University of Southern California Dissertations and Theses

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...

(USC Thesis Other)

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

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

1

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 study ARST0332

by

Qingyuan Wu

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

May 2016

2
Table of Contents

Acknowledgements ......................................................................................................................... 3
List of Figures ................................................................................................................................. 4
List of Tables .................................................................................................................................. 5
Abbreviations .................................................................................................................................. 6
Abstract ........................................................................................................................................... 7
Chapter 1: Introduction ................................................................................................................... 8
Chapter 2: Methods ....................................................................................................................... 10
Chapter 3: Results ......................................................................................................................... 16
Baseline patient characteristics of the analytic and comparison group .................................... 16
Characteristics of primary tumors for patients in analytic group ............................................. 19
Description of follow-up time for survivors in analytic and comparison groups ..................... 21
Comparison of survival between the two groups ...................................................................... 22
Analysis of potential prognostic factors for the analytic group ................................................ 27
Chapter 4: Discussion ................................................................................................................... 30
Reference ...................................................................................................................................... 32

3
Acknowledgements

I’m deeply grateful to my mentor Dr. Don Barkauskas for being incredibly instructive
and providing me with enormous help throughout the course of this project.
I would also like to thank my committee members Dr. Wendy Mack and Dr. Mark Krailo
for their expertise and guidance.
Additionally, I would like to thank Children's Oncology Group and Dr. Sheri Spunt for
providing such great project and high-quality data.
4
List of Figures

Figure 1 Experimental Design Schema ......................................................................................... 11
Figure 2 Treatment Regimens ....................................................................................................... 12
Figure 3 Distribution of Follow-up Time for Survivors ............................................................... 21
Figure 4 Comparison of Product-Limit Survival Estimates for Event-free Survival (with number
of subjects at risk) ................................................................................................................. 23
Figure 5 Hazard Functions for Event-free Survival ...................................................................... 24
Figure 6 Comparison of Product-Limit Survival Estimates for Overall Survival (with number of
subjects at risk) ..................................................................................................................... 25
Figure 7 Hazard Functions for Overall Survival ........................................................................... 26

5
List of Tables

Table 1 Baseline Patient Characteristics of Analytic and Comparison Group ............................. 17
Table 2 Histological Subtypes of Soft Tissue Sarcoma ................................................................ 19
Table 3 Primary tumor site ........................................................................................................... 20
Table 4 Follow-up Time for Survivors ......................................................................................... 21
Table 5 Critical Events .................................................................................................................. 22
Table 6 Survival Time .................................................................................................................. 22
Table 7 Univariate Prognostic Factor of Event-free Survival and Overall Survival for the
Analytic Group (n=82) .......................................................................................................... 28
Table 8 Multivariate Cox Model for Event-free Survival in the Analytic Group (n=82) ............. 29

6
Abbreviations

COG: Children's Oncology Group
EFS: event-free survival
FNCLCC: French Federation of Cancer Centers Sarcoma Group
NRSTS: nonrhabdomyosarcomatous soft tissue sarcomas
OS: overall survival
POG: Pediatric Oncology Group
RMS: rhabdomyosarcoma
RT: radiotherapy
STS: soft tissue sarcoma

7
Abstract

Background: Nonrhabdomyosarcomatous soft tissue sarcomas (NRSTS) are both a rare and
heterogeneous group of malignancies among children and adolescents, with an incidence of
around 500 cases per year comprising approximately 5% of childhood malignancies in the
United States in those <20 years of age. While radiation is widely used as an adjuvant treatment
after surgical removal of the tumor in prevention of recurrence, it can also cause long-term
complications.

Methods: ARST0332 was a groupwide phase III study with a primary aim to evaluate a novel
risk-based strategy for treating pediatric NRSTS and dual goals of limiting the toxicity of therapy
for low-risk patients and maximizing the efficacy of therapy for intermediate- and high-risk
patients. ARST0332 enrolled 551 eligible patients and assigned them to 4 treatment arms based
on risk: Arm A (observation), Arm B (adjuvant radiotherapy), Arm C (adjuvant chemotherapy
and radiotherapy) and Arm D (neoadjuvant chemoradiotherapy).
This thesis focused on whether patients <30 years of age with a ≤5 cm high-grade soft tissue
sarcoma could be cured with surgery alone without the adjuvant radiotherapy, so as to avoid the
consequences of radiation. The analytical group was 82 patients with ≤5 cm high-grade tumors
enrolled on ARST0332 Arm A treated with surgery only (all classified as low-risk), compared
with the remaining 152 patients identified as low-risk (non-metastatic, grossly resected tumors).
Endpoints included event-free survival (EFS) and overall survival (OS).

Results: The analytic group had a slightly higher proportion of patients experiencing EFS events
(including relapse, progression and second malignant neoplasm) and death than the comparison
group, although log-rank tests for comparison of both EFS and OS were not significant (log-rank
p=0.08 and 0.32, respectively). Depth of tumor was found to be a statistically significant
prognostic factor for event-free survival (p=0.029), with a hazard ratio of deep vs. superficial of
9.65(95%CI: 1.26-74.06), after adjusting for age and gender. No significantly factor was related
to overall survival.

Conclusion: Patients with high-grade small tumors treated with surgery only did not have worse
survival outcome than other low-risk patients treated with adjuvant radiotherapy.
8
Chapter 1: Introduction

Rhabdomyosarcoma (RMS), a tumor of striated muscle, is by far the most common soft
tissue sarcoma (STS) in children aged 0 to 14 years, accounting for around half of these tumors
in this age range [1]. In pediatrics, the remaining soft tissue sarcomas are commonly referred to
as nonrhabdomyosarcomatous soft tissue sarcomas (NRSTS), encompassing over 35 histologies
with a wide spectrum of local aggressiveness and metastatic potential [2].
The NRSTS are both a rare and heterogeneous group of malignancies among children
and adolescents, with an incidence of around 500 cases per year comprising approximately 5%
of childhood malignancies in the United States in those <20 years of age [1, 3]. Rareness and
heterogeneity have made NRSTS a difficult group of tumors to study in young populations. Only
three prospective clinical trials with fewer than 200 total patient enrollments were conducted in
the U.S. [4-6] prior to the conduct of ARST0332.
ARST0332 study’s primary aim was to evaluate a novel risk-based strategy for treating
pediatric NRSTS, with dual goals of limiting the toxicity of therapy for low-risk patients and
maximizing the efficacy of therapy for intermediate- and high-risk patients.
The COG risk classification system has been validated [3] with the following stratification:
(1) Low risk (~50% of population): Patients with non-metastatic, grossly resected, low-grade
tumors of any size or high-grade tumors ≤5 cm.
(2) Intermediate risk (~35% of population): Cases of non-metastatic tumors that are high
grade, >5 cm, or unresectable.
(3) High risk (~15% of population): Patients with nodal or distant metastases.

In this thesis, we focus on the event-free survival and overall survival of low-risk patients
with ≤5cm high-grade tumors who were treated with surgery only, and compare this group’s
survival with the remaining patients who were identified as low-risk with non-metastatic, grossly
resected tumors.
In adults, high-grade soft tissue sarcomas that are completely removed surgically are often
treated with radiotherapy to ensure the tumor does not recur; however this adjuvant treatment
increases the risk for long-term complications such as cardiovascular diseases, impairment of
organ and body tissues, and secondary cancers. Radiotherapy is particularly toxic in pediatric
9
population in that it may generate late effects contributing to a high burden of morbidity,
including a high probability of developing chronic health conditions and experiencing severe or
life-threatening complications during adulthood [7-9].
Published data suggest that more than 90% of the low risk patients become long-term
survivors, and that the risk of metastatic recurrence is extremely low [8], so the goal of treating
such patients is focused on diminishing the late effects of treatment. Recent studies in adults with
NRSTS also suggest that carefully selected patients with adequately resected tumors can be
treated successfully without radiation therapy [10-12], making possible the diminishment of the
likelihood of late effects by minimizing therapy.
In this study, we tested whether patients <30 years of age with a ≤5 cm high-grade soft tissue
sarcoma could be cured with surgery alone without the adjuvant radiotherapy, in the hope of
decreasing the consequences of radiation (growth impairment, stiffening of muscles and skin,
fractures, secondary cancers).

10
Chapter 2: Methods

Study design and participants:
ARST0332 was a groupwide phase III study comprised of newly diagnosed NRSTS
patients with age under 30 years, confirmed by central pathology review via concurrent
enrollment on protocol COG-D9902. The patients were enrolled between February 2007 and
February 2012, from 159 institutions in the United States, Australia, Canada, New Zealand and
Puerto Rico.
Among a total of 588 patients enrolled, 551 eligible patients were recruited and assigned
to 4 treatment arms based on risk: (1) Arm A (low or high risk, 212 subjects, observation):
grossly excised low-grade and 5 cm widely excised high-grade tumor; (2) Arm B (low risk,
19 subjects, adjuvant radiotherapy): 5 cm marginally resected high-grade tumor; (3) Arm C
(intermediate or high risk, 120 subjects, adjuvant chemotherapy and radiotherapy): > 5 cm
grossly resected tumor ± metastases; (4) Arm D (intermediate or high risk, 200 subjects,
neoadjuvant chemoradiotherapy): > 5cm unresected tumor ± metastases [13]. (Figure 1 and
Figure 2)

11
Figure 1 Experimental Design Schema

12
Figure 2 Treatment Regimens

The analytic subjects for this thesis are 82 patients with ≤5 cm high-grade (POG grade=3)
tumors enrolled on ARST0332 Arm A who were treated with surgery only (all fall in the low-
risk group), compared with the remaining 152 patients classified as low-risk, who had low-grade
tumor of any size or high-grade tumor 5 cm and were enrolled in treatment arms other than
Arm A. Theoretically, patients in the comparison group should only fall in Arm A or Arm B
which are the low-risk arms, but we did see several patients assigned to Arm C which should
include intermediate- or high-risk patients only. We adopted the ‘intent-to-treat’ approach for the
analysis in this thesis, with risk classification based on Central Imaging Review. The
inconsistency seen here is possibly due to disagreement between central review and the
institutions which assigned the patients to different treatments.

Exposure Variables [14]:
Demographic variables: age at enrollment, ethnicity, race, sex
13
Pathologic diagnosis
1
: histological subtypes of the STS
POG grade
2
: the grading system of the Pediatric Oncology Group used to categorize NRSTS
and predict outcomes for children, based on previous systems developed for adults but
accounting for pathologic entities only seen in young children. The POG grade was used for
treatment determination in this trial.
• Low – POG grade 1 or 2
• High – POG grade 3
FNCLCC grade
1
: the grading system of the French Federation of Cancer Centers used for adult
sarcomas, which uses grades 1 through 3 and is useful for predicting local aggressiveness and
distant metastasis for soft tissue sarcomas.
• Low – FNCLCC grade 1 or 2
• High – FNCLCC grade 3
Primary tumor site:
• Head and Neck Sites – head, neck
• Body Wall Sites – anterior chest wall, breast, posterior chest wall, paraspinal, intraspinal,
abdominal wall
• Visceral Sites – intrathoracic, intraperitoneal, retroperitoneal, pelvis, perineum
• Upper Extremity Sites – shoulder, upper arm, lower arm, hand
• Lower Extremity Sites – hip, thigh, leg, foot
Size of primary tumor (in cm): the maximal original diameter of the primary tumor measured at
the time of pathological evaluation
Depth of primary tumor:
• Superficial – tumor is superficial to and does not involve the superficial fascia

1
For detailed list of histological types, please see Fletcher CD, Unni KK, Mertens F. World Health
Organization Classification of Tumours. Pathology and Genetics. Tumours of Soft Tissue and Bone.
IARC Press, Lyon, 2002. Available at https://www.iarc.fr/en/publications/pdfs-online/pat-
gen/bb5/BB5.pdf
2
One of the pathology objectives of ARST0332 was to prospectively evaluate the POG and FNCLCC
grading systems. During central pathology review, each tumor was assigned a grade in both the POG
system and the FNCLCC system, but the POG grade was used for treatment determination. Analysis of
study outcomes according to both the POG and FNCLCC grade assignments would help determine which
grading system better predicts outcome in childhood NRSTS, and offers the possibility of identifying the
best grading system for use in future studies of childhood NRSTS.
14
• Deep – tumor is deep to or invades the superficial fascia.
Invasiveness of primary tumor:
• Non-invasive – Primary tumor does not touch or invade neurovascular structures or bone
• Neurovascular invasion (touching) – Primary tumor touches but does not surround
neurovascular structures.
• Neurovascular invasion (surrounding) – Primary tumor surrounds neurovascular
structures
• Bone invasion (touching) – Primary tumor touches but does not destroy bone
• Bone invasion (destroying) – Primary tumor destroys bone
Microscopic surgical margins:
• Negative – A cuff of non-malignant tissue measuring at least 5 mm in all directions
surrounds the tumor in the operative specimen. When the tumor abuts fascia or
periosteum and the fascia or periosteum is removed in continuity with the tumor
specimen, this margin will also be considered negative.
• Positive – The cuff of non-malignant tissue surrounding the tumor in the operative
specimen is < 5 mm in thickness in one or more plane.

Outcome Assessment:
Overall survival (OS) time: time from enrollment until death or date last seen/censored
Event-free survival (EFS) time: time from enrollment until first relapse/progression, second
malignant neoplasm or death; if no EFS event, the EFS time was time from enrollment until date
last seen/censored.

Statistical Analysis
All statistical analyses were performed using SAS software (Version 9.4; SAS Institute
Inc, Cary, NC).
A two-tailed alternative and a 5% Type I error rate were assumed for each hypothesis
test. Exploratory data analysis including frequency counts and frequency distributions were
reported in tables.
Fisher’s Exact Test was used for the comparison of categorical variables including age
group at enrollment, ethnicity, race, sex, FNCLCC grade, depth of primary tumor, invasiveness
15
of primary tumor, microscopic surgical margins between analytic and comparison groups; t-
test/Wilcoxon rank sum tests were used for the comparison of size of primary tumor. For
pairwise comparison between the variables within the analytic group, Fisher’s Exact Test was
used to determine correlations.
Median was used for the description of follow-up time for survivors of analytic and
comparison groups. For overall survival and event-free survival, the product-limit estimate was
used to generate the survival curve. A two-sided log-rank test was used to test the hypothesis that
there is no difference for the risk of EFS-event and the risk of death between the analytic and
comparison groups
We performed a prognostic factor analysis for overall survival and event-free survival
using univariate Cox proportional hazards model in the analytic group; significant (p<0.05) and
marginally significant factors (p<0.1) were subsequently included in a multivariable Cox
regression model. Hazard ratios and 95% confidence intervals were obtained from the Cox
models.

16
Chapter 3: Results

82 patients with ≤5 cm high-grade tumors enrolled on ARST0332 Arm A who were
treated with surgery only were classified as having a ‘high-grade small tumor’, and were
included in the analytic group. The 82 patients all fell in the low-risk group (non-metastatic,
grossly resected tumor), so an appropriate comparison group would be the remaining 152
patients in the low-risk group. Tumor characteristics for the comparison group were low-grade
tumors of any size or high-grade tumors ≤5 cm and enrolled in treatment arms other than Arm A.

Baseline patient characteristics of the analytic and comparison group:
Table 1 shows the baseline characteristics of the patients, including demographics and
primary tumor characteristics. The information for size of tumor, POG grade and treatment arm
are listed in the table but excluded from comparison since these variables were used to identify
the groups. Age at enrollment, ethnicity, race and depth of primary tumor were equivalent
between the two groups. Gender significantly differed between the two groups; females were
more likely than males to be in the high-grade small tumor group (p=0.019). FNCLCC grade was
significantly different between the two groups (p<0.0001). Invasiveness of the primary tumor
and microscopic surgical margins also significantly differed between the two groups (p<0.0001);
this could be due to the possibility that characteristics that define invasiveness such as
involvement of blood vessels also influence surgical approach.

17
Table 1 Baseline Patient Characteristics of Analytic and Comparison Group
High Grade Small Tumor

P-Value
2

Analytic
(n=82)
1

Comparison
(n=152)
1

Age at Enrollment (years)
Minimum
Maximum
Age Group at Enrollment
11.42 4.72
0.73
24.52
11.94 4.95
0.78
21.10
0.44

0.10
0-5 12(14.6%) 28(18.4%)
6-10 21(25.6%) 27(17.8%)
11-15 39(47.6%) 63(41.5%)
16-18
19-21
21-30
7(8.5%)
2(2.4%)
1(1.2%)
30(19.7%)
4(2.6%)
0(0.0%)

Ethnicity 0.91
Hispanic or Latino 12(14.6%) 19(12.5%)
Not Hispanic or Latino 67(81.7%) 127(83.6%)
Unknown 3(3.7%) 6(3.9%)
Race 0.85
White 64(78.1%) 117(77.0%)
Black or African American 7(8.5%) 17(11.2%)
Asian 4(4.9%) 4(2.6%)
American Indian or Alaskan 0(0.0%) 1(0.7%)
Unknown 7(8.5%) 13(8.6%)
Gender 0.0193
Male 29(35.4%) 79(52.0%)
Female 53(64.6%) 73(48.0%)
FNCLCC Grade
3

1

3(3.7%)

70(47.0%)
<0.0001
2 48(58.5%) 64(43.0%)
3
Indeterminate
31(37.8%)
0(0.0%)
9(6.0%)
6(4.0%)

Depth of primary tumor 0.12
Deep
Superficial
Invasiveness of primary tumor
47(57.3%)
35(42.7%)
103(67.8%)
49(32.2%)

<0.0001
18
Non-invasive
Invasive
Indeterminate
Microscopic surgical margins
4

66(80.5%)
11(13.4%)
5(6.1%)

98(64.5%)
54(35.5%)
0(0.0%)

<0.0001
Negative 70(85.4%) 75(49.7%)
Positive 12(14.6%) 76(50.3%)
Size of primary tumor (cm)

2.49 1.18 4.29 3.36
-
Minimum
Maximum
0.20
5.00
0.40
20.50

POG Grade

-
1
2
3
Indeterminate
-
-
82(100%)
-
59(38.8%)
71(46.7%)
20(13.2%)
2(1.3%)

Treatment Arm

-
A: No adjuvant treatment
B: Adjuvant radiotherapy
C: Adjuvant chemotherapy +
radiotherapy
82(100%)
-
-
123(80.9%)
19(12.5%)
10(6.6%)

1
For continuous variables (age at enrollment, size of primary tumor): mean SD; for categorical
variables (the rest of the variables): number(%)
2
P-value for continuous variables: obtained by t-test; for categorical variables: obtained by
Fisher’s exact test.
3
Frequency missing in comparison group, n=3; indeterminate subjects were excluded from the
comparison
4
Frequency missing in comparison group, n=1; data obtained from Central Surgery Review

19
Characteristics of primary tumors for patients in analytic group:
For the 82 patients in the analytic group, among the 16 pathology types of STS seen (3
cases unclassified), the most common types were synovial sarcoma (16 cases, 19.5%), epitheliod
sarcoma (13 cases, 15.9%) and undifferentiated sarcoma (11 cases, 13.4%). (Table 2)

Table 2 Histological Subtypes of Soft Tissue Sarcoma
Frequency Percent
Synovial sarcoma 16 19.5
Epitheliod sarcoma 13 15.9
Undifferentiated sarcoma 11 13.4
Alveolar soft part sarcoma 9 11.0
Pleomorphic 'MFH' (undifferentiated
pleomorphic sarcoma)
8 9.8
Leiomyosarcoma 4 4.9
Low grade fibromyxoid sarcoma
(hyalinizing spindle cell tumor)
3 3.7
Clear cell sarcoma of soft tissue 3 3.7
Malignant peripheral nerve sheath
tumour
3 3.7
Unclassified soft tissue sarcoma 3 3.7
Angiosarcoma of soft tissue 2 2.4
Parachordoma 2 2.4
Adult fibrosarcoma 1 1.2
Myxofibrosarcoma 1 1.2
Sclerosing epithelioid fibrosarcoma 1 1.2
Mesenchymal chondrosarcoma 1 1.2
Pleomorphic liposarcoma 1 1.2

20
Among the 82 patients in the analytic group, primary tumors were seen in 16 sites; the
most common sites were thigh (9 cases, 11.0%), abdominal wall (9 cases, 11.0%), and leg (8
cases, 9.8%). (Table 3)

Table 3 Primary tumor site
Frequency Percent
Thigh 9 11.0
Abdominal wall 9 11.0
Leg 8 9.8
Lower arm 7 8.5
Hand 7 8.5
Head 7 8.5
Shoulder 6 7.3
Foot 5 6.1
Neck 5 6.1
Upper arm 4 4.9
Chest wall – anterior 4 4.9
Unknown 3 3.7
Pelvis 2 2.4
Chest wall – posterior 2 2.4
Paraspinal 2 2.4
Perineum 1 1.2
Hip 1 1.2

21
Description of follow-up time for survivors in analytic and comparison groups:
The overall follow-up time for 77 survivors in the analytic group and 147 in the
comparison group is shown below in Figure 3 and Table 4.

Figure 3 Distribution of Follow-up Time for Survivors

Table 4 Follow-up Time for Survivors

High Grade Small Tumor
Analytic (n=77)
1
Comparison (n=147)
1

Follow-up Time for Survivors (years)
Minimum
Maximum
4.72
0.04
8.07
4.50
0.30
7.95
1
Median

Distribution of SURV_TIME
Kernel Normal
Analytic
Comparison
HigrdSm
0
10
20
30
Percent
0
10
20
30
Percent
Analytic
0
5
10
15
20
25
Percent
0
5
10
15
20
25
Percent
Comparison
-2.5 0.0 2.5 5.0 7.5 10.0
Overall Survival Time (years)
22
Comparison of survival between the two groups:
14 patients (17.1%) in the analytic group and 15 (9.9%) in the comparison group had EFS
events. Each group had 5 patients die, and the deaths were all seen after the patients were taken
off the protocol due to an EFS event. (Table 5)
Analytic Group: The average event-free survival time was 3.94 1.90 years, with a
median of 4.17 years, and range of 0.038 to 8.074 years. The average overall survival time was
4.32 1.59 years, with a median of 4.56 years, and range of 0.038 to 8.074 years.
Comparison Group: The average event-free survival time was 3.55 2.23 years, with a
median of 3.66 years, and range of 0.006 to 8.747 years. The average overall survival time was
4.09 2.00 years, with a median of 4.16 years, and range of 0.006 to 8.747 years. (Table 6)

Table 5 Critical Events

High Grade Small Tumor
Analytic (n=82)
1
Comparison (n=152)
1

First EFS Event
Relapse or Progression
Second Malignant Neoplasm
No Event
Death
Yes
No
13(15.9%)
1(1.2%)
68(82.9%)

5(6.1%)
77(93.9%)
14(9.2%)
1(0.7%)
137(90.1%)

5(3.3%)
147(96.7%)
1
Number(%)

Table 6 Survival Time

High Grade Small Tumor
Analytic (n=82)
1
Comparison (n=152)
1

Event-free Survival Time (years)
Minimum
Maximum
Overall Survival Time (years)
Minimum
Maximum
4.17
0.038
8.074
4.56
0.038
8.074
3.66
0.006
8.747
4.16
0.006
8.747
1
Median

23
Comparison of Event-free Survival:
The event-free survival distributions of two groups were marginally significant (log-rank
p=0.08), with the event-free survival of the comparison group seem slightly better than the
analytic group over the course of follow-up (Figure 4). The hazard function had a bandwidth
1
of
5.37 (years) for the analytic group and 6.46 (years) for the comparison group, and it appeared to
be reasonable enough to satisfy the proportional hazard assumption. Using a Cox proportional
hazard model, the risk of having an EFS event for the patients in the analytic group is estimated
to be 1.91(95%CI: 0.92-3.98) times as high as the patients in the comparison group (Figure 5).

Figure 4 Comparison of Product-Limit Survival Estimates for Event-free Survival (with
number of subjects at risk)

1
Hazard rate is generally expected to change smoothly over time, rather than jump around haphazardly.
Therefore, to accomplish this smoothing, the hazard function estimate at any time interval is a weighted
average of differences within a window of time that includes many differences, known as the bandwidth.
82 67 45 7 1
152 128 96 32 0
0 2 4 6 8
Event-Free Survival Time (years)
0.0
0.2
0.4
0.6
0.8
1.0
Survival Probability
Analytic
Comparison
Comparison Analytic High-grade Small Tumor
Product-Limit Survival Estimates
With Number of Subjects at Risk
82 67 45 7 1
152 128 96 32 0
0 2 4 6 8
Event-Free Survival Time (years)
0.0
0.2
0.4
0.6
0.8
1.0
Survival Probability
Analytic
Comparison
Comparison Analytic High-grade Small Tumor
Logrank p=0.0766
+ Censored
Product-Limit Survival Estimates
With Number of Subjects at Risk
24

Figure 5 Hazard Functions for Event-free Survival
1

1
bw=bandwidth for kernel smoothing
25
Comparison of Overall Survival:
The overall survival distributions of the two groups were not significantly different (log-
rank p=0.32), and the cumulative survival estimates of the two groups were close during the
course of follow-up (Figure 6). The hazard function had a bandwidth of 4.15 (years) for the
analytic group and 6.67 (years) for the comparison group. Using a Cox proportional hazard
model, the risk of death for the patients in the analytic group is estimated to be 1.87(95%CI:
0.54-6.47) times as high as the patients in the comparison group. (Figure 7).

Figure 6 Comparison of Product-Limit Survival Estimates for Overall Survival (with number
of subjects at risk)

82 75 50 8 1
152 138 101 35 0
0 2 4 6 8
Overall Survival Time (years)
0.0
0.2
0.4
0.6
0.8
1.0
Survival Probability
Analytic
Comparison
Comparison Analytic High-grade Small Tumor
Product-Limit Survival Estimates
With Number of Subjects at Risk
82 75 50 8 1
152 138 101 35 0
0 2 4 6 8
Overall Survival Time (years)
0.0
0.2
0.4
0.6
0.8
1.0
Survival Probability
Analytic
Comparison
Comparison Analytic High-grade Small Tumor
Logrank p=0.3162
+ Censored
Product-Limit Survival Estimates
With Number of Subjects at Risk
26
Figure 7 Hazard Functions for Overall Survival
1

1
bw=bandwidth for kernel smoothing
27
Analysis of potential prognostic factors for the analytic group:
Using Cox proportional hazards models, we analyzed factors related to EFS and OS
univariately for the analytic group. Depth of primary tumor was significantly associated with
EFS (p=0.0022), with patients having deep tumors at 10.86(95%CI: 1.42-83.06) times the risk of
experiencing EFS events than patients having superficial tumors. The same direction of effect
was seen in OS, but the association was not statistically significant, with deep vs. superficial
having a hazard ratio of 2.80(95%CI: 0.31-25.07). Gender was marginally significantly
associated with both EFS and OS (both p=0.07), with females having lower failure rates than
males. Age at enrollment was marginally associated with EFS (p=0.08); the risk of having an
EFS event increased by 1.12(95%CI: 0.99-1.26) per year of age. (Table 7)
Including depth of primary tumor and marginally significant variables (age and gender)
in the multivariate Cox model for EFS, we obtained a multivariate cox model for event-free
survival in the analytic group (Table 8). After adjusting for age and gender, depth of primary
tumor was still significantly associated with the risk of EFS event (p=0.029), with a hazard ratio
of deep vs. superficial being 9.65(95%CI: 1.26-74.06).

28
Table 7 Univariate Prognostic Factor of Event-free Survival and Overall Survival for the Analytic Group (n=82)
Hazard Ratio
Event-free Survival
1
P-Value Overall Survival
1
P-Value
Age at Enrollment (per year) 1.12(0.99-1.26) 0.076 1.10(0.90-1.37) 0.33
Ethnicity
Hispanic or Latino vs. Not Hispanic or
Latino
2.59(0.81-8.27) 0.11 1.49(0.17-13.41) 0.72
Race
White vs. Asian - 0.99 - 1.00
Black or African American vs. Asian - 0.99 - 1.00
Gender
Female vs. Male 0.37(0.13-1.08) 0.070 0.13(0.02-1.19) 0.072
FNCLCC Grade
1 vs. 3

-

0.99

-

1.00
2 vs. 3 0.85(0.30-2.46) 0.77 0.99(0.17-5.90) 0.99
Depth of primary tumor
Deep vs. Superficial
Invasiveness of primary tumor
10.86(1.42-83.06) 0.002

2.80(0.31-25.07) 0.36

Invasive vs. Non-invasive
Microscopic surgical margins
1.68(0.47-6.05)

0.43

1.53(0.17-13.68)

0.70

Negative vs. Positive 1.89(0.25-14.51) 0.54 - 0.99
1
Hazard Ratio(95% CI)

29
Table 8 Multivariate Cox Model for Event-free Survival in the Analytic Group (n=82)
Chi-Square Pr > ChiSq Hazard Ratio 95% Hazard Ratio Confidence Limits
Age at Enrollment 3.0154 0.0825 1.113 0.986 1.256
Gender Female 2.8277 0.0927 0.399 0.136 1.164
Depth of Primary Tumor Deep 4.7549 0.0292 9.652 1.258 74.062

30
Chapter 4: Discussion

Baseline patient characteristics showed some differences between the two groups on
gender, FNCLCC grade (a possible explanation for this is that a correlation was found between
the POG-assigned grade and the FNCLCC-assigned grade in previous publication [15]),
invasiveness of primary tumor and microscopic surgical margins, making the analytic and
comparison groups less comparable.
There were a wide variety of 16 pathology types of STS and 16 sites seen in the primary
tumors of the patients in the analytic group.
The follow-up time for survivors was similar between the two groups, reducing the
possibility of seeing different EFS and OS solely due to different follow-up times.
The analytic group had a higher proportion of patients experiencing EFS events and death
than the comparison group. However, log-rank tests for comparison of both EFS and OS were
not significant (log-rank p=0.077 and 0.32, respectively), providing no evidence that the different
treatments adopted in the two groups generated different survival outcomes.
We only found one statistically significant prognostic factor for EFS, which was depth of
tumor. Marginally significant factors were age and gender. No significant factor was found for
OS, and gender was only marginally significantly associated with OS. The small number of
events observed (14 EFS events and 5 deaths) severely reduced power to detect any associations
between the factors and the outcomes.
Previous studies suggested that patients with non-metastatic, grossly resected tumors
were expected to have excellent long-term outcomes with 5-year survival exceeding 90% [16-
18], and such patients were categorized as ‘low-risk’ [3]. The patients included in this thesis
were all identified as low-risk, and the results were consistent with previous findings, with
survival of 93.9% in the analytic group and 96.7% in the comparison group (enrolled between
February 2007 and February 2012). Event-free survival for analytic and comparison groups was
82.9% and 90.1% respectively. While overall and event-free survival for the comparison group
were both slightly better than the analytic group, log-rank tests showed no statistically significant
survival differences. Therefore, insufficient evidence is seen in our data that adjuvant
radiotherapy is needed to treat low-risk patients. This result aligns with a study conducted in St
Jude Children’s Research Hospital [16], which pointed out that adjuvant therapy was not
31
recommended for all pediatric patients with resected NRSTS and adjuvant radiotherapy may be
useful to avoid disfigurement and functional failure in conjunction with limited surgery.
Previous publications report that the prognosis of NRSTS (including overall survival,
local and distant recurrence) varies greatly depending on the following factors: (histologic)
tumor grade, invasiveness of the tumor, tumor size, site of the primary tumor, presence of
metastases, resectability of the tumor, use of radiation therapy [16-19].
In the prognostic analysis for the analytic group, several factors associated with the
survival outcomes. Higher age was positively associated with the risk of experiencing EFS
events, which is supported by previous findings [20, 21]. Gender was marginally associated with
both EFS and OS with females at lower risk; no previous studies have reported this. Gender was
also observed to be a significant patient characteristic in that females were more likely to have
high-grade small tumors compared with other low-risk patients. Given that the sample size is
considerably large for this kind of disease, we suspect this association to be true, although the
pathological reason for this is not clear. Deep tumor was a significant risk factor for EFS, with a
high hazard ratio of 10.86 (95%CI: 1.42-83.06) compared with superficial tumors, and the
hazard ratio was 2.8 (95%CI: 0.31-25.07) for OS. Invasiveness of the tumor was not a prognostic
factor in the analytical group, in contrast to published results. Nevertheless, invasiveness and
depth of tumors were highly correlated in the analysis of pairwise comparison among analytic
group (p=0.0016), which may explain our result. Although a previous study also indicated that
positive surgical margins was a significant prognostic factor for recurrence [16], our results did
not show evidence for this. A possible explanation for this could be that the analytic group was a
low-risk subgroup of the NRSTS patients, and the result may not hold in this subgroup. Other
established prognostic factors such as tumor size, tumor grade, presence of metastases,
resectability of the tumor were used to identify the analytic group and thus were not included in
the prognostic analysis.
In summary, our data support the treatment of low-risk NRSTS patients with minimized
therapies to avoid adverse consequences associated with these therapies. We also identified some
prognostic factors for patients with low-risk high-grade small tumor, and the results largely
resembled previous findings.
32
Reference:

1 Ries, Lynn A. Gloeckler, et al. "Cancer incidence and survival among children and
adolescents: United States SEER Program 1975-1995." Cancer incidence and survival
among children and adolescents: United States SEER Program 1975-1995 (1999).
2 Pizzo, Philip A., and David G. Poplack. Principles and practice of pediatric oncology.
Lippincott Williams & Wilkins, 2015: 827-854
3 Waxweiler, Timothy V., et al. "Non-Rhabdomyosarcoma Soft Tissue Sarcomas in Children:
A Surveillance, Epidemiology, and End Results Analysis Validating COG Risk
Stratifications." International journal of radiation oncology, biology, physics 92.2 (2015):
339-348.
4 Pratt, Charles B., et al. "Treatment of unresectable or metastatic pediatric soft tissue
sarcomas with surgery, irradiation, and chemotherapy: a Pediatric Oncology Group
study." Medical and pediatric oncology 30.4 (1998): 201-209.
5 Pratt, Charles B., et al. "Role of adjuvant chemotherapy in the treatment of surgically
resected pediatric nonrhabdomyosarcomatous soft tissue sarcomas: A Pediatric Oncology
Group Study." Journal of Clinical Oncology17.4 (1999): 1219-1219.
6 Pappo, A. S., M. Devidas, and J. Jenkins. "Ifosfamide (I), Doxorubicin (D), and G-CSF (G)
for pediatric unresected metastatic non-rhabdomyosarcomatous soft tissue Sarcomas
(NRSTS): A Pediatric Oncology Group (POG) Study." Proc Am Soc Clin Oncol. Vol. 20.
2001.
7 Armstrong, Gregory T., et al. "Aging and risk of severe, disabling, life-threatening, and fatal
events in the childhood cancer survivor study." Journal of clinical oncology (2014): JCO-
2013.
8 Geenen, Maud M., et al. "Medical assessment of adverse health outcomes in long-term
survivors of childhood cancer." Jama 297.24 (2007): 2705-2715.
9 Hudson, Melissa M., et al. "Clinical ascertainment of health outcomes among adults treated
for childhood cancer." Jama 309.22 (2013): 2371-2381.
10 Baldini, Elizabeth H., et al. "Long-term outcomes after function-sparing surgery without
radiotherapy for soft tissue sarcoma of the extremities and trunk." Journal of Clinical
Oncology 17.10 (1999): 3252-3259.
33
11 Rydholm, Anders, et al. "Limb-sparing surgery without radiotherapy based on anatomic
location of soft tissue sarcoma." Journal of clinical oncology 9.10 (1991): 1757-1765.
12 Fabrizio, Patrick L., Scott L. Stafford, and Douglas J. Pritchard. "Extremity soft-tissue
sarcomas selectively treated with surgery alone." International Journal of Radiation
Oncology* Biology* Physics 48.1 (2000): 227-232.
13 Spunt, Sheri L., et al. "Risk-based treatment for nonrhabdomyosarcoma soft tissue sarcomas
(NRSTS) in patients under 30 years of age: Children's Oncology Group study
ARST0332." ASCO Annual Meeting Proceedings. Vol. 32. No. 15_suppl. 2014.
14 "Risk-based treatment for nonrhabdomyosarcoma soft tissue sarcomas (NRSTS) in patients
under 30 years of age: Children's Oncology Group study ARST0332." Protocol. Study
Chair: Spunt, Sheri L.. Version Date: 04/13/11; Amendment #3.
15 Khoury, Joseph D., et al. "Grading of nonrhabdomyosarcoma soft tissue sarcoma in children
and adolescents." Cancer 116.9 (2010): 2266-2274.
16 Spunt, Sheri L., et al. "Prognostic factors for children and adolescents with surgically
resected nonrhabdomyosarcoma soft tissue sarcoma: an analysis of 121 patients treated at St
Jude Children's Research Hospital." Journal of Clinical Oncology 17.12 (1999): 3697-3705.
17 Pisters, P. W., et al. "Analysis of prognostic factors in 1,041 patients with localized soft
tissue sarcomas of the extremities." Journal of Clinical Oncology 14.5 (1996): 1679-1689.
18 Spunt, Sheri L., et al. "Clinical features and outcome of initially unresected nonmetastatic
pediatric nonrhabdomyosarcoma soft tissue sarcoma." Journal of clinical oncology 20.15
(2002): 3225-3235.
19 Ferrari, Andrea, et al. "Adult-type soft tissue sarcomas in pediatric-age patients: experience
at the Istituto Nazionale Tumori in Milan." Journal of clinical oncology 23.18 (2005): 4021-
4030.
20 Qureshi, Sajid S., and Monica Bhagat. "Non-rhabdomyosarcoma soft-tissue sarcomas in
children: Contemporary appraisal and experience from a single centre." Journal of Indian
Association of Pediatric Surgeons 20.4 (2015): 165.
21 Spunt, Sheri L., Stephen X. Skapek, and Cheryl M. Coffin. "Pediatric nonrhabdomyosarcoma
soft tissue sarcomas." The oncologist 13.6 (2008): 668-678.

Abstract (if available)

Abstract Background: Nonrhabdomyosarcomatous soft tissue sarcomas (NRSTS) are both a rare and heterogeneous group of malignancies among children and adolescents, with an incidence of around 500 cases per year comprising approximately 5% of childhood malignancies in the United States in those <20 years of age. While radiation is widely used as an adjuvant treatment after surgical removal of the tumor in prevention of recurrence, it can also cause long-term complications. ❧ Methods: ARST0332 was a groupwide phase III study with a primary aim to evaluate a novel risk-based strategy for treating pediatric NRSTS and dual goals of limiting the toxicity of therapy for low-risk patients and maximizing the efficacy of therapy for intermediate- and high-risk patients. ARST0332 enrolled 551 eligible patients and assigned them to 4 treatment arms based on risk: Arm A (observation), Arm B (adjuvant radiotherapy), Arm C (adjuvant chemotherapy and radiotherapy) and Arm D (neoadjuvant chemoradiotherapy). ❧ This thesis focused on whether patients <30 years of age with a ≤5 cm high-grade soft tissue sarcoma could be cured with surgery alone without the adjuvant radiotherapy, so as to avoid the consequences of radiation. The analytical group was 82 patients with ≤5 cm high-grade tumors enrolled on ARST0332 Arm A treated with surgery only (all classified as low-risk), compared with the remaining 152 patients identified as low-risk (non-metastatic, grossly resected tumors). Endpoints included event-free survival (EFS) and overall survival (OS). ❧ Results: The analytic group had a slightly higher proportion of patients experiencing EFS events (including relapse, progression and second malignant neoplasm) and death than the comparison group, although log-rank tests for comparison of both EFS and OS were not significant (log-rank p=0.08 and 0.32, respectively). Depth of tumor was found to be a statistically significant prognostic factor for event-free survival (p=0.029), with a hazard ratio of deep vs. superficial of 9.65(95%CI: 1.26-74.06), after adjusting for age and gender. No significantly factor was related to overall survival. ❧ Conclusion: Patients with high-grade small tumors treated with surgery only did not have worse survival outcome than other low-risk patients treated with adjuvant radiotherapy.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A novel risk-based treatment strategy evaluated in pediatric head and neck non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) patients: a survival analysis from the Children's Oncology Group study...

PDF

Risk factors and survival outcome in childhood alveolar soft part sarcoma among patients in the Children’s Oncology Group (COG) Phase 3 study ARST0332

PDF

Comparative study of the POG and FNCLCC grading systems in non-rhabdomyosarcoma soft tissue sarcomas

PDF

Extremity primary tumors in non-rhabdomyosarcoma soft tissue sarcoma: survival analysis

PDF

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Infants in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Distant metastatic nonrhabdomyosarcoma soft tissue sarcomas in children and adolescents: clinical features and survival outcome among patients in Children's Oncology Group Phase 3 Study ARST0332

PDF

Body wall primary tumors in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Effects of demographic and tumor characteristics on outcomes in children with malignant peripheral nerve sheath tumors

PDF

An analysis of disease-free survival and overall survival in inflammatory breast cancer

PDF

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

PDF

An assessment of impact of early local progression on subsequent risk for the treatment failure in adolescent and young adult patients with non-metastatic osteosarcoma

PDF

Carboplatin and vincristine chemotherapy for progressive low grade gliomas in pediatric patients with or without neurofibromatosis type 1 (NF1)

PDF

Contemporary outcomes for adult congenital heart surgery in an adult tertiary care hospital

PDF

Explore risk and protective factors of undifferentiated embryonal sarcoma of the liver

PDF

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

PDF

Long-term effects on vertebral body height growth of dose sculpting intensity modulated radiation therapy for children with neuroblastoma

PDF

Associations between isoflavone soy protein (ISP) supplementation and breast cancer in postmenopausal women in the Women’s Isoflavone Soy Health (WISH) clinical trial

PDF

Statistical analysis of a Phase II study of AMG 386 versus AMG 386 combined with anti-VEGF therapy in patients with advanced renal cell carcinoma

PDF

Randomized clinical trial generalizability and outcomes for children and adolescents with high-risk acute lymphoblastic leukemia

Asset Metadata

Creator Wu, Qingyuan (author)

Core Title 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...

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 04/22/2016

Defense Date 04/22/2016

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

Tag clinical trial,non-rhabdomyosarcoma soft tissue sarcoma,OAI-PMH Harvest,pediatric,survival analysis

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Barkauskas, Donald (committee chair), Krailo, Mark (committee member), Mack, Wendy (committee member)

Creator Email qingyuan@usc.edu,qyuan.w@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-242773

Unique identifier UC11278078

Identifier etd-WuQingyuan-4365.pdf (filename),usctheses-c40-242773 (legacy record id)

Legacy Identifier etd-WuQingyuan-4365-0.pdf

Dmrecord 242773

Document Type Thesis

Format application/pdf (imt)

Rights Wu, Qingyuan

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