Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
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
(USC Thesis Other)
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
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
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
by
Fan Zhang
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOSTATISTICS)
May 2018
Copyright 2018
Fan Zhang
ii
DEDICATION
To my friends and family, for their continual support and encouragement.
iii
ACKNOWLEDGEMENTS
This work was completed under the direction and supervision of my thesis committee chair,
Dr. Donald Barkauskas. I would like to express my sincerest gratitude and appreciation to
Dr. Barkauskas for his encouragement and guidance throughout the duration of my
research and the writing of my thesis. I also want to extend my appreciation to each of my
thesis committee members, Dr. Meredith Franklin, Dr. Christianne Lane, and Dr. Wendy
Mack, for their insight and suggestions throughout the preparation of this manuscript.
iv
TABLE OF CONTENTS
Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
Introduction
Methods
Study Population
Risk Classification System
Treatment
Prognostic Factors and Outcome Measurement
Statistical Methods
Results
Baseline Characteristics
Overall Effects of Distant Metastases
Effects of Prognostic Factors within Distant Metastases
Discussion
References
ii
iii
v
vi
1
2
4
4
4
5
7
8
10
10
12
13
21
23
v
LIST OF TABLES
Table 1. Comparison of characteristics across patient groups with and
without distant metastatic disease
Table 2. Log-rank tests for group differences in overall survival among
patients with distant metastatic NRSTS: Screening variables for Cox
regression
Table 3. Multivariate Cox proportional hazard ratios and 95% confidence
intervals for overall survival among distant metastatic patients
Table 4. Log-rank tests for group differences in event-free survival
among patients with distant metastatic NRSTS: Screening variables for
Cox regression
11–12
14
15
15
vi
LIST OF FIGURES
Figure 1. Experimental Design Schema (taken from Children’s Oncology
Group ARST0332 study protocol
13
)
Figure 2a. Kaplan-Meier estimated overall survival for patients stratified
by distant metastatic disease.
Figure 2b. Kaplan-Meier estimated event-free survival for patients
stratified by distant metastatic disease.
Figure 3a. Kaplan-Meier estimated overall survival for all patients
stratified by sites of metastases.
Figure 3b. Kaplan-Meier estimated event-free survival for all patients
stratified by sites of metastases.
Figure 4. Kaplan-Meier estimated overall survival for patients with
distant metastases stratified by race category.
6
16
17
18
19
20
1
ABSTRACT
To better understand the risk factors among nonrhabdomyosarcoma soft tissue sarcomas
(NRSTS) in children and adolescents diagnosed with distant metastatic disease, the present
study investigated the differences between patients with distant metastatic NRSTS (n=72)
and all other patients with NRSTS (n=456) with respect to diagnostic characteristics as
well as overall and event-free survival, and attempted to identify certain prognostic features
associated with greater risk of mortality among patients with distant metastatic NRSTS.
Survival probabilities were calculated using the Kaplan-Meier method and tests of
significance for prognostic factors were conducted using nonparametric log-rank tests.
Univariate associations significant at p<0.20 were entered into a multivariate Cox stepwise
regression model. For NRSTS patients with distant metastases compared to those without,
risk factors were found to be significantly different in all disease characteristics, including
primary tumor size, primary tumor depth, primary tumor invasiveness, tumor histologic
grade, whether lymph node metastases were present, and whether the primary tumor was
unresectable. Patients with distant metastases fared significantly worse in both overall and
event-free survival when compared to all other patients, as expected. Among distant
metastatic patients, the multivariate regression models for both overall and event-free
survival yielded no statistically significant prognostic factors, although race was found to
be marginally significant in overall survival.
2
INTRODUCTION
Soft tissue sarcomas cover about 4%–8% of all childhood cancers in total, with
approximately 900 new cases diagnosed every year.
1,2
Of these, about 50%–60% are
diagnosed as rhabdomyosarcoma (RMS), while the remaining are nonrhabdomyosarcoma
soft tissue sarcomas (NRSTS).
3-5
Nonrhabdomyosarcoma soft tissue sarcomas differ
greatly from their rhabdomyosarcoma counterparts with respect to histologic distributions,
clinical features, prognostic characteristics, and treatment response.
4,6
The present study
focuses on the survival outcome and identification of prognostic features for survival
among NRSTS patients.
The majority of studies published on NRSTS place emphasis on adult patients. Prognostic
data from these studies may be used to provide some information regarding treatments;
however, they cannot be directly applied to the treatment of pediatric NRSTS due to
differences in tumor histologic behavior and greater risks of treatment in children.
7,8
The
standard treatment for NRSTS in pediatric cases is surgical resection of the primary tumor,
with or without radiotherapy, since pediatric NRSTSs have been shown to be relatively
resistant to chemotherapy.
4,6,9
Most cases of NRSTS are localized at presentation, and over
70% of patients with non-metastatic surgically resected disease are expected to achieve
long-term survival.
4,6,10,11
However, only half of those patients with large, high-grade tumors and/or unresectable
disease will become long-term survivors.
4,6
Additionally, the 15%–30% of NRSTS patients
3
diagnosed with metastatic disease have much poorer prognoses with less than 10%
expected to become long-term survivors, despite the use of a combination of treatments,
including surgery, chemotherapy, and radiotherapy.
4,10
Additionally, for those patients who
survive their disease, the late effects of treatment may be detrimental to their health.
6,7
Few
studies have been published investigating the treatment effects of radiation therapy and
chemotherapy on childhood development in patients with NRSTS.
12
Due to these risks
posed by treatment, the use of certain therapies, including radiotherapy and chemotherapy,
are limited to patients who are classified as high risk of disease recurrence and/or patients
with high-grade tumors and unresectable disease.
6,7
Despite the favorable outcome for children with non-metastatic resectable disease, the
outcome for patients with metastatic non-resectable disease is much poorer, and requires
further investigation. In the present study, we strive to determine the survival outcome and
better understand the treatment strategy for pediatric NRSTS patients, specifically in
patients with distant metastases. We evaluate and compare the survival of patients
diagnosed with distant metastatic NRSTS to all other NRSTS patients, as well as to those
patients diagnosed with regional lymph node metastases only. We also investigate the
prognostic factors contributing to survival in patients diagnosed with distant metastatic
disease. This will assist in the identification and confirmation of certain prognostic factors
used to predict survival in pediatric NRSTS cases.
4
METHODS
Study Population
Patients were enrolled on the Children’s Oncology Group trial ARST0332, a group-wide
Phase 3 risk-based clinical trial for NRSTS in patients under 30 years of age.
13
Patients
were required to meet a number of eligibility criteria prior to being enrolled, and patient
characteristics were collected prior to the start of treatment.
13
The patient population of interest in the present study was taken from the above clinical
trial. Patients with distant metastases (n=72) were analyzed and compared to non-
metastatic patients (n=456). Patients with distant metastatic disease were also analyzed
independently to determine prognostic factors contributing to survival.
Risk Classification System
Few studies evaluating the prognostic factors of pediatric NRSTS have been performed.
Of those published, the most important prognostic factors for survival are the presence of
metastases, whether the tumor was resectable, tumor grade (low vs. high), and tumor size
(≤5 cm vs. >5 cm in diameter).
4,6,10,14,15
Past studies have classified and analyzed patients
by surgical resection, where patients with complete resection (with negative margins) were
compared to patients with evidence of residual tumor (whether microscopic or gross).
11
In
order to guide treatment based on certain risk factors, the ARST0332 study classified
patients into three distinct risk categories
13
:
5
Low risk — This group was comprised of patients with non-metastatic surgically
resected disease. Patients with tumors classified as both high grade and >5 cm in
size were not placed in this group. Over 90% of patients in this group are expected
to achieve long-term survival (≥5 years).
4,6,10,11
Intermediate risk — This group was comprised of patients with either non-
metastatic surgically resected disease tumors classified as both high grade and >5
cm in size or patients with non-metastatic and surgically unresectable tumors. Only
half of these patients are expected to become long-term survivors.
4,6
High risk — This group was comprised of patients diagnosed with metastatic
disease (to regional lymph nodes and/or distant sites). Patients in this group have
very poor prognoses, and less than 10% are expected to become long-term
survivors.
4,10
The present study focused on patients in the high-risk category, particularly those patients
with distant metastases. The prognostic impact of certain risk factors on survival, including
maximal tumor diameter, histologic grade, presence/absence of lymph node metastases,
tumor invasiveness, and whether the tumor was resectable, was evaluated.
Treatment
A number of clinical features, namely whether metastatic disease is present, whether the
tumor was resectable, whether the tumor grade is low or high, and whether the tumor size
6
is ≤5 cm or >5 cm in diameter, determined the risk categories detailed above and the
assigned treatment arm for all pediatric patients in this study.
Figure 1. Experimental Design Schema (taken from Children’s Oncology Group ARST0332
study protocol
13
) shows the risk stratification and treatment assignment for all possible
combinations given the prognostic factors above.
Figure 1 shows that all patients with distant metastatic disease are classified as high risk.
Among patients in this category, patients were split into 3 groups in order to determine the
treatment that they would receive. Patients with low grade tumors and complete resection
of disease would not have received any experimental therapy and would have been placed
on Treatment Arm A for observation, although no high-risk patients were placed on
7
Treatment Arm A in the trial. Patients who had their primary tumor grossly resected were
placed on Treatment Arm C, and received chemotherapy and adjuvant radiotherapy. Lastly,
patients who had an unresectable primary tumor were placed on Treatment Arm D, and
received the experimental chemoradiotherapy treatment. A study conducted at St. Jude
Children’s Research Hospital revealed that patients with surgically unresectable pediatric
NRSTS had a higher probability of survival when given both chemotherapy and
radiotherapy together, when compared to either chemotherapy or radiotherapy alone.
6
Thus, a new combined chemotherapy and radiotherapy approach was utilized for these
patients. More details for each treatment can be found in the Children’s Oncology Group
ARST0332 study protocol.
13
Prognostic Factors and Outcome Measurement
The main outcomes analyzed in this study were: (1) overall survival, where the events
monitored included deaths only, and (2) event-free survival, where the events monitored
included death, disease relapse and/or progression, and second malignant neoplasm. For
patients who did not experience an event, censoring occurred at the patient’s last follow-
up time.
Possible prognostic factors included personal risk factors and disease characteristics. The
personal risk factors were age at enrollment, sex, race, and ethnicity; while the disease
characteristics were primary tumor size, primary tumor depth, tumor grade, primary tumor
invasiveness, whether lymph node metastases were present, and whether the tumor was
unresectable.
8
For the personal risk factors, age at first enrollment was analyzed both as a continuous
variable, as well as categorized into three groups: 0–<10 years of age; 10–<18 years of age;
and 18–<30 years of age. Race was divided into four categories: White; African American;
Other; and Unknown. Ethnicity was classified into three categories: Hispanic/Latino; Not
Hispanic/Latino; and Unknown. For the disease characteristic risk factors, primary tumor
size was analyzed both as a continuous variable, as well as categorized into 4 groups: 5
cm; 5–10 cm; 10– 15 cm; and >15 cm. Primary tumor depth was split into two categories:
Superficial; and Deep. For tumor grade, two different pathology grading systems were
utilized — Pediatric Oncology Group (POG) and Fédération Nationale des Centres de
Lutte Contre le Cancer (French Federation of Centers in the Fight Against Cancer;
FNCLCC). Both pathologic grading scales range from 1 to 3, with increasing severity. For
both POG and FNCLCC histologic grades, grades of 1 or 2 were classified as “low grade”
in our analyses, while grades of 3 were classified as “high grade”. Tumors that could not
be graded remained indeterminate. Primary tumor invasiveness was classified as either
invasive or non-invasive. Whether lymph node metastases were present and whether the
tumor was unresectable were classified as binary (yes/no) variables, and analyzed as such.
Statistical Methods
Comparisons of prognostic factors between distant metastatic patients and all other patients
with NRSTS were performed using t-tests for continuous variables and Fisher’s exact test
for proportions for categorical variables. Overall survival was defined as the time interval
from date of enrollment on the study to either date of death or date of last contact, if the
9
patient was lost to follow-up. Event-free survival (EFS) was defined as the time interval
from the date of enrollment to either the date of disease relapse and/or progression, date of
second malignant neoplasm (SMN), date of death, or date of last contact, whichever
occurred first, in that order, in alignment with previous work from Children’s Oncology
Group. Patients who experienced relapse and/or progression, SMN, or death were
considered to have had an EFS event. All other patients were considered as censored at
time of last contact. The probabilities of overall survival and event-free survival were
estimated using the Kaplan-Meier product limit method.
16
Log-rank test statistics were
used to test for differences between strata of individual prognostic factors for both overall
and event-free survival. Observations with unknown and/or indeterminate results were
excluded from their respective univariate analyses. In order to prescreen our variable set,
those variables with an -level below 20% on the log-rank tests were entered into a Cox
regression model for multivariable analyses, utilizing stepwise selection to obtain a final
model. A significance level to enter of 0.05 and a significance level to stay of 0.05 was
utilized. Associations with survival are presented as hazard ratios as estimates of relative
risks, along with 95% confidence intervals.
All data were obtained from Children’s Oncology Group Phase 3 clinical trial
ARST0332.
13
This trial opened on February 5, 2007, and closed on February 10, 2012.
Data current to June 30, 2017, were used in these analyses. The analyses were conducted
in SAS (Version 9.4; SAS Institute, Cary, NC). Proportional hazard assumptions for the
Cox regression model were explored graphically using the proc lifetest command and
deemed to be sufficient for our analyses.
10
RESULTS
Baseline Characteristics
There were 528 eligible and evaluable patients enrolled in the Children’s Oncology Group
ARST0332 study, 72 of whom were diagnosed with distant metastatic disease. Selected
personal risk factors and disease characteristics of interest for patients stratified by
presence of distant metastases are presented in Table 1. Among personal risk factors,
patients who had distant metastases were significantly older at enrollment when compared
to all other patients, both when age at enrollment is classified as a continuous variable and
as a three-group categorical variable (p<0.0001 and p=0.0018, respectively). All disease
characteristics were significantly different between the two groups. When measured on a
continuous scale, primary tumor size was significantly larger in distant metastatic patients
(p<0.0001). A greater proportion of patients with distant metastases had primary tumor
size in the 15+ cm group (26.4% in distant metastases vs. 9.0% in all other patients).
Patients with distant metastases differed from all other patients in both primary tumor depth
and primary tumor invasiveness, with distant metastatic patients generally having deeper
and more invasive tumors (p<0.0001 for both). Patients with distant metastases were
significantly more likely to have a high-grade tumor in the POG classification (94.4% in
distant metastases vs. 68.9% in all other patients; p<0.0001). Similar results (62.5% high-
grade in distant metastases vs. 41.7% in all other patients, p=0.0002) were found for the
FNCLCC classification system. Whether lymph node metastases were present and whether
the tumor was unresectable also differed between patients with distant metastases and all
other patients, with patients with distant metastatic disease having a greater percentage of
11
lymph node metastases (16.7% in distant metastasis vs. 1.8% in all other patients) and a
greater percentage of tumors that were unresectable (73.6% in distant metastases vs. 29.4%
in all other patients) (p<0.0001 for both). Personal risk factors of sex, race, and ethnicity
did not differ between patient groups (p=0.38, p=0.23, p=0.22, respectively).
Table 1. Comparison of characteristics across patient groups with and without distant metastatic
disease
Patient Characteristic
Mean (SD) or n (%)
Distant
Metastases
(n = 72)
All Other
Patients
(n = 456)
p-value* for
difference
across
groups
Age at enrollment, years (continuous) 16.03 (4.82) 12.53 (5.52) <0.0001
Age at enrollment, years 0.0018
[0-10) 9 (12.5) 136 (29.8)
[10-18) 45 (62.5) 256 (56.1)
[18-30) 18 (25.0) 64 (14.0)
Sex 0.38
Male 37 (51.4) 208 (45.6)
Female 35 (48.6) 248 (54.4)
Race 0.23
White 51 (70.8) 323 (70.8)
African American 15 (20.8) 65 (14.3)
Other 3 (4.2) 21 (4.6)
Unknown 3 (4.2) 47 (10.3)
Ethnicity 0.22
Hispanic or Latino 11 (15.3) 69 (15.1)
Not Hispanic or Latino 61 (84.7) 368 (80.7)
Unknown 0 (0) 19 (4.2)
Primary tumor size, cm (continuous) 12.56 (5.34) 7.09 (5.15) <0.0001
Primary tumor size, cm <0.0001
≤5 4 (5.6) 191 (41.2)
(5-10] 21 (29.2) 146 (32.0)
(10-15] 28 (38.9) 78 (17.1)
≥15 19 (26.4) 41 (9.0)
Primary tumor depth <0.0001
Superficial 2 (2.8) 90 (19.7)
Deep 70 (97.2) 366 (80.3)
POG Histologic Grade <0.0001
Low 4 (5.6) 142 (31.1)
High 68 (94.4) 314 (68.9)
FNCLCC Histologic Grade 0.0002
Low 26 (36.1) 266 (58.3)
High 45 (62.5) 190 (41.7)
Indeterminate 1 (1.4) 0 (0)
Primary tumor invasiveness <0.0001
Non-invasive 15 (20.8) 214 (46.9)
Invasive 57 (79.2) 242 (53.1)
12
Patient Characteristic
Mean (SD) or n (%)
Distant
Metastases
(n = 72)
All Other
Patients
(n = 456)
p-value* for
difference
across
groups
Lymph Node Metastases <0.0001
Yes 12 (16.7) 8 (1.8)
No 60 (83.3) 448 (98.3)
Unresectable Tumor <0.0001
Yes 53 (73.6) 134 (29.4)
No 19 (26.4) 322 (70.6)
*p-values were obtained using t-tests for continuous characteristics and Fisher’s exact test for categorical
characteristics.
Overall Effects of Distant Metastases on Survival
The median follow-up time for those patients alive at last contact was 5.69 years (IQR:
4.57, 7.15). Among NRSTS patients diagnosed with distant metastases, the median follow-
up time for those patients alive at last contact was 5.97 years (IQR: 3.79, 7.43), while for
all other NRSTS patients, the median follow-up time was 5.66 years (IQR: 4.57, 7.08).
Using the Kaplan-Meier survival method, we observed a statistically significant difference
in overall survival rate of NRSTS with those diagnosed with distant metastatic disease
compared to all other patients (
2
=176.4, df = 1, p<0.0001). NRSTS patients who did not
have distant metastases had significantly higher survival rates than those with distant
metastases. The 5-year overall survival rate was 87.1% (95% CI: (83.4%, 90.0%)) for those
without distant metastatic disease compared to 30.9% (95% CI: (20.3%, 42.0%)) for those
with distant metastatic disease. Similarly, we observed a statistically significant difference
in event-free survival rate of NRSTS between patients with distant metastases and all other
patients (
2
=183.7, df = 1, p<0.0001). The 5-year event-free survival rate was 75.7% (95%
CI: (71.3%, 79.5%)) for those without distant metastatic disease compared to 17.7% (95%
CI: (9.8%, 27.4%)) for those with distant metastatic disease. Figures 2A and 2B show the
13
Kaplan-Meier survival curves for overall survival and event-free survival, respectively,
stratified by whether distant metastatic disease was present.
A four-way comparison of distant metastases against the presence of lymph node
metastases was performed on overall survival, yielding a statistically significant
association (
2
=188.7, df = 3, p<0.0001). Similar results were found with event-free
survival (
2
=189.2, df = 3, p<0.0001). Figures 3A and 3B show the Kaplan-Meier survival
curves for overall survival and event-free survival, respectively. A Cox proportional
hazards regression model including the presence of distant metastases, the presence of
lymph node metastases, and their interaction, revealed that only presence of distant
metastases was significant (p=0.01).
Prognostic Factors among Patients with Distant Metastases
The univariate nonparametric associations for prognostic factors predicting overall
survival among 72 patients with distant metastatic NRSTS are provided in Table 2. Race,
ethnicity, primary tumor depth, and presence of lymph node metastases were significantly
associated with overall survival outcome at the p<0.20 level. Among all of the above, only
race was found to be statistically significant at the =0.05 level (p=0.029).
Upon running our stepwise multivariate Cox regression selection model for overall
survival, with a significance level to enter and significance level to stay set at 0.05, none
of the four variables were selected. Race entered the model first (p=0.0288 by the Score
test), before being taken out of the model immediately (p=0.0503 by the Wald test). The
14
final hazard ratios and 95% confidence intervals in the multivariate Cox proportional
hazards regression model can be found in Table 3. Due to race being entered into the model
and removed immediately after, we were unable to test for the significance of the other
three variables in our statistical analysis software; thus, a separate Cox regression selection
model excluding race was performed with the three remaining variables, of which none
were selected to be kept in the multivariate model (p=0.74 for ethnicity, p=0.16 for primary
tumor depth, p=0.17 for lymph node metastases). Among the categories of race, African
American patients (n=15) were found to have slightly higher risk when compared to White
patients (n=51) in terms of overall survival, although not statistically significant (p=0.57).
Patients classified under ‘Other’ race category (n=3) had significantly higher risk
compared to White patients in overall survival (HR=4.60, 95% CI: (1.35, 15.73))
(p=0.0149). The Kaplan-Meier survival curve for overall survival by race category is
shown in Figure 4.
Table 2. Log-rank tests for group differences in overall survival among patients with distant
metastatic NRSTS: Screening variables for Cox regression
Variables p-value
Age at Enrollment (years) 0.7718
Age Category at Enrollment (years) 0.8571
Sex 0.5417
Race 0.0286*
Ethnicity 0.1389*
Primary Tumor Size (cm) 0.9643
Primary Tumor Size Category (cm) 0.9831
Primary Tumor Depth 0.1781*
POG Histologic Grade 0.6752
FNCLCC Histologic Grade 0.8467
Primary Tumor Invasiveness 0.3661
Lymph Node Metastases 0.1233*
Unresectable Tumor 0.4329
*refers to p-value less than 0.20.
15
Table 3. Multivariate Cox proportional hazard ratios and 95% confidence intervals for overall
survival among distant metastatic patients
Variables N
Hazard Ratio
(95% CI)
Individual
p-value
Overall
Wald χ
2
Overall
df
Overall
p-value
Race White
(Baseline)
51 –– ––
5.98 2 0.0503*
African
American
15 1.23 (0.61, 2.50) 0.5652
Other 3 4.60 (1.35, 15.73) 0.0149
*p-value from score test = 0.0288 when entered into the model.
The univariate nonparametric associations for prognostic factors predicting event-free
survival are presented in Table 4. Among all the risk factors considered, only race was
significantly associated with event-free survival outcome at the <0.20 level. No variables
were statistically significant at the =0.05 level. Upon running our stepwise multivariate
Cox regression selection model for event-free survival, we did not select to keep race in
the model (p=0.16).
Table 4. Log-rank tests for group differences in event-free survival among patients with distant
metastatic NRSTS: Screening variables for Cox regression
Variables p-value
Age at Enrollment (years) 0.2519
Age Category at Enrollment (years) 0.8902
Sex 0.8468
Race 0.1434*
Ethnicity 0.5501
Primary Tumor Size (cm) 0.4975
Primary Tumor Size Category (cm) 0.6052
Primary Tumor Depth 0.3487
POG Histologic Grade 0.3711
FNCLCC Histologic Grade 0.4732
Primary Tumor Invasiveness 0.2503
Lymph Node Metastases 0.2804
Unresectable Tumor 0.9035
*refers to p-value less than 0.20.
16
Figure 2a. Kaplan-Meier estimated overall survival for patients stratified by distant metastatic
disease. Patients with distant metastases (n=72, shown in red) fared significantly poorer in overall
survival when compared to all other patients (n=456) (p<0.0001).
Kaplan-Meier plot of Overall Survival for all patients by distant metastases
With number of subjects at risk
17
Figure 2b. Kaplan-Meier estimated event-free survival for patients stratified by distant metastatic
disease. Patients with distant metastases (n=72, shown in red) fared significantly poorer in event-
free survival when compared to all other patients (n=456) (p<0.0001).
Kaplan-Meier plot of Event-Free Survival for all patients by distant metastases
With number of subjects at risk
18
Figure 3a. Kaplan-Meier estimated overall survival for all patients stratified by sites of
metastases. Overall survival was found to be significantly different across all four groups
(p<0.0001).
Kaplan-Meier plot of Overall Survival for all patients by sites of metastases
19
Figure 3b. Kaplan-Meier estimated event-free survival for all patients stratified by sites of
metastases. Event-free survival was found to be significantly different across all four groups
(p<0.0001).
Kaplan-Meier plot of Event-Free Survival for all patients by sites of metastases
20
Figure 4. Kaplan-Meier estimated overall survival for patients with distant metastases stratified by
race category. Overall survival was found to differ significantly between the Other (n=3) and White
(n=51) race categories (p=0.0149) but not between the White and African American (n=15) race
categories (p=0.57).
Kaplan-Meier plot of Overall Survival for distant metastatic patients by race
21
DISCUSSION
The analyses from the present study indicated that NRSTS patients with distant metastases
had a significantly worse rate of survival when compared to all other patients with NRSTS.
This was expected to be the case since a greater risk of mortality would be present if the
primary disease spreads to a different organ in the body.
For the Cox proportional hazards regression for overall survival, we noticed that race was
being entered and then taken out immediately after. Upon more careful inspection, race
was entering the regression model as significant calculated by the Score test statistic
(p=0.0288), while it was being removed from the regression model as calculated by the
Wald test statistic (p=0.0503). Due to this, the other three variables in our model never got
a chance to enter the model, so race was temporarily removed in order to assess whether
any combination of ethnicity, primary tumor depth, and/or presence of lymph node
metastases was significant enough to enter the model. Even though the final model ended
up empty, we decided to keep race in the final multivariate model for overall survival, due
to the above discrepancy.
Looking more carefully at the race association, we observe the difference between the
hazard ratios between African American race vs. White race and Patients of ‘Other’ race
vs. White race. The risk of mortality was significantly greater in the patients of ‘Other’
race when compared to White patients, but not significantly different between African
American and White patients. This indicates that the significance of race in our multivariate
22
regression model may be most likely due to the statistical significance found in patients
classified as ‘Other’ race category (n=3). Given the small number of patients in the ‘Other’
race category and its highly significant result, the inclusion of race in our multivariate
model (See Table 3) may not be fully appropriate. Thus, our multivariate model of race
alone as a prognostic factor for overall survival within patients with distant metastases
should be taken with reservation.
The nonsignificant results obtained for both overall survival and event-free survival
indicate a number of things. Firstly, it could be that none of these prognostic factors
provided additional information that is significant to survival outcome once a patient has
been diagnosed with distant metastatic disease. In other words, we expected and observed
a significantly large difference between the survival of distant metastatic patients and all
other patients, possibly suggesting that among patients with distant metastases, the risk
factors that we investigated no longer make a difference. Another possibility to explain the
lack of significant results is that there may be additional prognostic factors that were not
analyzed in this study that should have an effect on the risk of mortality within patients
with distant metastases. Whether or not the case, further investigation of prognostic factors
on the survival in patients with distant metastatic disease is warranted. Given that the
prognostic factors tested for in this study were found to be insignificant, future research
should focus on certain elements not covered, including influence of tumor location and
genetic markers.
23
REFERENCES
1. Loeb D, Thornton K, Shokek O. Pediatric Soft Tissue Sarcomas. Surgical Clinics of
North America. 2008;88(3):615-627. doi:10.1016/j.suc.2008.03.008.
2. Smith K, Indelicato D, Knapik J et al. Adjuvant Radiotherapy for Pediatric and
Young Adult Nonrhabdomyosarcoma Soft-Tissue Sarcoma. International Journal of
Radiation Oncology*Biology*Physics. 2011;81(1):150-157.
doi:10.1016/j.ijrobp.2010.05.005.
3. Sangkhathat S. Current management of pediatric soft tissue sarcomas. World Journal
of Clinical Pediatrics. 2015;4(4):94. doi:10.5409/wjcp.v4.i4.94.
4. Spunt S, Hill D, Motosue A et al. Clinical Features and Outcome of Initially
Unresected Non-metastatic Pediatric Nonrhabdomyosarcoma Soft Tissue Sarcoma.
Journal of Clinical Oncology. 2002;20(15):3225-3235. doi:10.1200/jco.2002.06.066.
5. Qureshi S, Bhagat M. Non-rhabdomyosarcoma soft-tissue sarcomas in children:
Contemporary appraisal and experience from a single centre. Journal of Indian
Association of Pediatric Surgeons. 2015;20(4):165. doi:10.4103/0971-9261.154664.
6. Spunt S, Poquette C, Hurt Y 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. 1999;17(12):3697-3705. doi:10.1200/jco.1999.17.12.3697.
7. Spunt S, Skapek S, Coffin C. Pediatric Nonrhabdomyosarcoma Soft Tissue Sarcomas.
The Oncologist. 2008;13(6):668-678. doi:10.1634/theoncologist.2007-0182.
24
8. Dillon P, Maurer H, Jenkins J et al. A prospective study of nonrhabdomyosarcoma
soft tissue sarcomas in the pediatric age group. Journal of Pediatric Surgery.
1992;27(2):241-245. doi:10.1016/0022-3468(92)90320-7.
9. Rao B. Nonrhabdomyosarcoma in children: Prognostic factors influencing survival.
Seminars in Surgical Oncology. 1993;9(6):524-531. doi:10.1002/ssu.2980090611.
10. Pappo A, Rao B, Jenkins J et al. Metastatic nonrhabdomyosarcomatous soft-tissue
sarcomas in children and adolescents: The St. Jude Children's Research Hospital
experience. Medical and Pediatric Oncology. 1999;33(2):76-82.
doi:10.1002/(sici)1096-911x(199908)33:2<76::aid-mpo3>3.0.co;2-b.
11. Pratt C, Pappo A, Gieser P 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 Oncology. 1999;17(4):1219-
1219. doi:10.1200/jco.1999.17.4.1219.
12. Merchant T, Parsh N, Lezamo del Valle P et al. Brachytherapy for pediatric soft-
tissue sarcoma. International Journal of Radiation Oncology*Biology*Physics.
2000;46(2):427-432. doi:10.1016/s0360-3016(99)00381-8.
13. Children’s Oncology Group. Risk-Based Treatment For Non-Rhabdomyosarcoma Soft
Tissue Sarcomas (NRSTS) In Patients Under 30 Years Of Age: A Groupwide Phase III
Study.
14. Marcus K, Grier H, Gebhardt M et al. Childhood soft tissue sarcoma: 20-year
experience. International Journal of Radiation Oncology*Biology*Physics.
1994;30:204. doi:10.1016/0360-3016(94)90713-7.
25
15. McGrory J, Pritchard D, Arndt C, Nascimento A, Remstein E, Rowland C.
Nonrhabdomyosarcoma Soft Tissue Sarcomas in Children. The Mayo Clinic
experience. Clinical Orthopaedics and Related Research. 2000;374:247-258.
doi:10.1097/00003086-200005000-00022.
16. Klein J, Moeschberger M. Survival Analysis: Techniques for Censored and Truncated
Data. 2
nd
ed. New York, NY: Springer; 2005.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma
PDF
Extremity primary tumors in non-rhabdomyosarcoma soft tissue sarcoma: survival analysis
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
Infants in non-rhabdomyosarcoma soft tissue sarcoma
PDF
Body wall primary tumors in non-rhabdomyosarcoma soft tissue sarcoma
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
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
Effects of demographic and tumor characteristics on outcomes in children with malignant peripheral nerve sheath tumors
PDF
Comparative study of the POG and FNCLCC grading systems in non-rhabdomyosarcoma soft tissue sarcomas
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
Asset Metadata
Creator
Zhang, Fan
(author)
Core Title
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
School
Keck School of Medicine
Degree
Master of Science
Degree Program
Biostatistics
Publication Date
02/10/2018
Defense Date
01/25/2018
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Adolescents,ARST0332,Children,distant metastases,distant metastatic disease,EFS,event-free survival,lymph node metastases,mortality,nonrhabdomyosarcoma,non-rhabdomyosarcoma,nonrhabdomyosarcoma soft tissue sarcomas,NRSTS,OAI-PMH Harvest,prognostic features,soft tissue sarcomas,Survival,tumor invasiveness,unresectable tumor
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Barkauskas, Donald (
committee chair
), Franklin, Meredith (
committee member
), Lane, Christianne Joy (
committee member
), Mack, Wendy Jean (
committee member
)
Creator Email
ffzhang@childrensoncologygroup.org,zhan541@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-473277
Unique identifier
UC11268459
Identifier
etd-ZhangFan-6025.pdf (filename),usctheses-c40-473277 (legacy record id)
Legacy Identifier
etd-ZhangFan-6025.pdf
Dmrecord
473277
Document Type
Thesis
Rights
Zhang, Fan
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
Tags
ARST0332
distant metastases
distant metastatic disease
EFS
event-free survival
lymph node metastases
mortality
nonrhabdomyosarcoma
non-rhabdomyosarcoma
nonrhabdomyosarcoma soft tissue sarcomas
NRSTS
prognostic features
soft tissue sarcomas
tumor invasiveness
unresectable tumor