TGCT were included in the analysis
of concordance of histology of first and second primary TGCTs: among these
men, both diagnoses were identified ascertained by the CCR for 177, one
diagnosis was ascertained through CCR records and one diagnosis reported by
the participant for 42, both diagnoses were ascertained exclusively from SEER
records for 311, and both diagnoses were identified from referral sources for 20
men.

For each of these men both TGCT diagnoses were documented with histologic
type and date of diagnosis. This information was obtained entirely from registry
records for 488 men, but for 62 men it was abstracted from the original pathology
report for at least one diagnosis. For each of the 550 men the following data
were extracted from registry records or provided by self-report: race/ethnicity,
date of birth, and dates of first and second TGCT diagnoses. Dates were coded
as month, day and year for men identified through the CCR or by referral. Day or
month and day components of date were imputed for men identified exclusively
through SEER as follows: date of each diagnosis was imputed as the first day of
the month and year provided by SEER, and date of birth was imputed as June 1
of the year of birth provided by SEER. By the same methods we imputed missing
the day and month of diagnosis of two TGCTs and day only of one TGCT
diagnosed among men in the other sets.

10

For tumors not reported by a registry, diagnosis date was defined as calendar
date of orchiectomy recorded on the pathology report, except in two instances in
which the date recorded on the biopsy report was used and 15 in which date of
diagnosis of one TGCT was provided by the participant. For the 1018 tumors
reported by a registry, specific histologic type was abstracted as ICDO code from
registry records. For other 82 tumors, we assigned the ICDO code
corresponding to specific histology recorded on the pathology report.

We conducted a second analysis in order to compare characteristics of men with
bilateral TGCT to men diagnosed with one TGCT (“unilateral TGCT”). For this
purpose we used data from SEER 17 (2006), identifying individuals with either
one or two TGCTs as defined above. For each man included in this analysis we
extracted data on the following characteristics of each primary TGCT: registry
reporting the diagnosis, patient ID (assigned by reporting registry), race, NHIA
derived Hispanic Origin, year of birth, month of diagnoses, year of diagnoses,
and ICD-0 histology code. NHIA Derived Hispanic Origin is used by SEER to
assign Hispanic ethnicity using the NAACCR (North American Association of
Central Cancer Registries) Hispanic Identification Algorithm which uses a
combination of variables, including surname to directly or indirectly classify cases
as Hispanic for analytic purposes. In this manner 31,147 TGCTs were identified,
including multiple TGCTs occurring in the same individual (bilateral cases).
11

All participants were assigned a birth date of June 1st for their year of birth. For
the 39 cases in which month of diagnosis was unknown, June was assigned for
this variable. Day of diagnosis assigned as the first of day of the reported month
of diagnosis for all TGCTs. By matching on Patient ID we identified 443
individuals with a history of bilateral TGCT for the analysis, and treated the
remaining 30,704 as having a presumptive history of unilateral TGCT. We
included in the analysis 440 bilateral cases for whom at least one presumed
unilateral case matched on year of birth was identified in the same reporting
registry, among all unilateral cases matched on year of birth and reporting
registry, a total of 440 bilateral cases and 16,204 presumed unilateral cases were
included.

This study was approved by the Institutional Review Board at the University of
Southern California.

Definitions of variables

Traditional broad categories of histologic type were created by grouping TGCTs
on ICDO code as follows: seminoma (dysgerminoma, 9060; pure seminoma,
9061; anaplastic seminoma, 9062; spermatocytic seminoma, 9063; germinoma
9064), and non-seminoma/mixed GCT (germ cell tumor, non-seminomatous,
9065; embryonal carcinoma, 9070; yolk sac tumor, 9071; teratoma benign NOS,
12

9080; teratocarcinoma/mixed embryonal carcinoma and teratoma, 9081;
malignant teratoma undifferentiated, 9082; malignant teratoma intermediate,
9083; teratoma with somatic type malignancies, 9084; mixed germ cell tumor,
9085; choriocarcinoma NOS, 9100; choriocarcinoma with other germ cell
elements, 9101).

Age in years at each diagnosis was calculated by subtracting date of birth from
date of diagnosis (days/365.25). Interval between diagnoses was calculated by
subtracting date of first diagnosis from date of second diagnosis. Diagnoses were
defined as synchronous where the interval was less than 60 days. For analysis
purposes in instances of synchronous cases where the same date is given for
the left and right diagnoses the right is coded as the first TGCT. Mean and range
of ages at first and second diagnoses were reported, stratified on reporting
registry and histologic type. To maintain confidentiality and report statistically
stable estimates, these values are not reported in tabular form on strata
corresponding to 10 or fewer men.

Estimates of percentage of seminomatous and non-seminomatous TGCT

To calculate the percentage of men diagnosed with testis cancer we used the
SEERStat software to access SEER 17 data. We selected case listings for all
cancers meeting each of two criteria (1) site location was testis NOS, testis
13

descended, or testis undescended, and (2) histology code was one of the
following: seminoma specified NOS, 9061; seminoma anaplastic 9062;
spermatocytic seminoma, 9063; germ cell tumor non-seminomatous, 9065;
embryonal carcinoma NOS, 9070; yolk sac tumor, 9071; teratoma, malignant
NOS, 9080; teratocarcinoma, 9081; malignant teratoma undifferentiated, 9082;
malignant teratoma, intermediate, 9083; teratoma with malignant transformation,
9084; mixed germ cell tumor, 9085; choriocarcinoma, 9100; choriocarcinoma
combined with other germ cell elements, 9101; dysgerminoma, 9060 or
germinoma, 9064. Selected cases listings were then assigned to one of the
following categories based on age in years at diagnosis: under 14, 14-45, over
45. We calculated percentages of total for each of those groups.

Statistical Analyses

Analyses were performed using SAS 9.1 software (SAS Cary, NC). In the first
analysis we sought to learn among cases of bilateral TGCT whether the histology
of the second tumor can be predicted by the histology of the first tumor. Using a
dataset of only bilateral cases, N=550, we performed unconditional logistic
regression, regressing the broad histologic type of the second diagnosis
(seminoma or non-seminoma/mixed germ cell tumor) on broad histologic type of
the first diagnosis. We first estimated crude odds ratios, then addressed possible
effects of length of interval between diagnoses and ages at first and second
14

diagnoses by including these variables as covariates in subsequent regression
models.

In the second analysis we studied unilateral and bilateral TGCT cases, seeking
to learn if tumor histology (of first, or only tumor) could predict bilateral cases.
Since the incidence rates for the different histologic subtypes have been
changing over time, we controlled for birth cohort by matching bilateral cases to
unilateral cases on year of birth. We also matched on reporting registry because
registries differ widely in origin date, and those with earlier cases have greater
time to follow up to allow development of bilateral TGCT among unilateral cases.
Bilateral cases were frequency matched to unilateral cases by individual year of
birth and reporting registry within the SEER 17 group using m to n matching. We
assessed the relationship between histology and bilateral disease using
conditional logistic regression, stratifying on the frequency matched sets

In this set of analyses we first regressed the outcome, bilateral TGCT, on the
broad histologic type of the first (or only) diagnosis of TGCT using a model with
no other covariates. In subsequent analyses we used models including age at
only/first diagnosis, race, and ethnicity. All analyses were based on two-sided p
values.

15

Literature Review
We conducted a literature review to determine to identify earlier publications that
had addressed these questions. We queried Ovid and PubMed databases
including articles published between January 1, 1980 and April1st 2010 using
keywords "testis cancer” and “bilateral", and reviewed all retrieved publications to
learn whether the authors had attempted to address the association between
broad histologic type of first and second TCGT, or whether this association could
be calculated from the data presented in the report.

16

Chapter Three: Results

Estimation of distribution of TGCT by age groups

Using the SEERStat software to obtain case listing data we located 33,265 cases
of TGCT in the SEER 17 data. From those cases we found 276 cases diagnosed
under age 14, 27,602 between the ages of 14 and 45, and 5,387 cases
diagnosed over age 45. Thus 0.83 percent of all incident TGCTs in SEER 17
were diagnosed in boys under age 14, 83.0 percent among boys and men ages
14-45, and 1.01 percent among men over 45.

Predictive value of the histologic type of a first TGCT on the histologic type
of a second TGCT

A total of 550 men with a history of two histologically-confirmed primary TGCTs
were identified, 530 ascertained through a population-based cancer registry and
20 by referral. The broadly defined histologic types of first and second primary
TGCTs (seminomas and non-seminomas) of these men are presented in Table
2. Among the men with bilateral TGCT the study population included 483 non-
Hispanic Whites (88%), 45 Hispanic whites (8%), seven Pacific Islanders, (1%),
six Asians (1%),five African Americans (<1%), two Native Americans (<1%), and
two individuals of unknown race/ethnicity (<1%), (data not shown).
17

Mean ages at first and second TGCT diagnosis were 31.4 years and 36.0 years,
respectively, for all histologic subtypes combined. Seminomas comprised 60.6
percent of first primary and 70.6 percent of second primary TGCTs. As shown in
Table 2, these attributes were similar for subgroups of cases obtained by each
ascertainment method.

Table 2. 550 Men with History of Two Primary Testicular Germ Cell Tumors (TGCTs), by Method of
Identification, Broadly Categorized Tumor Histology, and Age at Diagnosis of each TGCT

N
First TGCT Diagnosis Second TCGT Diagnosis
Age in years
Mean (range)
Percent
Seminoma
Age in years
Mean (range)
Percent
Seminoma
Population-based cases 530 31.4 (14.3-71.7) 60.6 35.9 (17.5-79.1) 70.4
Both diagnoses identified through CCR
1
177 31.0 (15.0-52.2) 66.1 35.1 (17.5-55.8) 72.8
One diagnosis identified through CCR
1
,
one diagnosis self-reported and
confirmed by medical records
42 32.7 (20.8-71.7) 54.8 39.6 (23.6-75.4) 69.0
Both diagnoses identified through SEER
2

excluding of those identified through CCR
1

311 31.4 (14.3-71.7) 58.2 35.9 (17.6-79.1) 69.1
Additional cases
3
20 32.7 (18.9-51.3) 60.0 38.5 (26.0-53.4) 75.0
All Cases 550 31.4 (14.3-71.7) 60.6 36.0 (17.5-79.1) 70.6
1
The California Cancer Registry from 1988 onward, and all 10 California Regional Registries from the inception of each
through 2006.
2
All Registries participating in the Surveillance Epidemiology and End Results (SEER) 17 Program except California,
from the inception of each through 2006.
3
Referred by medical care provider or support group (not population-based)

18

19

Broadly defined histologic types of first and second TGCTs are presented in
Table 3. Men for whom the first tumor was a seminoma tended to be older at
the time of first diagnosis than those whose first tumor was a non-
seminoma/mixed GCT (mean age at first diagnosis 34.1 versus 27.3 years).
Within each of these groups, age at diagnosis of the second TGCT tended to be
older for those whose second TGCT was a seminoma (e.g., among those with a
first seminoma, average age at second diagnosis was 40.1 for the subgroup with
a seminomatous second TGCT versus 33.9 for the subgroup with a non-
seminomatous second TCGT; among men whose first tumor was non-
seminomatous, corresponding average ages at second diagnosis were 27.9
versus for those with a seminomatous second TGCT and 26.3 for non-
seminomatous second tumor, and for second diagnosis 33.2 for a seminomatous
second tumor and 30.2. for non-seminomatous second tumor). Appendix A-1
describes further differences in age at diagnosis for some subsets of men
defined by specific histologic type of each tumor. For example, age at first
diagnosis of anaplastic seminoma (9062) tended to be later than first occurrence
of other types of seminoma. However, these more detailed comparisons were
limited by infrequent occurrence of some combinations of specific histologic
types of first and second TGCTs.

20
Table 3. Broad Histologic Type and Age at Diagnosis of First and Second Tumors Among 550 Men
with History of Two Primary Testicular Germ Cell Tumors

First Diagnosis Testicular Germ Cell Tumor

Second Diagnosis Testicular Germ Cell
Tumor
Histologic Type
N
Synchronous
Diagnoses
N (percent)
Age in Years:
Mean (Range)
Histologic Type
N
Age in Years:
Mean (Range)
Seminoma (all types)
Followed by:
Seminoma
Non-Seminoma/Mixed GCT
333

247
86
97 (29.13)

76 (30.77)
21 (24.42)
34.1 (18.0-71.7)

35.5 (18.0-71.7)
30.1 (18.0-51.3)
Any

Seminoma
Non-Seminoma/
Mixed GCT
333

247
86
38.5 (19.2-79.1)

40.1 (22.2-79.1)
33.9 (19.2-52.6)
Non-Seminoma/Mixed GCT
Followed by:

Seminoma
Non-Seminoma/Mixed GCT
217

141
76
57 (26.27)

40 (28.37)
17 (22.37)
27.3 (14.3-58.4)

27.9 (16.5-58.4)
26.3 (14.3-53.6)
Any

Seminoma
Non-Seminoma/
Mixed GCT
217

141
76
32.2 17.5-58.4)

33.2 (19.5-58.4)
30.2 (17.5-53.6)
All First TCGT 550 154 (28.00) 31.4 (14.3-71.7) Any Second
TGCT
550 36.0, (17.5-79.1)

21

To learn whether bilateral cases are more likely to develop one broadly defined
histologic type in both testes, we calculated odds ratio estimates of the
association between a seminomatous first primary and seminomatous second
primary TGCT. In the univariate analysis in which we considered only histologic
type of each tumor, men for whom the initial TGCT was a seminoma were more
likely to also have seminoma as the histology of the second TGCT (OR=1.55;
95% CI=1.07-2.26) (Table 4a), indicating that histologic type of first and second
primary TGCTs are related. In subsequent analyses, we adjusted for age at
each TGCT diagnosis and length of the interval between diagnoses. Adjusting
for length of interval between diagnoses only, did not substantially alter the odds
ratio estimate (OR=1.59; 95% CI=1.09-2.31). However, adjusting for age at first
diagnosis eliminated any apparent association between histologic types of first
and second tumors (OR=0.98; 95% CI=0.68-1.48), as did adjusting for age at
second diagnosis (OR=0.98; 95% CI=0.65-1.47) or ages at both diagnoses and
interval (OR=0.95; 95% CI=0.63-1.45). Results did not materially differ when
adjusted for age at diagnosis and interval between diagnoses
simultaneously.(Table 4b)

22

Table 4a: 2x2 table for unadjusted Odds Ratio estimate of seminomatous
first diagnosis predicting a seminomatous second diagnosis
First Diagnosis Second Diagnosis Totals
Seminomatous Non
seminomatous

Seminomatous 247 86 333
Non
seminomatous
141 76 217
Totals 388 162 550
OR=1.55 95% CI =1.07-
2.24
Kappa=0.096 95% CI=0.014-
0.178

Table 4b: Estimated Odds Ratios for a seminomatous first diagnosis
predicting a seminatous second diagnosis by unconditional logistic
regression among all bilateral cases (N=550)
Controlling for: OR 95% CL
No other variables 1.55 1.07-2.26
Interval between diagnoses 1.59 1.09-2.31
Age at first diagnosis 0.98 0.68-1.48
Age at second diagnosis 0.98 0.65-1.47
Interval and age at first diagnosis 0.95 0.63-1.45
Interval and age at second diagnosis 0.95 0.63-1.45

Results were not materially different when we repeated these analyses in
subsets of the data from which we excluded either referred cases (N=530) (Table
4c), men with any diagnosis of spermatocytic seminoma (N=547), or those for
whom a presumed seminoma had been reported as germinoma or
dysgerminoma (N=533) (results not shown).

23

Table 4c: Estimated Odds Ratios for a seminomatous first diagnosis
predicting a seminatous second diagnosis by unconditional logistic
regression among population based bilateral cases (N=530)
Controlling for: OR 95% CL
No other variables 1.57 1.08-2.29
Interval between diagnoses 1.61 1.10-2.36
Age at first diagnosis 0.99 0.65-1.50
Age at second diagnosis 0.98 0.65-1.48
Interval and age at first diagnosis 0.95 0.62-1.46
Interval and age at second diagnosis 0.95 0.62-1.46

TGCTs were diagnosed synchronously (within 60 days of each other) for 154
men (28%), and asynchronously for the remaining 396. We sought to address
the question of whether synchronous cases of bilateral TGCT were more or less
likely than asynchronous cases to share a common etiology. To accomplish this
we conducted the analyses described above on the subsets consisting
exclusively of synchronous or asynchronous cases. We found very similar
patterns in each of these subgroups (Tables 4d and 4e). Among synchronous
cases, N=154, we estimated that the univariate OR of the of the second, or left,
TGCT being seminomatous if the first (or right) TGCT was seminomatous was
1.53 (95% CI=0.73-3.24), similar in magnitude to the estimate in the overall data,
but without achieving statistical significance. When covariates were added to the
model to control for the interval between diagnoses, less than 60 days, the OR
did not change perceptively. When covariates were added to the model to
address the age at first diagnosis and the age at second diagnosis both

24

individually and together the point estimate of the odds ratio approached 1 (e.g.,
OR= 1.17; 95% CI=0.53-2.56 in analyses adjusted for age at first diagnosis). In
the analysis of only asynchronous cases, N=396, the results closely matched
those of the analysis of synchronous and asynchronous cases combined, with
univariate analyses revealing a marginally significant association between
histology of first and second TGCTs (OR= 1.57; 95% CI 1.00-2.36) that
dissipated with adjustment for age at first diagnosis (OR= 0.90; 95% CI 0.55-
1.42).
Table 4d: Estimated Odds Ratios for a seminomatous first diagnosis
predicting a seminatous second diagnosis by unconditional logistic
regression among synchronous cases (N=154)
Controlling for: OR 95% CL
No other variables 1.53 0.73-3.24
Interval between diagnoses 1.54 0.73-3.25

Age at first diagnosis 1.17

0.53-2.56
Age at second diagnosis 1.17 0.53-2.56
Interval and age at first diagnosis 1.18 0.54-2.60

Interval and age at second diagnosis 1.18 0.54-2.60

Table 4e: Estimated Odds Ratios for a seminomatous first diagnosis
predicting a seminatous second diagnosis by unconditional logistic
regression among asynchronous cases (N=396)
Controlling for: OR 95% CL
No other variables 1.57 1.00-2.36
Interval between diagnoses 1.64 1.06-2.55
Age at first diagnosis 0.90 0.55-1.48
Age at second diagnosis 0.88 0.54-1.42
Interval and age at first diagnosis 0.91 0.55-1.52
Interval and age at second diagnosis 0.91 0.55-1.52

25

Comparison of men with bilateral TGCT to others with TGCT
Bilateral cases were matched to presumed unilateral cases on year of birth and
reporting registry, and after excluding three bilateral cases for whom a matched
control was not available, 16,204 presumed unilateral and 440 bilateral TGCT
cases were included in this analysis. Both unilateral and bilateral case groups
were predominantly non-Hispanic white, 82.0 and 88.9 percent, respectively
(Appendix A-2). Unilateral cases tended to be somewhat older at diagnosis,
mean (range) 33.8 (0.67-78.3) than bilateral cases at first diagnosis, 31.25
(14.33-65.75). This pattern held true within racial/ethnic group and strata defined
by broad histologic type (Figure 1). However, Hispanic whites tended to be
younger than non-Hispanic whites at the time of diagnosis: among those with
presumed unilateral diagnosis, mean (range) age was 31.6 (14.7-66.5) for
Hispanic whites and 34.1 (0.67-78.3) among their non-Hispanic white
counterparts, while among those with bilateral TGCT age a first diagnosis was
28.6 (18.0-44.7) among Hispanic whites and 31.5 (14.3-65.8) among non-
Hispanic whites.

26

Across registries, mean age at diagnosis for unilateral cases was consistent, as
was the mean age at first diagnosis for bilateral cases (Appendix A-3). The racial
makeup of the SEER registries was fairly consistent across registries that joined
SEER at different years, except for those that joined in 1988, which included all
of California except for San Francisco/Oakland. These California registries differ
from the others in having the highest percentage of Hispanic whites (30.8 in
unilateral cases and 21.4 in bilateral cases) than any of the other registry
groupings by year. In the California registries the percentage of non-Hispanic
0
5
10
15
20
25
30
35
40
Unilateral Bilateral
Mean age at diagnosis
Figure 1: Comparison of mean age at first/only
diagnosis of unilateral and bilateral cases by
broad histologic type
Seminomatous
Non-Seminomatous

27

whites was lower 62.0 for unilateral and 74.3 for bilateral cases (Appendix A- 3)
than in the other registries,.

Examining effects of broadly defined histologic type of first TGCT without
addressing additional covariates, we found little indication of an association
between histology of the first/only TGCT and risk of subsequently developing
bilateral TGCT. However, in analyses in which we adjusted for effects of age at
first/only TGCT diagnosis, we found risk of bilateral TGCT to be greater when
histology of the first/only TGCT was seminoma (OR=1.4: 95%CI=1.1-1.7) (Table
5a). Increasing age at first diagnosis, calculated in years, appeared to have a
slightly protective effect (OR= 0.93; 95%CI=0.92-0.95 per year increase in age at
diagnosis)(results not show), regardless of whether the racial/ethnic variables
were included in the analysis model. Seminoma as histology of the first TGCT
remained associated with risk of a second primary TGCT in analyses in which
effects of race (White versus other) and ethnicity (Hispanic versus other) were
included in the model (OR = 1.4; 95% CI=1.1-1.7). White race and non-Hispanic
ethnicity both appeared to be associated with developing a second TGCT
(OR=2.0; 95%CI=1.0-2.29; and OR= 1.5; 95%CI=1.2-3.3, respectively). To
address the possibility that bias may have resulted from shorter survival times
before usage of platinum based chemotherapies became common, we repeated
these analyses, in subsets of bilateral cases diagnosed either before 1980 or

28

before 1990. The restrictions did not materially change results. (Tables 5b and
5c).
Table 5a: Analysis of Unilateral cases and Bilateral cases matched by
year of birth and registry of diagnosis, estimated Odds Ratio of a
seminomatous first/only diagnosis predicting an outcome of bilateral
TGCT (N=16644)
Controlling for: OR 95% CL
No other variables
1.03 0.84-1.25
Age at first diagnosis 1.38 1.12-1.71
Age and white race 1.40 1.13-1.72
Age, white race, non-Hispanic ethnicity 1.40 1.14-1.73

Table 5b: Analysis of Unilateral cases and Bilateral cases matched by
year of birth and registry of diagnosis, estimated Odds Ratio of a
seminomatous first/only diagnosis predicting an outcome of bilateral
TGCT diagnosed after 1980 (N=15292)
Controlling for: OR 95% CL
No other variables
0.95 0.77-1.18
Age at first diagnosis
1.26 1.00-1.57
Age and white race
1.27 1.01-1.58
Age, white race, non-Hispanic ethnicity
1.27 1.01-1.59

Table 5c: Analysis of Unilateral cases and Bilateral cases matched by
year of birth and registry of diagnosis, estimated Odds Ratio of a
seminomatous first/only diagnosis predicting an outcome of bilateral
TGCT diagnosed after 1990 (N=11715)
Controlling for: OR 95% CL
No other variables
1.39 1.04-1.85
Age at first diagnosis
1.64 1.21-2.20
Age and white race
1.65 1.23-2.23
Age, white race, non-Hispanic ethnicity
1.67 1.24-2.23

29

To further address the association of a first/only seminomatous TGCT as a risk
factor for the development of bilateral TGCT we investigated the individual
histologic subtypes that compose the broadly defined seminomatous type. We
introduced into the model covariates that addressed heterogeneity for only pure
seminoma (9061), for pure seminomas (9061) with dysgerminoma (9060) and
germinoma (9064) -- because the latter two histologic types are presumed to be
previous coding conventions for pure seminoma -- and for anaplastic seminoma
(9062). Each of these terms was coded as dichotomous, each tumor was
labeled as one of the above histologic types or was not (i.e., any other histology)
in the new covariate. The term coding for pure seminoma (9061), dysgerminoma
(9064) and germinoma (9064) to address heterogeneity within the seminomatous
returns an OR= 1.03 (95% CI=0.84-1.26) while the term for other seminomas
returned OR=1.00 (95% CI=0.56-1.79), p=0.919 with no other adjustments With
adjustments for the age at first/only diagnosis for the narrow definition becomes
OR=1.39 (95% CI=1.13-1.72) with that for other seminomas OR=1.23 (95%
CI=0.69-2.20), p=0.688. With adjustment for age at first diagnosis, race and
ethnicity the narrowly defined term returns OR=1.41 (1.14-1.75) and other
seminomas OR=1.25 (95% CI=0.70-2.24), p=0.680, indicating no significant
heterogeneity within the seminomatous/non-seminomatous term when
germinomas, dysgerminomas and pure seminomas are grouped together and
compared to anaplasitc seminomas, spermatocytic seminomas (Table 6a).

30

Table 6a: Comparison of terms addressing heterogeneity within
broadly defined seminomatous TGCT as a risk factor for
development of bilateral TGCT, comparing pure seminoma, 9061,
dysgerminoma 9060, and germinoma 9064 to other seminomas
Adjusting for OR for pure seminoma,
9061, dysgerminoma 9060,
germinoma 9064
(95% CI)
OR for other
seminoma
(95% CI)
p value
none 1.03 (0.84-1.26) 1.00 (0.56-1.79) 0.919
Age at first diagnosis 1.39 (1.13-1.72) 1.23 (0.69-2.20) 0.668
Age, white 1.41 (1.14-1.74) 1.24 (0.69-2.22) 0.667
Age, white, non-
Hispanic
1.41 (1.14-1.75) 1.25 (0.70-2.24) 0.680

Pure seminoma, 9061, comprises the largest single histologic type in this
sample, 52.97 percent of all histologic types and 92.23 percent of seminomatous
types. The term to coding for pure seminoma (9061) to address heterogeneity
within the seminomatous group returns an OR=1.02 (95% CI= 0.84-1.25) while
the term for other seminomas OR=1.06 (95% CI= 0.68-1.65) returns with no
other adjustments p=0.890. With adjustments for the age at first/only diagnosis
for the narrow definition becomes OR=1.43 (95% CI=1.15-1.78) and the term for
other seminoma OR=1.09 (95% CI=0.670-1.70), p=0.227. With adjustment for
age at first diagnosis, race, and ethnicity the term adjusting heterogeneity the
term for pure seminoma OR=1.45 (95% CI=1.16-1.80) and the term for other
seminomas OR=1.12 (0.72-1.75), p=0.258, indicating no significant
heterogeneity within the seminomatous/non-seminomatous term when only
seminoma NOS (9061) was evaluated against other seminomas within the broad
histologic type. (Table 6b).

31

Table 6b: Comparison of terms addressing heterogeneity within
broadly defined seminomatous TGCT as a risk factor for development
of bilateral TGCT, comparing pure seminoma 9061, to other seminomas
Adjusting for OR for pure seminoma,
9061
(95% CI)
OR for other seminoma
(95% CI)
p
value
none 1.02 (0.84-1.25) 1.06 (0.68-1.65) 0.890
Age at first diagnosis 1.43 (1.15-1.78) 1.09 (0.70-1.70) 0.227
Age, white 1.44 (1.16-1.79) 1.11 (0.71-1.73) 0.239
Age, white, non-
Hispanic
1.45 (1.16-1.80) 1.12 (0.75-1.75) 0.258

The term to coding for anaplastic seminoma (9062) to address heterogeneity
within the seminomatous group also fails to demonstrate significant
heterogeneity from the broad histologic group . (Table 6c).

Table 6c: Comparison of terms addressing heterogeneity within
broadly defined seminomatous TGCT as a risk factor for development
of bilateral TGCT, comparing anaplastic seminoma 9062, to other
seminomas
Adjusting for OR for anaplastic
seminoma 9062
(95% CI)
OR for other
seminoma
(95% CI)
p value
none 1.06 (0.58-1.93) 1.03 (0.84-1.25) 0.924
Age at first diagnosis 1.24 (0.68-2.272) 1.39 (1.12-1.72) 0.706
Age, white 1.25 (0.68-2.295) 1.40 (1.14-1.74) 0.709
Age, white, non-
Hispanic
1.26 (0.69-2.314) 1.41 (1.14-1.74) 0.719

Literature Review

In our publication search we found 18 studies that reported on bilateral testis
cancer and provided information on the histologic types of both the first and
second diagnoses (Table 7). Thirteen of these studies provided data that

32

allowed us to calculate a crude OR of the association between broadly defined
histologic types of first and second TGCTs. None of these studies provided
sufficient information to adjust for age at diagnosis. Because these studies did
not always report the interval between diagnoses we could not reliably identify
synchronous diagnoses as defined by less than 60 days between, therefore we
only calculated OR for those cases identified in the published data as
asynchronous. From the data reported by Osterlind and colleagues (1991)
(N=68 asynchronous cases) we calculated a positive association which was
marginally statistically significant (OR=2.79; 95% CI=1.03-7.55). Miles et all
1985 reports only on synchronous cases "not an independent event", and from
their data we calculated an OR of 126.67 (95% CI=8.79-1489.54). From data
published by Fossa and colleagues (2005) (N= 287 asynchronous cases,
including many included in our own analysis) we calculated an OR=1.66; 95% CI
(1.013-2.734), which did archive marginal statistical significance; however, we
were not able to adjust for age at diagnosis in analyses of these published data.
The other five reports did provide sufficient information on the histologic types of
both first and second diagnoses needed calculate an OR.

33

Table 7: Previously published studies addressing histologic types in
bilateral TGCT
Study (first
author, year
published)
Number of
bilateral cases
Total
Asynchronous
Synchronous
Odds Ratio
(95%CI)
Relating first
seminoma to
second
seminoma
Reported, or
Calculated
by us?
Did
analysis
control for
age at
diagnosis
(yes/no)?
Overlap
with
other
studies
Albers 1999 30 total

Not evaluated unable no no
Bokemeyer,
1993
27 total
24 asynchronous
3 synchronous (all
discordant)
1.875
(0.372-9.451)
For
asynchronous
only
Calculated no no
Che 2002 24 total
20 asynchronous
4 synchronous
3 concordant
1 discordant

6.0
(0.885-39.124)
For
asynchronous
only
Calculated
Reported only
p value,

no no
Colls 1996 14 total
9 asynchronous
5 synchronous
3 concordant
2 discordant

0.286
0.009-5.704
For
asynchronous
only
Calculated
reported “not
concordant"
no no
Coogan
1998
20 total
15 asynchronous
5 synchronous
4 concordant
1 discordant
0.56
0.004-0.905
For
asynchronous
only
Calculated no no
Dieckmann
1986
(repeated
below)
9 total 12.0
0.51-280.10)
Calculated no Very
likely

Diekmann
1993
21 total
18 asynchronous
3 synch
1.80
0.334-19.674
Include both
Calculated no Very
likely
Fordham
1990
38 total
19 concordant
19 discordant
unable no no
Fossa
2005
462 Total
287 asynchronous
175 synchronous
1.664
(1.013-2.734)
For
Asynchronous
only
Calculated no no
Hellbardt
1990
3 total Report 100 %
concordance
unable no no

34

Table 7, continued
Kleinerman
1985
6 total Report “almost
all 2
nd
different
cell type from
1
st
”
Unable no no
Miles
1985
34 total
34 synchronous
126.67
8.787-1489.54

Calculated
Report:
Not
independent
event
no no
Osterlind
1991
73 total
68 asynchronous
5 synchronous
2.787
(1.028-7.548)
for
Asynchronous
only
Calculated no no
Patel 1990 16 total
12 asynchronous
4 synchronous
all concordant
3.00
0.177-39.65
for
Asynchronous
only
Calculated no no
Reinberg
1991
6 100 non
concordance in
6 bilateral GCT
Case report no no
Scheiber
1987
20 total
19 asynchronous
1 synchronous
1 discordant
1.875
0.321-10.969
asynchronous
only

Calculated
Reports “no
association”

no no
Thompson
1988
9 total
8 asynchronous
1 synchronous
1 discordant
2.0
0.132-32.645
Asynchronous
only
Calculated no no
Tekin
2000

11 total
7 asynchronous
4 synchronous
All concordant
1.5
0.093-25.586
for
Asynchronous
only
Calculated no no

Previous studies examining associations between histology type of first and
second primary TGCTs were based on numbers of participants that were not
adequate to reliably detect such an association, and did not address possible
effects of age at diagnosis, in spite of the strong dependence TGCT risk on age

35

Chapter Four: Discussion

Our analysis of data from the largest number of bilateral TGCT cases studied to
date found that after controlling for age at diagnosis, there is no association
between broad histologic subtype of first and primary TGCTs among bilateral
cases. Previous studies examining associations between histologic type of first
and second primary TGCTs were based on numbers of participants that were not
adequate to reliably detect such an association, and did not address possible
effects of age at diagnosis, in spite of the strong dependence TGCT risk on age.
Men with bilateral TGCT provide an opportunity to better understand whether
seminomas and non-seminomas are likely to share common causes, even when
the causes have not been identified and therefore cannot be measured directly.
Examination of the association between the histologic type of first and second
bilateral tumors provides important clues to address the question of shared
etiology in all TGCT. Elucidation of the factors leading to development of TGCT
and whether seminomas and non-seminomas share these factors is important to
the development of treatment and prognostic methods.

Our finding that the association between histology of the first and second tumors
is confounded by age at first diagnosis suggests that some age-related factor is
determinate in the distinction between the two broad histologic types of TGCT.
While it has already been reported in unilateral TGCT that there exists an age

36

disparity in seminomatous and non-seminomatous tumors, this is the first time
that it has been reported that age confounds the relationship between the
histologic types of first and second bilateral TGCT. This finding, along with the
existing knowledge of the age difference between the broad histologic types in
unilateral TGCT should direct future research towards the investigation of
biologic factors that are age related. First and foremost should be investigated
whether the age disparity is due to one histologic type, non-seminomas,
developing at a more rapid pace leading to earlier clinical presentation or they
begin the development to malignancy at a earlier age. Additionally research
should be conducted focusing on biologic factors that are known to vary with age,
such as the accumulation of chromosomal damage and changes in methylation
of the genome to hormonal factors.

Our analysis of unilateral and bilateral cases finds that the first or only TGCT
being a seminoma is a risk factor for the development of bilateral TGCT.
However, this association is also confounded by age at first diagnosis. Fossa et
al reported seminoma as a risk factor for development of bilateral TGCT, but the
investigators did not examine potential confounding effects of age at diagnosis.
This confounding by age could reflect a greater length at time at risk for those
men with a first diagnosis at a younger age, as both seminomatous and non-
seminomatous bilateral cases are younger at first diagnosis than unilateral
cases. This finding that seminomatous first tumors are a risk factor or the

37

development of bilateral TGCT suggests that clinicians should follow patients
with seminomas as closely as non-seminomatous cases for recurrence despite
the currently accepted view that non-seminomas are more aggressive.

While our data support a common etiology for both histologic subtypes, there are
some findings which are not consistent with this theory. There have been
differing secular trends reported for each subtypes. One study reported a
plateau of non-seminoma incidence rates while seminoma rates continue to
increase (Stang 2009) while another study found that that both groups have
experienced increasing rates(Bray 2006).

It remains to be discerned how those men who develop bilateral testis cancer
differ from those with unilateral testis cancer. We would further investigate
unilateral and bilateral cases of TGCT to compare exposures, therapies, obtain
tumor tissue, social factors etc to elucidate differences and perhaps predictive
factors in unilateral cases of TGCT which go on to become bilateral cases. In an
ideal situation a long term cohort study of unilateral cases, including samples of
DNA from peripheral blood and tumor tissues from the first diagnosis, would be
conducted to investigate age related changes to DNA, such as methylation in
order to further address questions regarding the etiologies of the broad histologic
type, although this may not be feasible because of the rarity of the disease.

38

References
1. Albers P, G oll A, Bierhoff E, Schoeneich G, M öller S. Clinical course and
histopathologic risk factor assessment in patients with bilateral testicular germ cell
tumors. Urology 1999;54:714-8.
2. Bokemeyer C, Schmoll H, Sch offski P, Harstrick A, Bading M, Poliwoda H. Bilateral
testicular tumours: prevalence and clinical implications. Eur J Cancer 1993;29A:874-6.
3. Bray F, Ferlay J, Devesa S, McGlynn K, M öller H. Interpreting the international trends
in testicular seminoma and nonseminoma incidence. Nat Clin Pract Urol 2006;3:532-43.
4. Che M, Tamboli P, Ro J, Park D, Ro J, Amato R, Ayala A. Bilateral testicular germ
cell tumors: twenty-year experience at M. D. Anderson Cancer Center. Cancer
2002;95:1228-33.
5. Colls B, Harvey V, Skelton L, Thompson P, Frampton C. Bilateral germ cell testicular
tumors in New Zealand: experience in Auckland and Christchurch 1978-1994. J Clin
Oncol 1996;14:2061-5.
6. Coogan C, Foster R, Simmons G, Tognoni P, Roth B, Donohue J. Bilateral testicular
tumors: management and outcome in 21 patients. Cancer 1998;83:547-52.
7. Cortessis VK. Epidemiologic Insights into the Occurrence and Causes of Testicular
Cancer in 2003. In: D R. American Cancer Society Atlas of Clinical Oncology, Germ Cell
Tumorsed. London: BC Decker, 2003:16-29.
8. Dieckmann K, Boeckmann W, Brosig W, Jonas D, Bauer H. Bilateral testicular germ
cell tumors. Report of nine cases and review of the literature. Cancer 1986;57:1254-8.
9. Dieckmann K, Loy V, B ottner P. Prevalence of bilateral testicular germ cell tumours
and early detection based on contralateral testicular intra-epithelial neoplasia. Br J Urol
1993;71:340-5.
10. Fordham M, Mason M, Blackmore C, Hendry W, Horwich A. Management of the
contralateral testis in patients with testicular germ cell cancer. Br J Urol 1990;65:290-3.
11. Fossa S, Chen J, Schonfeld S, McGlynn K, McMaster M, Gail M, Travis L. Risk of
contralateral testicular cancer: a population-based study of 29,515 U.S. men. J Natl
Cancer Inst 2005;97:1056-66.

39

12. Garner M, Turner M, Ghadirian P, Krewski D. Epidemiology of testicular cancer: an
overview. Int J Cancer 2005;116:331-9.
13. Hellbardt A, Mirimanoff R, Obradovic M, Mermillod B, Paunier J. The risk of second
cancer (SC) in patients treated for testicular seminoma. Int J Radiat Oncol Biol Phys
1990;18:1327-31.
14. Horwich A, Shipley J, Huddart R. Testicular germ-cell cancer. Lancet 2006;367:754-
65.
15. Kleinerman R, Liebermann J, Li F. Second cancer following cancer of the male
genital system in Connecticut, 1935-82. Natl Cancer Inst Monogr 1985;68:139-47.
16. Looijenga L, Gillis A, Stoop H, Hersmus R, Oosterhuis J. Chromosomes and
expression in human testicular germ-cell tumors: insight into their cell of origin and
pathogenesis. Ann N Y Acad Sci 2007;1120:187-214.
17. Miles B, Kiesling VJ, Belville W. Bilateral synchronous germ cell tumors. J Urol
1985;133:679-80.
18. Nichols C. In: D R. American Cancer Society Atlas of Clinical Oncology, Germ Cell
Tumorsed. London: BC Decker, 2003: 86-92
19. Oosterhuis J, Looijenga L. Testicular germ-cell tumours in a broader perspective.
Nat Rev Cancer 2005;5:210-22.
20. Osterlind A, Berthelsen J, Abildgaard N, Hansen S, Hjalgrim H, Johansen B, Munck-
Hansen J, Rasmussen L. Risk of bilateral testicular germ cell cancer in Denmark: 1960-
1984. J Natl Cancer Inst 1991;83:1391-5.
21. Patel S, Richardson R, Kvols L. Synchronous and metachronous bilateral testicular
tumors. Mayo Clinic experience. Cancer 1990;65:1-4.
22. Reinberg Y, Manivel J, Zhang G, Reddy P. Synchronous bilateral testicular germ
cell tumors of different histologic type. Pathogenetic and practical implications of
bilaterality in testicular germ cell tumors. Cancer 1991;68:1082-5.
23. Ries LAG, Melbert D, Krapcho M, Stinchcomb DG, Howlader N, Horner MJ, Mariotto
A, Miller BA, Feuer EJ, Altekruse SF, Lewis DR, Clegg L, Eisner MP, Reichman M,
Edwards BK (eds). SEER Cancer Statistics Review, 1975-2005, National Cancer
Institute. Bethesda, MD, http://seer.cancer.gov/csr/ 1975_2005/, based on November
2008 SEER data submission, posted to the SEER web site, 2009.

40

24. Scheiber K, Ackermann D, Studer U. Bilateral testicular germ cell tumors: a report of
20 cases. J Urol 1987;138:73-6.
25. Skakkebaek N, Rajpert-De Meyts E, Main K. Testicular dysgenesis syndrome: an
increasingly common developmental disorder with environmental aspects. Hum Reprod
2001;16:972-8.
26. Stang A. Etiology of testicular germ cell tumors: lumping or splitting? A plea against
lumping. Eur J Epidemiol 2009;24:65-7.
27. Tekin A, Aygun Y, Aki F, Ozen H. Bilateral germ cell cancer of the testis: a report of
11 patients with a long-term follow-up. BJU Int 2000;85:864-8.
28. Thompson J, Williams C, Whitehouse J, Mead G. Bilateral testicular germ cell
tumours: an increasing incidence and prevention by chemotherapy. Br J Urol
1988;62:374-6.
29. Walsh T, Croughan M, Schembri M, Chan J, Turek P. Increased risk of testicular
germ cell cancer among infertile men. Arch Intern Med 2009;169:351-6.
30. Zheng T, Holford T, Ma Z, Ward B, Flannery J, Boyle P. Continuing increase in
incidence of germ-cell testis cancer in young adults: experience from Connecticut, USA,
1935-1992. Int J Cancer 1996;65:723-9.

41
Appendices

Table A-1. Specific Histologic Type and Age at Diagnosis of First and Second Tumors Among 550 Men with History of Two Primary
Testicular Germ Cell Tumors
First Diagnosis Testicular Germ Cell Tumor Second Diagnosis Testicular Germ Cell Tumor
Histologic Type
N
Synchronous
Diagnoses
N (percent)
Age in years: Mean
(Range)
Histologic Type
N
Age in years:
Mean (Range)
Seminoma
1
316 34.0 (18.0-67.5) Seminoma
2
229 39.9 (22.2-79.1)
Anaplastic Seminoma (9062) 5 *
Spermatocytic Seminoma (9063) 1 *
Embryonal Carcinoma (9070) 32 35.0 (21.3-49.6)
Yolk Sac Tumor (9071) 1 *
Teratoma benign NOS (9080) 4 *
Teratocarcinoma (9081) 7 *
Malignant teratoma undifferentiated
(9082)
1 *
Mixed Germ Cell Tumor

(9085) 34 34.6 (19.2-52.6)
Choriocarcinoma, NOS (9100) 1 *
Choriocarcinoma combined with other
germ cell elements (9101)
1 *
Anaplastic Seminoma (9062) 15 37.6 (21.8-71.7) Pure Seminoma (9061) 7 *
Anaplastic Seminoma (9062) 3 *
Embryonal Carcinoma (9070) 1 *
Yolk Sac Tumor (9071) 1 *
Teratocarcinoma (9081) 1 *
Malignant teratoma undifferentiated
(9082)
1 *
Mixed Germ Cell Tumor

(9085) 1 *
Spermacytic Seminoma (9063) 2 * Pure Seminoma (9061) 2 *

42

Appendix A-1, continued
Germ Cell Tumor, non-
seminomatous (9065)
1 * Pure Seminoma (9061) 1 *
Embryonal Carcinoma, NOS
(9070)
78 28.2 (17.2-44.9) Seminoma
3
47 34.4 (19.5-46.7)
Anaplastic Seminoma (9062) 1 *
Embryonal Carcinoma (9070) 11 31.6 (23.7-39.9)
Teratoma benign NOS (9080) 3 *
Teratocarcinoma (9081) 5 *
Mixed Germ Cell Tumor

(9085) 10 31.9 (22.1-40.0)
Choriocarcinoma combined with other
germ cell elements (9101)
1 *
Yolk Sac Tumor (9071) 4 * Pure Seminoma (9061) 3 *
Yolk Sac Tumor (9071) 1 *
Teratoma, benign NOS (9080) 18 26.4 (15.0-41.4) Seminoma
4
12 33.4 (21.3-54.4)
Embryonal Carcinoma (9070) 2 *
Teratoma benign NOS (9080) 1 *
Mixed Germ Cell Tumor

(9085) 2 *
Choriocarcinoma combined with other
germ cell elements (9101)
1 *
Teratocarcinoma/Mixed
embryonal carcinoma and
teratoma (9081)
48 25.8 (14.3-58.4) Pure Seminoma (9061) 31 33.4 (22.3-58.4)
Embryonal Carcinoma (9070) 8 *
Teratocarcinoma (9081) 5 *
Mixed Germ Cell Tumor

(9085) 3 *
Choriocarcinoma, NOS (9100) 1 *

43

Appendix A-1, continued
Malignant teratoma,
intermediate (9083)
1 * Embryonal Carcinoma (9070) 1 *
Teratoma with somatic type
malignancies (9084)
2 * Pure Seminoma (9061) 2 *
Mixed Germ Cell Tumor

(9085) 54 27.8 (18.4-40.6) Pure Seminoma (9061) 36 32.2 (20.4-48.4)
Embryonal Carcinoma (9070) 3 *
Teratoma benign NOS (9080) 1 *
Teratocarcinoma (9081) 1 *
Mixed Germ Cell Tumor

(9085) 10 32.3 (20.6-41.6)
Choriocarcinoma combined with other
germ cell elements (9101)
1 *
Choriocarcinoma combined with
other germ cell elements (9101)
11 26.3 (18.3-39.8) Pure Seminoma (9061) 8 *
Embryonal Carcinoma (9070) 1 *
Teratoma benign NOS (9080) 1 *
Choriocarcinoma combined with other
germ cell elements (9101)
1 *
1
Includes 304 reported as Pure Seminoma (9061), six reported as Germinoma (9064) and six reported as Dysgerminoma (9060)
2
Includes 225 reported as Pure Seminoma (9061) and four reported as Germinoma (9064)
3
Includes 46 reported as Pure Seminoma (9061) and one reported as Germinoma (9064)
4
Includes 11reported as Pure Seminoma (9061) and one reported as Dysgerminoma (9060)
* not reported for cells containing fewer than 10 observations

44
Appendix A-2: Comparison of Unilateral and Bilateral cases by Race and Ethnicity.
Unilateral N=16204 Bilateral N=440
Race Number
(percent)
Mean (Range)
age at only
diagnosis in
years
Percent
Seminomato
us at
diagnosis
Number
(percent)
Mean (Range) age
at first dx in years
Percent
Seminomatous
at first
diagnosis
White 15155
(93.53 )
Total
33.8 (0.67-78.3)
5
Seminomatous
N=8665, 36.2 (12.7-78.3)
Non Seminomatous
N=6490, 30.52 (0.67-70.9)
57.18 422 (95.91
)
Total
31.3, (14.3-65.8)
Seminomatous
N=245, 33.9 (21.3-65.8)
Non Seminomatous
N=177, 27.7 (14.3-58.3)

58.06
Asian

445
1
(2.75)
Total
34.1 (1.59-57.3)
2
Seminomas
N=288, 35.71 (17.8-57.3)
Non Seminomas
N=157, 31.3 (1.59-53.2)
64.72 12
3
(2.73) 29.0, (18.0-40.4)
Seminomatous
N=8, 30.1 (18.0-40.4)
Non Seminomatous
N=4, 26.8 (19.1-31.8)

66.67
Black 352
(2.17)
Total
33.8 (0.83-58.6)
3
Seminomas
N=231, 35.9 (19.1-58.6)
Non Seminomas
N=121, 29.8 (0.83-46.1)
65.63 2 (0.45) * 50.00
American
Indian/Alaska
Native
64
(0.39)
Total
32.5 (16.4-50.8)
Seminomas
N=39, 34.5 (17.6-50.8)
Non Seminomas
N=2,5 29.2(16.4-43.3)

60.94 1 (0.23) * *

45
Appendix A-2, continued
Other 28 (0.17) Total
30.6 (2.08-54.8)
Seminomas
N=19, 33.9 (22.4-54.8)
Non Seminomas
N=9, 23.6 (2.08-32.3)
67.86 1 (0.23) * *
Unknown 160 (0.99) Total
34.7 (19.9-59.0)
Seminomas
N=90, 37.1 (22.2-59.0)
Non Seminomas
N=70, 31.7(19.9-51.5)

56.25 2 (0.45) * 100.
Total 16204
9
33.8, (0.67-78.3)
8
Seminomas
N=9332, 36.2 (12.7-78.3)
Non Seminomas
N=6872, 30.5 (0.67-70.9)
57.59 440 Total
31.3, (14.3-65.8)
Seminomatous
N=257, 33.8(18.0-65.8)
Non Seminomatous
N=183, 27.7 (14.3-58.3)

58.41
Ethnicity
Non-
Hispanic
White
13289
(82.0 )
34.1, (0.67-78.3)
5
Seminomas
N=7646, 36.5 (12.7-78.3)
Non Seminomas
N=5643, 30.8 (0.67-70.9)
57.54 391 (88.9 ) Total
31.5, (14.3-65.8)
Seminomatous
N=225, 34.2 (21.3-65.8)
Non Seminomatous
N=166, 27.8 (14.3-58.3)

57.54
Spanish/
Hispanic/
Latino
derived
White
1866 (11.5) 31.6, (14.7-66.5)
Seminomas
N=1019, 34.1 (14.8-66.5)
Non Seminomas
N=847, 28.6 (14.7-54.9)

54.6 31 (7.05) Total
28.6, (18.0-44.7)
Seminomatous
N=20, 30.0 (28.8-42.6)
Non Seminomatous
N=11, 26.1(18.0-44.7)

64.52

46
Appendix A-2, continued
Spanish/
Hispanic/
Latino
derived non-
white
39
6
32.7, (20.3-50.1)
Seminomas
N=26, 34.6 (22.2-50.1)
Non Seminomas
N=13, 29.0 (20.3-44.0

66.67 1
7
* *
Total 16204
9
33.8, (0.67-78.3)
8
Seminomas
N=9332, 36.2 (12.7-78.3)
Non Seminomas
N=6872, 30.5 (0.67-70.9)
57.59 440 Total
31.3, (14.3-65.8)
Seminomatous
N=225, 34.2 (21.3-65.8)
Non Seminomatous
N=166, 27.8 (14.3-58.3)

58.41
1
includes 31 Asian Indian/Pakistani, 1 Chamoran, 101 Chinese, 63 Filipino, 1 Guamanian, 44 Hawaiian, 106 Japanese, 1
Kampuchean, 14 Korean, 3 Laotian, 3 Micronesian, 1 New Guinean, 50 Other Asian, 3 Pacific Islanders NOS, 1
Polynesian, 2 Samoan, 1 Thai, 1 Tongan, 18 Vietnamese.
2
if adjusted to include cases only greater than 2 years of age at diagnosis N=444, mean, median (range) 34.2,33.9 (15.6-
57.3)
3
includes 3 Chinese, 2 Filipino, 4 Hawaiian, 2 Japanese, 1 Other Asian
4
if adjusted to include cases only greater than 2 years of age at diagnosis N=351, mean, median (range) 33.9, 33.6 (2.1-
58.6)
5
if adjusted to include cases only greater than 2 years of age at diagnosis N=13282, mean, median (range) 34.1, 33.7
(2.00-78.3)
6
includes 5 American Indian/Alaska Natives, 12 Blacks, 3 Asians, 3 Other, and 16 Unknowns.
7
1 Asian
8
if adjusted to include cases only greater than 2 years of age at diagnosis N=16195, mean, median (range) 33.8, 33.3
(2.00-78.3)
9
Does not include 39 the above Spanish/Hispanic/Latino of races other than white because already list in the above races
* Data omitted do to small numbers N< 10.

47
Appendix A-3: Comparison of Unilateral and Bilateral Cases by year of SEER registry origin
All
Cases
Unilateral Cases Bilateral Cases
Mean,
Median,
Range
age at
only
diagnosis
in year
Percent
Seminomat-
ous
Diagnosis
Race : N (percentage) Mean,
Median,
Range age
at first
diagnosis
in years
Percent
Seminomat
-ous at first
diagnosis
Race : N (percentage)
SEER
Registries
founded in
1973
1
N=8627
33.46,
32.92
(0.671-
78.25)
N=8365
57.06 Non-Hispanic White 7333 (87.66)
Hispanic White 480 (5.74)
Asian 262 (3.13)
Black 175 (2.09)
Unknown 65 (0.78)
American Indian/AK 30 (0.36)
Other 20 (0.24)
30.69, 29.04
(14.33-65.75)
N=262
53.82 Non-Hispanic White 237
(90.46)
Asian 12
(4.58)
Hispanic White 11
(4.20)
Black 1
(0.38)
Other 1
(0.38)

SEER
Registries
founded in
1974,
1975 and
1979
2
N=3706

34.3, 33.8
(1.33-67.1)
N=3609
58.71 Non-Hispanic White 3280 (90.88)
Hispanic White 140 (3.88)
Black 97 (2.69)
Asian 49 (1.36)
American Indian/AK Native 19 (0.53)
Unknown 19 (0.53)
Other 5 (0.14)
31.6, 30.4
(17.2-51.2)
N=97
65.98 Non-Hispanic White 90
(93.81)
Hispanic White 5
(5.15)
Unknown 1
(1.03)

48
Appendix A-3, continued
SEER
Registries
Founded in
1988
3
N=4081
33.8, 33.7
(14.7-
57.3)
N=4011
57.49 Non-Hispanic White 2486 (61.98)
Hispanic White 1237 (30.84)
Asian 134 (3.34)
Unknown 70 (1.75)
Black 66 (1.65)
American Indian/AK Native 15 (0.37)
Other 3 (0.07)
32.6, 31.9
(16.9-50.3)
N=70
62.86 Non-Hispanic White 52
(74.29)
Hispanic White 15
(21.43)
American Indian/AK Native 1
(1.43)
Black 1
(1.43)
Unknown 1
(1.43)
SEER
Registries
Founded in
1995
4
N=230
34.6, 33.0
(23.7-
52.7)
N=219
61.19 Non-Hispanic White 190 (86.76)
Black 14 (6.39)
Hispanic White 9 (4.11)
Unknown 6 (2.74)
33.4, 31.8
(25.8-49.9)
N=11
72.73 Non-Hispanic White 11
(100.00)
1
Registries include Connecticut, Detroit (Metropolitan), Hawaii, Iowa, New Mexico, San Francisco-Oakland SMSA, and Utah
2
Registries include Seattle (Puget Sound) 1974, Atlanta (Metropolitan) 1975, and New Jersey 1979
3
Registries include California excluding SF/SJM/LA, Los Angeles, and San Jose-Monterey
4
Registries include Kentucky and Louisiana

Abstract (if available)

Abstract Testicular germ cell tumors (TGCT) are the most commonly diagnosed neoplasm of young men 15-45 years of age, and incidence of these tumors has been increasing in recent decades for reasons currently unknown. Histologic type of TGCT can be divided into the broad categories of seminomas and non-seminomas. It currently remains to be elucidated if these histologic types share a common etiology. In this study we analyzed the largest set of bilateral cases of TGCT ever assembled, 550, to determine if the histologic type of the first tumor predicts that of the second. Presuming an individual with a history of two primary tumors has the same risk factors for both tumors, a concordance of histologic types in the same individual would suggest that the risk factors for seminomas and non-seminomas differ. The data show no association between the histologic type of the first and second tumors when the analysis is adjusted for the age at first diagnosis (odds ratio (OR)=0.95) This finding suggests that the two broad histologic types share a common etiology, but some age related factor impacts differentiation between the two. We also compared 16,204 presumed unilateral cases of TGCT to 440 bilateral cases matched on year of birth and region of diagnosis to determine if histologic type of the first/only TGCT confers risk of developing a second TCGT. This analysis suggests that a first diagnosis of seminoma is a risk factor for developing testicular cancer after adjustment for age at first or only diagnosis (OR=1.401

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Trends of testicular cancer within the United States by histological subtype amongst whites and blacks, 1973-2003

PDF

Risk factors of testicular germ cell tumor

PDF

The role of steroid hormones in the etiology of urologic diseases

PDF

The study of germ cell tumors and related conditions: an analysis of self-reported data with characterization and comparison of Family History Questionnaires respondents and non-respondents

PDF

Incidence and survival rates of the three major histologies of renal cell carcinoma

PDF

A comparison of three different sources of data in assessing the adolescent and young adults cancer survivors

PDF

Genes and environment in prostate cancer risk and prognosis

PDF

Association of comorbidity with prostate cancer tumor characteristics in African American men

PDF

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

PDF

The development of type 2 diabetes mellitus (type 2 DM) in California twins

PDF

Origins of the gender disparity in bladder cancer risk: a SEER analysis

PDF

Cardiovascular disease risk factors and cognitive function

PDF

Carcinogenic exposures in racial/ethnic groups

PDF

The effects of tobacco exposure on hormone levels and breast cancer risk among young women

PDF

Survival of children and adolescents with low-risk non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) treated with surgery only: an analysis of 234 patients from the Children’s Oncology Group stud...

PDF

Effects of stress and the social environment on childhood asthma in the children' s health study

PDF

The effect of inhibiting the NFkB pathway on myeloid derived suppressor cell cytokine production

PDF

Genetic and environmental risk factors for childhood cancer

PDF

Genomic risk factors associated with Ewing Sarcoma susceptibility

PDF

An assessment of necrosis grading in childhood osteosarcoma: the effect of initial treatment on prognostic significance

Asset Metadata

Creator Thomas, Jennifer Phay (author)

Core Title Bilateral testicular germ cell tumors: distribution of histologic types

School Keck School of Medicine

Degree Master of Science

Degree Program Epidemiology

Publication Date 08/09/2010

Defense Date 07/01/2010

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

Tag bilateral,cancer,Epidemiology,histologic types,OAI-PMH Harvest,Testis

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cortessis, Victoria K. (committee chair), Hamilton, Ann S. (committee member), Siegmund, Kimberly D. (committee member)

Creator Email jennifer.thomas@med.usc.edu,jennphay@hotmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3352

Unique identifier UC1502804

Identifier etd-Thomas-3484 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-378608 (legacy record id),usctheses-m3352 (legacy record id)

Legacy Identifier etd-Thomas-3484.pdf

Dmrecord 378608

Document Type Thesis

Rights Thomas, Jennifer Phay

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu