University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
Explore Risk and Protective Factors of Undifferentiated Embryonal Sarcoma of the Liver

by

Zhongjie Cai

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 2016

Table of Contents
Abstract
Introduction
Methods
Results
Discussion
Reference
Appendix

i
1
2
4
6
8
9

i
Abstract
Background: Undifferentiated embryonal sarcoma of the liver (UESL) was recognized as a
distinct entity in 1978. ARST0332 is the first U.S. prospective study to enroll more than a small
handful of patients with UESL, and to gather centrally-reviewed data on imaging and pathologic
response to treatment. The aim of this paper is to explore risk and protective factors related to
event-free survival and overall survival.
Methods: Kaplan-Meier (KM) curves were constructed to visualize event-free survival and
overall survival. Fisher’s Exact Test was performed to evaluate relationship between various
patient and tumor characteristics. Cox proportional hazards models were used to explore
significant prognostic factors that related to the event-free survival.
Results: Except for covariates that are built upon one another (such as risk level and metastasis),
only the association between Tumor Invasiveness and Surgical Resection Margin was found
marginally significant (p=0.077). Due to the very small number of events, we failed to find any
risk or protective factors associated with event-free survival.
Conclusion: No significant risk or protective factors were found associated with event-free
survival. The very high percentage of UESL patients who are disease free and alive shows that
the treatments applied were highly successful.
1

Introduction
Non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) include an assorted group of tumors
accounting for 4-7% of all childhood malignancies and around 50% of all sarcomas in children.
NRSTS is estimated to affect approximately 500 children under 20 in the U.S. annually [1, 2].
Due to the very low incidence of different subtypes, clinical trials have not been conducted on
one single tumor type and NRSTS have only been analyzed as a whole group.
In this study, we performed analysis on patients with undifferentiated embryonal sarcoma of the
liver (UESL) using data obtained from study ARST0332 of the Children’s Oncology Group
(COG). UESL constitutes less than 5% of all malignant hepatic tumors among children [3, 4].
UESL was first described by Stocker and Ishak in 1978 as a distinct entity and was characterized
as an aggressive tumor with a very poor prognosis [5]. One prospective study held by the Italian
and German Soft Tissue Sarcoma Cooperative Groups showed that over 50% of patients with
UESL treated with multimodality therapy survived, suggesting huge potential of combination
therapy [6].
ARST0332 is the first U.S. prospective therapeutic study to enroll relatively a large number of
patients with centrally-reviewed data on imaging and pathologic response. The treatment strategy
in this study has proven successful because of the high percentage of patients who are alive and
event-free.
Our aim is to do a descriptive study for these patients with UESL about their demographics and
treatment received. We would also like to determine factors that associate with event-free
survival.

2

Methods
Patients
The patient population comprises of 39 children with UESL. They were aged below 30 at the
time of the biopsy that established the diagnosis of UESL and without infantile fibrosarcoma if
under the age of 2. The patient must have enrolled on the COG study D9902 in order to confirm
the diagnosis by central pathology review. The enrollment on D9902 must have occurred either
earlier than or on the same date as the enrollment on ARST0332.
Risk Classification and Treatment Schema
The risk of patients’ UESL was determined based on the current known prognostic factors with
higher risk representing lower expected event-free survival. The detailed break-up is shown in
the flow chart below [5].

3

Treatment Strategy
All of the UESL patients were assigned into two treatment groups: Treatment Arm C and
Treatment Arm D. Major difference between the two treatment arms are the order of therapies.
Treatment Arm C gives adjuvant chemotherapy followed by radiation therapy while Treatment
Arm D gives neoadjuvant chemotherapy.
Prognostic Factors to Evaluate
A list of covariates of interest was provided by the Study Chair, Sherri Sherri SpundSpunt, MD,
and their influences on outcome were evaluated. These characteristics include disease extent
(metastatic vs. non-metastatic), maximum diameter of the tumor (≤5cm vs. >5cm), tumor
invasiveness (non-invasive vs. invasive), extent of resection of the primary tumor and metastases,
microscopic surgical margins, and histologic tumor grade. Three records exist for tumor diameter,
tumor invasiveness, and surgical margin: Central Imaging Review, Central Surgical Review, and
On Study. Central Imaging Review is of top priority. Central Surgical Review is used if Central
Imaging Review has missing or indeterminate data. Only if both Central Imaging Review and
Central Surgical Review are missing or indeterminate is On Study used because On Study data is
site-reported. The two tumor grading systems used are the Pediatric Oncology Group (POG)
Non-rhabdomyosarcomatous Soft Tissue Sarcoma Grading System and the French Federation of
Cancer Centers (FNCLCC) Sarcoma Group Grading System. The FNCLCC system uses grades
1 through 3 and is an indicator for local aggressiveness and distant metastasis for soft tissue
sarcomas. The POG system is based on previous systems developed for adults but accounts for
pathologic entities only seen in young children [5].
Analysis
4

Fisher’s Exact Test was used to compare prognostic factors between UESL and non-UESL
groups and within the UESL group. A demographics table was created to outline patient
characteristics. Kaplan-Meier (KM) curves were constructed to visualize event-free survival and
overall survival. Variances for the pointwise survival estimates were calculated using
Greenwood’s formula. Cox proportional hazards models were used to explore significant
prognostic factors that related to the event-free survival. Event-free survival is defined as the
length of time from enrollment to an event (recurrence, secondary malignancy, or death)
occurrence or last date seen (censored). A stepwise model selection was performed using SAS,
with Breslow’s method to handle tied event times. Although suggested, stratification by patients’
risk level was not performed due to sparse patient with high risk. A two-sided Type I error rate
of 5% is employed throughout the study.

Results
One covariate, Surgical Resection Margin, was left out because it contains over 50% of missing
data. Looking at all patients with UESL, we have approximately equal number of male and
female patients. The majority of the patients are white, non-Hispanic, and aged 6-16. Patients
with UESL all have tumors with POG grade 3 and mostly FNCLCC grade 3. Mostly sized 10-
20cm in diameter, the majority of the tumors are non-metastatic, invasive and of intermediate
risk classification. The patient characteristics and tumor characteristics are shown in Table 1a
and Table 1b respectively.
We did pairwise comparisons for the variables listed above between the two groups using
Fisher’s Exact test (Table 1a and Table 1b). We find significant association between UESL
status and age group (p=0.0012), POG (p<0.0001) and FNCLCC (p<0.0001) tumor grade, tumor
5

diameter (p<0.0001), risk group (p<0.0001), treatment arm (p<0.0001), and tumor invasiveness
(p<0.0001).
The comparison group, which comprises of all other patients enrolled on ARST0332, has very
different tumor and patient characteristics. The age group distribution is roughly bell-shaped
with mode at 11-16. Females take a larger percentage in this group compared to the UESL group.
White and non-Hispanic are still dominant despite the slight decrease in percentage compared to
the UESL group. Regarding the tumor grade (high vs. low), the POG grade and FNCLCC grade
have a much larger difference: roughly 60% in comparison group vs. 13% in UESL group are
low grade. There is also a decreasing trend in the number of patients in the tumor diameter
groups, with 206 (40%) having tumor diameter less than or equal to 5cm and only 8 (2%) having
tumor diameter 20-25cm. Approximately 15% patients have metastases at diagnosis, and by
definition these patients are the high risk group. Among the remaining 85%, they lie in low and
intermediate risk groups almost equally. Around 80% patients received Treatment Arm A or
Treatment Arm D.
Another pairwise comparison was used within the group of UESL patients. As Tumor POG
histological grade was 3 for all patients, this variable was not included in the comparison.
Therefore a total of 10 variables were tested: Age Group, Gender, Race, Ethnicity, Tumor
FNCLCC Histologic Grade, Tumor Invasiveness, Metastatic Status, Risk Group, Treatment Arm,
and Tumor Diameter. The results are shown in Table 2. We find the following
significant/marginally significant associations: Race and Ethnicity (p=0.001), Risk Group and
Tumor Diameter (p=0.016), and Treatment Arm and Primary Tumor Resectability (p<0.001).
KM curves were used to visualize event-free survival and overall survival. Patients with
metastatic UESL have an event-free survival rate of 50% (95% CI: 0.6% - 91.0%) at 4.5 years
6

and patients with non-metastatic UESL have an event-free survival rate of approximately 80%
(95% CI: 62.3% - 89.9%) at 3.1 years (Figure 1). Both of these event-free survival rates are
higher than the corresponding rates among non-UESL patients who have 16% (95% CI: 7.5% -
26.2%) at 6.0 years and 72% (95% CI: 66.3% - 77.0%) at 6.0 years for non-metastatic and
metastatic patients, respectively. It should be noted that the estimate for metastatic UESL
patients is questionable as it is based on 3 patients with 1 event. We also find that the difference
between the four curves is significant (p<0.0001). The difference for overall survival rate is
much larger (Figure 2) and we also find significant difference between the survival curves of the
four groups (p<0.0001).
We explored potential risk/protective factors using the Cox Proportional Hazards Model. Due to
a small number of deaths in the UESL population, we were only able to fit models for event-free
survival. Univariate analysis suggests that only Tumor Diameter has a marginally significant
association with event-free survival (Table 4a and Table 4b). We failed to identify any risk or
protective factors.

Discussion
Using Fisher’s Exact Test we find that UESL status is associated with Age Group, Tumor POG
Histologic Grade, Tumor FNCLCC Histologic Grade, Tumor Diameter, Risk Group, Treatment
Arm, and Tumor Invasiveness (p<0.05). This suggests that, except for the variables listed above,
we have similar data distribution for the other variables in the UESL and non-UESL groups.
The pairwise Fisher’s Exact Test for variables within the UESL group reveals the following
significant/marginally significant associations:
7

 Race and Ethnicity (p=0.001).
 Risk Group and Tumor Diameter (p=0.016). Tumor diameter 10-15cm only exists in the
intermediate risk group. 16 other cases in the intermediate risk group scatter in the 15-
20cm and 20-25cm groups with13 in the former and 3 in the latter. The last 3 cases are in
the high-risk group with one 15-20cm case and two 20-25cm cases (Table 3a).
 Treatment Arm and Primary Tumor Resectability (p<0.001). All but two treatments are
consistent with the protocol with 22 received Treatment Arm C and 15 received
Treatment Arm D (Table 3b).
This gives us a list of variables to look for potential confounders when we do the model fitting
afterward.
Cox Proportional Hazards Model was fitted and none of the variables had a significant
association with event-free survival.
The very high percentage of UESL patients who are disease free and alive shows that treatments
of adjuvant chemotherapy followed by radiation therapy and neoadjuvant chemotherapy are
highly successful.

8

Reference
1. Gurney JG, Young JL, Roffers SD, Smith MA, Bunin GR. Soft Tissue Sarcomas. In: Ries
LA, Smith MA, Gurney JG, Linet M, Tamra T, Young JL, editors. Cancer incidence and
survival among children and adolescents: United States SEER Program 1975-1995,
National Cancer Institute, SEER Program. NIH Pub. No. 99-4649. Bethesda, Maryland;
1999. p. 111-24.
2. Ries LA, Smith MA, Gurney JG, et al: Cancer incidence and survival among children and
adolescents: United States SEER Program 1975-1995. Bethesda, National Cancer
Institute, SEER Program. NIH Pub. No. 99-4649, 1999
3. Stocker JT, Ishak KG. Undifferentiated (embryonal) sarcoma of the liver: report of 31
cases. Cancer 1978;42:336-48
4. Lack EE, Schloo BL, Azumi N, et al. Undifferentiated (embryonal) sarcoma of the liver.
Clinical and pathologic study of 16 cases with emphasis on immunohistochemical
features. Am J Surg Pathol 1991; 15:1-16.
5. Spunt S, et al: ARST0332: Risk-Based Treatment for Non-Rhabdomyosarcoma Soft
Tissue Sarcomas (NRSTS) in Patients Under 30 Years of Age. Version Date: 04/13/11
Amendment #3, downloaded from http://members.childrensoncologygroup.org on
12/2/15
6. Bisogno G, Pilz T, Perilongo G, et al.: Undifferentiated sarcoma of the liver in childhood:
a curable disease. Cancer 94 (1): 252-7, 2002
9

Appendix
Table 1a: Patient demographics
UESL Non-UESL Fisher's Exact p
Variables Frequency Percentage Frequency Percentage
Age Group
<6 7 17.95% 63 12.30%
0.0012
6-11 17 43.59% 90 17.58%
11-16 11 28.21% 204 39.84%
16-19 2 5.13% 96 18.75%
19-22 0 0.00% 32 6.25%
>22 2 5.13% 27 5.27%
Gender
Male 19 48.72% 237 46.29%
0.868
Female 20 51.28% 275 53.71%
Race
White 24 61.54% 366 71.48%
0.3829
Black or African American 8 20.51% 75 14.65%
Asian 1 2.56% 17 3.32%
American Indian or Alaska 1 2.56% 5 0.98%
Unknown 5 12.82% 49 9.57%
Ethnicity
Not Hispanic or Latino 34 87.18% 414 80.86%
0.5593 Hispanic or Latino 5 12.82% 78 15.23%
Unknown 0 0.00% 20 3.91%

10

Table 1b: Tumor characteristics
UESL Non-UESL Fisher's Exact p
Variables Frequency Percentage Frequency Percentage
POG Grade
1 0 0.00% 59 11.52%
<.0001
2 0 0.00% 93 18.16%
3 39 100.00% 357 69.73%
Indeterminate 0 0.00% 3 0.59%
FNCLCC Grade
1 0 0.00% 79 15.43%
<.0001
2 5 12.82% 217 42.38%
3 34 87.18% 208 40.63%
Indeterminate 0 0.00% 8 1.56%
Tumor Diameter
≤5cm 0 0.00% 206 40.23%
<.0001
5-10cm 0 0.00% 178 34.77%
10-15cm 20 51.28% 87 16.99%
15-20cm 14 35.90% 33 6.45%
20-25cm 5 12.82% 8 1.56%
Metastasis
No 36 92.31% 433 84.57%
0.2461
Yes 3 7.69% 79 15.43%
Risk
Low 0 0.00% 234 45.70%
<.0001 Intermediate 36 92.31% 199 38.87%
High 3 7.69% 79 15.43%
Treatment Arm
A 0 0.00% 212 41.41%
<.0001
B 0 0.00% 19 3.71%
C 23 58.97% 97 18.95%
D 16 41.03% 184 35.94%
Resection Margin
Negative 0 0.00% 1 0.69%
0.3039 Positive 1 7.69% 5 3.45%
Unknown 12 92.31% 139 95.86%
Tumor
Unresectable

No 23 58.97% 336 65.63%
0.3902
Yes 16 41.03% 176 34.38%
Tumor Invasiveness
Non-invasive 4 10.26% 228 44.53%
<.0001 Invasive 35 89.74% 278 54.30%
Indeterminate 0 0.00% 6 1.17%
11

Table 2: Pairwise Fisher’s Exact test between factors of patient with UESL
p-values Gender
Age
Group
Race Ethnicity
Tumor
FNCLCC
Histologic
Grade
Tumor
Diameter
Metastatic
Status
Risk
Group
Treatment
Arm
Primary
Tumor
Resectability
Tumor
Invasiveness
Gender 1.000
Age Group 0.226 1.000
Race 0.503 0.666
1.00
0
Ethnicity 1.000 0.818
0.00
1 1.000
Tumor
FNCLCC
Histologic
Grade 0.661 0.473
0.38
0 0.517 1.000
Tumor
Diameter 0.744 0.265
0.65
6 0.670 1.000 1.000
Metastatic
Status 0.605 0.370
1.00
0 0.345 1.000 0.016 1.000
Risk Group 0.605 0.370
1.00
0 0.345 1.000 0.016 0.000 1.000
Treatment
Arm 1.000 0.935
0.37
0 1.000 0.139 0.811 1.000 1.000 1.000
Primary
Tumor
Resectability 1.000 0.935
0.83
9 1.000 0.139 0.309 1.000 1.000 <0.001 1.000
Tumor
Invasiveness 0.342 0.271
1.00
0 1.000 1.000 0.596 0.284 0.284 0.130 0.130 1.000
12

Table 3a: Risk Group vs. Tumor Diameter contingency table
Fisher’s Exact Test
p=0.016 Tumor Diameter
Risk Group 10-15cm 15-20cm 20-25cm
Intermediate 20 13 3
High 0 1 2

Table 3b: Treatment Arm vs. Primary Tumor Resectability contingency table
Fisher’s Exact Test
p<0.001
Primary Tumor
Resectability
Treatment Arm 0 1
C 22 1
D 1 15

13

Table 4a: Hazard Ratio and 95% CI for univariate association between patient demographics and
event-free survival
Variables n
Hazard
ratio 95% CI P value Overall p
Age Group
<6 7 0 0 . 0.9945
0.4406
6-11 17 0.126 0.012 1.358 0.0876
11-16 11 0.099 0.008 1.278 0.0763
16-19 2 0.283 0.015 5.229 0.3966
>22 2 1
Gender
Male 19 1
0.8837
Female 20 0.902 0.225 3.608 0.8837
Race
White 24 1
0.1916
Black or African American 8 5.282 0.882 31.633 0.0684
Asian 1 0 0 . 0.9951
American Indian or Alaska 1 15.924 1.379 183.886 0.0266
Unknown 5 5.592 0.784 39.874 0.0859
Ethnicity
Not Hispanic or Latino 34 1
0.254
Hispanic or Latino 5 2.544 0.511 12.666 0.254

14

Table 4b: Hazard Ratio and 95% CI for univariate association between tumor characteristics and
event-free survival via Cox Proportional Hazard Regression
Variables n
Hazard
ratio 95% CI P value Overall p
FNCLCC
2 5 0 0 . 0.9955
0.9955
3 34 1
Tumor Diameter
10-15cm 20 0.146 0.024 0.875 0.0352
0.0927 15-20cm 14 0.304 0.061 1.516 0.1464
20-25cm 5 1
Metastasis
No 36 0.668 0.082 5.435 0.7062
0.7062
Yes 3 1
Risk
Intermediate 36 1
0.7062
High 3 1.497 0.184 12.174 0.7062
Treatment Arm
C 23 1.995 0.402 9.888 0.3979
0.3979
D 16 1
Tumor Unresectable
No 23 2.157 0.435 10.694 0.3468
0.3468
Yes 16 1
Tumor Invasiveness
Non-invasive 4 1
0.1060
Invasive 35 0.265 0.053 1.325 0.1060

15

Figure 1: Kaplan-Meier curves for event-free survival

16

Figure 2: Kaplan-Meier curves for overall survival

Abstract (if available)

Abstract Background: Undifferentiated embryonal sarcoma of the liver (UESL) was recognized as a distinct entity in 1978. ARST0332 is the first U.S. prospective study to enroll more than a small handful of patients with UESL, and to gather centrally-reviewed data on imaging and pathologic response to treatment. The aim of this paper is to explore risk and protective factors related to event-free survival and overall survival. ❧ Methods: Kaplan-Meier (KM) curves were constructed to visualize event-free survival and overall survival. Fisher’s Exact Test was performed to evaluate relationship between various patient and tumor characteristics. Cox proportional hazards models were used to explore significant prognostic factors that related to the event-free survival. ❧ Results: Except for covariates that are built upon one another (such as risk level and metastasis), only the association between Tumor Invasiveness and Surgical Resection Margin was found marginally significant (p=0.077). Due to the very small number of events, we failed to find any risk or protective factors associated with event-free survival. ❧ Conclusion: No significant risk or protective factors were found associated with event-free survival. The very high percentage of UESL patients who are disease free and alive shows that the treatments applied were highly successful.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

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

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

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

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

PDF

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Body wall primary tumors in non-rhabdomyosarcoma soft tissue sarcoma

PDF

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

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

The longitudinal risk factors of diabetic retinopathy: the Los Angeles Latino Eye Study

PDF

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

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

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

PDF

Infants in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Comparative study of laparoscopy vs. laparotomy for ovarian mass removal

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

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

PDF

A cohort study of air-pollution and childhood obesity incidence

PDF

Association of bone density and bone metabolism markers: the Women’s Isoflavone Soy Health (WISH) trial

PDF

Factors influencing the decision and timing to undergo breast reconstruction after mastectomy for breast cancer in public hospital vs. private medical center from 2007 to 2013: a retrospective co...

Asset Metadata

Creator Cai, Zhongjie (author)

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

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Publication Date 04/04/2016

Defense Date 03/23/2016

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

Tag explore,OAI-PMH Harvest,UESL

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Barkauskas, Donald (committee chair), Azen, Stanley (committee member), Krailo, Mark (committee member)

Creator Email terry.zhongjiecai@gmail.com,zhongjic@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-224401

Unique identifier UC11277126

Identifier etd-CaiZhongji-4225.pdf (filename),usctheses-c40-224401 (legacy record id)

Legacy Identifier etd-CaiZhongji-4225.pdf

Dmrecord 224401

Document Type Thesis

Format application/pdf (imt)

Rights Cai, Zhongjie

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