University of Southern California Dissertations and Theses

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

(USC Thesis Other)

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Lymph Node Metastases in Non-Rhabdomyosarcoma Soft Tissue Sarcoma

By

Jiayi He

A Thesis Presented to the
Faculty of Graduate School
University of Southern California
In Partial Fulfillment of the
Requirement for the Degree of
Master of Science
In Applied Biostatistics and Epidemiology

May 2016

Copyright 2016 Jiayi He

TABLE OF CONTANTS
DEDICATION ----------------------------------------------------------------------------- iii
ACKNOWLEDGEMENT ---------------------------------------------------------------- iv
ABSTRACT ------------------------------------------------------------------------------ v-vi
INTRODUCTION AND BACKGROUND ----------------------------------------------1
METHODS -----------------------------------------------------------------------------------5
RESULTS -------------------------------------------------------------------------------------6
DISCUSSION ------------------------------------------------------------------------------10
CONCLUSION -----------------------------------------------------------------------------11
REFERENCE -------------------------------------------------------------------------------12
TABLES -------------------------------------------------------------------------------------14
FIGURES ------------------------------------------------------------------------------------24

iii

DEDICATION
I would like to dedicate my thesis to my beloved parents, who supported me each
step of the way to my education.

iv

ACKNOWLEDGEMENT
I would like to thanks all the people who helped and supported me in the
completion of this research project.

First, I would like to express my deepest appreciation to my committee chair Dr.
Donald Barkauskas, who has been guiding me through this educational journey.
Without his guidance this research project would not have been possible.

Second, I would like to thank my committee member Dr. Todd A. Alonzo and Dr.
Christianne J. Lane for their expertise and guidance.

Lastly, I would like to thank the entire faculty and employee from the Department
of Preventive Medicine for creating an amazing educational experience at the
University of Southern California.

v

ABSTRACT
Background
Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma (NRSTS) are uncommon
in pediatric patients. Previous studies suggested that the pattern of lymph node involvement in
pediatric patients was similar to adult patients in NRSTS, indicating that patients with epithelioid
sarcoma and clear cell sarcoma have increased incidence of lymph node involvement at initial
diagnosis. Hence, in order to further understand the pattern of lymph node involvement among
pediatric patients with NRSTS, we analyzed data from a large cohort study (ARST0322), which
represented the largest prospectively treated cohort of pediatric NRSTS patients in the U.S. It
offered us opportunity to further study the characteristics and survival outcomes for NRSTS
pediatric patients.
Methods
The target population consisted of 551 patients aged less than 30 years old, all of whom had met
the eligibility requirements and received a novel risk-based treatment. Our study population was
the patients with lymph node metastases (in total 22 people). Fisher’s exact test and two-sample
t-test were used to describe demographics and tumor clinical features between lymph node
metastases patients and non-lymph node metastases patients. Kaplan–Meier survival plots with
log-log transformed confidence intervals and log-rank test stratified by presence or absence of
distant metastases were used to evaluate event-free survival and overall survival. Finally, Cox
proportional-hazard regression was performed to select potential risk factors for EFS and OS for
lymph node metastases patients.

vi

Results
Synovial sarcoma (6 cases) and epithelioid sarcoma (5 cases) were the most frequent histotypes
among NRSTS pediatric patients with lymph node involvement. Lymph node metastases
pediatric patients had significantly larger maximum diameter tumor size [mean (SD) 11.2 (6.3)
cm] compared to patients without lymph node involvement [7.6 (5.3) cm, Two-Sample T-Test:
p=0.0020]. During mean 4.6 (1.7) years of follow up time, the 5-year event-free survival (EFS)
and overall survival (OS) for pediatric patients without lymph node metastases and distant
metastases were 75.8% ± 2.1% and 86.4% ± 1.8%, respectively. For pediatric patients with both
lymph node metastases and distant metastases, the EFS and OS were both 16.7% ± 10.8%. We
found presence of distant metastases at diagnosis was a strong risk factor for event-free survival
and overall survival among lymph node pediatric patients, and presence of lymph node
metastases at diagnosis was an independent risk factor from presence of distant metastases for
overall survival among all patients.
Conclusions
Pediatric patients with NRSTS lymph node involvement had different tumor size and survival
outcomes compared to patients without lymph node metastases. Moreover, presence or absence
of lymph node metastases only showed significant difference in overall survival among all
pediatric patients. We suggest lymph node involvement play an important role in the failure of
overall survival even in the presence of distant metastases.
Key words
Pediatric, Non-rhabdomyosarcoma soft tissue sarcoma (NRSTS), Lymph node metastases,
Distant metastases, Tumor grade, Histotype, Event-free survival, Overall survival.

1

INTRODUCTION AND BACKGROUND
The non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) are a heterogeneous group of
mesenchymal cell neoplasms, most of which are named for the mature tissue that the tumor most
resembled.
[1]
They comprise about 4% of childhood malignancies and affect approximately 500
children under the age of 20 years in the United States each year.
[2]
However, these tumors
among pediatric patients have lowest survival outcome and have not been studied systematically
to date, with only three prospective clinical trials with fewer than 200 total patient enrollments
having been conducted in the U.S.
[3-6]
Hence, efforts to evaluate the characteristics and survival
outcomes among pediatric patients with NRSTS are essential and promising, since it may give
insight into improving the survival outcome and quality for NRSTS pediatric patients.
Furthermore, study of metastatic involvement in pediatric NRSTS is important since patients
with metastases in NRSTS, which included both lymph node metastases and distant metastases,
are considered a high risk group due to their poor survival outcome compared to other NRSTS
patients in previous studies.
[3, 7]
What is more, traditional radiotherapy and chemotherapy
treatment for NRSTS pediatric patients have generated large amount of incurable serious side
effect, and few studies have been published for pediatric patients with lymph node metastases,
only within one study suggesting that pediatric patients with lymph node involvement presented
with different histologic subtype sarcoma compared to adults.
[7, 8, 9]
Hence, efforts to identify a
novel treatment for pediatric NRSTS patients to limit the toxicity and maximize the efficacy of
treatment delivered are necessary. Our study, a phase III clinical study with novel risk-based
treatment for pediatric NRSTS patients, offer us a good opportunity to further understand the
characteristics, survival outcome as well as the efficacy of the novel risk-based treatment for
NRSTS pediatric patients with lymph node metastases.
2

We analyzed data from ARST0332, a group-wide phase III study from Children’s Oncology
Group, which was a novel risk-based treatment study for NRSTS in patients under 30 years of
age and representing the largest prospectively treated cohort of pediatric NRSTS patients in the
U.S. Our aim was to describe the demographics, clinical features, and survival outcome for
NRSTS pediatric patients, identify potential risk predictors for event-free survival, overall
survival, local recurrence and distant recurrence for pediatric NRSTS patients with lymph node
metastases.
Study design and participants
The primary aim of this study was to evaluate the survival outcome of the novel risk-based
strategy among pediatric NRSTS patients, with dual goals of limiting the toxicity of therapy for
low-risk patients and maximizing the efficacy of therapy for intermediate-risk and high-risk
patients.
[10]
Eligibility

was defined as patients aged < 30 years at the time of the biopsy that established
the diagnosis of an eligible NRSTS (see protocol for list of eligible types
[10]
), with gross total
resection of the primary tumor when possible and no prior anthracycline (e.g., doxorubicin,
daunorubicin) or ifosfamide chemotherapy. Moreover, patients should have had no radiotherapy
to tumor-involved sites prior to enrollment, and should have a life expectancy of at least 3
months and meet organ function requirements described in protocol according to their relevant
therapy assigned.
[10]

3

Histologic Grading Criteria
Tumor grade was assigned according to the POG (Pediatric Oncology Group) system, which
takes into account tumor necrosis, mitotic activity, cellularity, and histology. All pathologic
material was centrally reviewed by three pathologists.
[11]
Risk Group Assessment and Treatment Assignment
A recent retrospective analysis of pediatric patients treated at St. Jude Children’s Research
Hospital was used as a reference for risk classification.
[12]
This study suggested that patients
with grossly resected non-metastatic tumors with low clinical histologic grade (POG grade 1 or 2)
and tumor maximum diameter ( ≤ 5cm) were defined as low-risk group and should receive
surgery with or without adjuvant radiotherapy for the goal to diminish the likelihood of late side
effects (in total, 212 patents on treatment A and 19 patients on treatment B). Patients with
initially un-resected non-metastatic tumors with clinical high histologic grade (POG grade 3) and
tumor maximum diameter (> 5cm) were defined as intermediate-risk group and received
adjuvant chemotherapy and radiotherapy (in total, 120 patients in treatment C). Patients with
metastatic tumors were defined as high-risk group and they should receive neo-adjuvant
chemotherapy and radiotherapy (in total, 220 patients in treatment D) based on definitive tumor
resection for the goal to improve the response rate and improve survival outcome (Figure 1,
Table 2).
[13]

Demographics Assessment

Patient characteristics included age at enrollment that would be treated as categorical
variable (<5, 6-10, 11-15, 16-18, 19-21, 22-30 years) and continuous variable for further
understanding on the distribution of age, gender treated as binary variable (male, female), race
4

treated as categorized variable (white, Black or African American, Asian, American Indian, or
Alaska) and ethnicity treated as binary variable (Hispanic or Latino, not Hispanic or Latino).
Tumor Feature Assessment
Tumor information was collected for eligible patients after enrollment from physical
examination, diagnostic imaging, biopsy of sites and operative data, which included as described
in Table 1.
Table 1.Items for tumor feature assessment
Locations of tumor to a primary site
Disease extent depending on metastatic status at diagnosis
Distant metastases anatomic sites
Tumor maximum diameter size (<5, 5.1–10, 10.1-15, >15 cm)
Tumor depth (superficial or deep)
Tumor invasiveness (invasive, non-invasive, indeterminate)
Tumor histologic POG grade
Tumor histologic French Federation of Cancer Centers Sarcoma Group (FNCLCC) grade
Pathologic diagnosis of tumor

Outcome Assessment
Event-free survival time is defined as the time period from the date of enrollment to the date
of relapse, second malignancy, death or last follow-up time (censored); overall survival time is
5

defined as the time period from the date of enrollment to the date of death or last follow-up time
(censored).
METHODS
Statistical Analysis
Mean (standard deviation) and proportions were used to describe characteristics for continuous
and categorized variables, respectively among all patients and lymph node metastases-only
patients. Bivariate comparison was conducted to study clinical feature distribution among lymph
node metastases-only patients. Fisher’s exact test and Two-sample t-tests were used to test for
significant difference for categorical and continuous variables between patents with or without
lymph node metastases, respectively using p-value under the significance level of 0.05. Kaplan–
Meier survival plots with log-log transformation to obtain the pointwise confidence intervals for
the survivor function in addition to the confidence intervals for the quartiles of the survival times,
[14]
and log-rank test stratified by presence or absence of distant metastases were used to evaluate
event-free survival and overall survival among all patients.
[15]
Univariate Cox proportional-
hazard model was performed to select potential risk factors for event-free survival, overall
survival, local and distant recurrence among patients with lymph node metastases. Cox
proportional-hazard model using stepwise selection with both stay and entry significant level of
0.05, and the covariates with univariate p-values no larger than 0.20
[10]
as potential model
components were used to find predictors. All data were analyzed using Statistical Analysis
System software version 9.4 (SAS institute, Inc., Cary, North Carolina).

6

RESULTS
Characteristics of Study Population
Table 2 shows the characteristics of patients enrolled on ARST0332. A total of 588 patients
participated in this study and 37 patients were excluded due to the failure to meet all eligibility
requirements, resulting in a total of 551 patients included in the final analysis. Patients were
divided into two groups depending on their lymph node metastases status. A total of 22 patients
with lymph node metastases aged mean 13.4 (6.4) years with primarily 68.2 % (15) white and
81.2% (18) not Hispanic/Latino. There are 529 patients without lymph node metastases aged
mean 13.1 (5.5) years with mainly 70.9% (375) white and 81.3% (430) not Hispanic/Latino. No
statistically significant difference for age at enrollment between these two groups was found
(p=0.79 as continuous variable and p=0.75 as categorical variable). In contrast, a statistically
significant difference was found in the size of primary tumor (p=0.002 for continuous variable
and p=0.003 for categorized variable), mean 11.2 (6.3) cm for lymph node metastases patients,
and mean 7.6 (5.3) cm for non-lymph node metastases patients. No significant difference was
found in gender, race, ethnicity, depth of primary tumor, histologic POG grade, histologic
FNCLCC grade, and invasiveness between these two groups. In addition, as for the treatment
arm assignment, lymph node metastases patients with the majority of having un-resected tumor
before enrollment (72.7%) were mainly assigned treatment arm D (72.7%); while most non-
lymph node metastases patients mainly had resected tumor before enrollment (66.7%) were
assigned treatment arm A (39.9%) . A statistically significant difference was found between
these two groups for treatment arm delivered (p=0.0006) and presence of un-resected tumor
before enrollment (p=0.0003).
7

Among 22 Lymph node metastases patients, we were able to identify significant differences
between patients with distant metastases (N=12) compared to those without (N=10) at
enrollment (p=0.01, Table 3). Patients with distant metastases aged mean 17.0 (4.7) years and
patients without distant metastases aged mean 9.0 (5.5) years. Moreover, in Table 4, FNCLCC
grade 2 were predominately presented in males while FNCLCC grade 3 mainly presented in
females (p=0.02). As for invasiveness, it was predominately presented in patients with larger
primary tumor size (10.1-15cm and >15.1 cm; p=0.008), patients with un-resected tumor (p=0.01)
before enrollment and patients treated with treatment arm D (p=0.01) respectively.
In Table 5 we provide the clinical feature distribution for pediatric patients with lymph node
metastases. Patients had distant metastases in both lungs (8), Liver (3), Bone (2), Mesentery (1)
and Pancreas (1). As for the primary sites of tumor, we could find in Pelvis (4), Shoulder (1),
Upper arm (2), Lower arm (1), Hand (1), Hip (1), Thigh (3), Leg (1), Foot (1), Neck (1), Chest
wall – anterior (1), Chest wall – posterior (1), Intraperitoneal (2) and Retroperitoneal (2). The
distribution of histologic diagnosis was as follows: 6 Synovial sarcoma, 5 Epithelioid sarcoma, 3
unclassified soft tissue sarcoma, 3 undifferentiated sarcoma, 1 Angiosarcoma of soft tissue, 1
Parachordoma, 1 Malignant peripheral nerve sheath tumor, 1 Clear cell sarcoma of soft tissue
and 1 High grade sarcoma.
Event-Free Survival and Overall Survival
For all eligible patients in event-free survival (Figure 2) and overall survival (Figure 3), with
mean 4.6 (1.7) years follow up time for all patients, the best survival was seen for patients
without either lymph node metastases or distant metastases (giving 5-year event-free survival
75.8% ± 2.1% and 5-year overall survival 86.4% ± 1.8%), while the worst was seen for patients
8

with both lymph node metastases and distant metastases (giving 5-year event-free survival 16.7%
± 10.8% and 5-year overall survival 16.7% ± 10.8%).
Log-rank tests were performed for event-free survival (χ
2
=2.27, df=1, p-value=0.13) and
overall survival χ
2
=4.61, df=1, p-value=0.03) among all patients for the effect of lymph node
metastases stratified by the presence or absence of distant metastases, which provided evidence
that lymph node metastases involvement stratified by distant metastases status did not show
significant difference in event-free survival but did show significant difference in overall
survival. However, if we tested the overall survival for lymph node metastases patients in distant
metastases and non-distant metastases groups separately, no statistical difference was found in
each strata (Figure 4, N=479, χ
2
=2.70, df =1, p=0.10 for non-distant metastases group; Figure 5,
N=72, χ
2
=2.62, df=1, p-value=0.11 for distant metastases group).
Potential Predictors
Univariate Cox proportional hazard models for event-free survival, overall survival and
distant recurrence were performed to explore potential risk factors among 22 lymph node
metastases patients (table 6, table 7 and table 8). We discovered that only distant metastases was
statistically significant for event-free survival failure and overall survival failure separately, no
significant potential risk factor was found for distant recurrence. Moreover, since
positive/negative margin had large amount of missing data (N=197 missing data) and the sample
size distribution for depth of primary tumor was extreme (N=21 for deep tumor and N=1 for
indeterminate tumor), we did not include them in the univariate model analysis. Moreover, only
15 Lymph node metastases patients were found to have relapse among our 22 patients, with 1
patient having had local recurrence and 14 having had distant recurrence, no predictive model
9

was performed for local recurrence due to limit sample size. In addition, since there was no
significant variable found under the significant p-value of 0.05 in the stepwise selection for
distance recurrence, predictive model for distant recurrence was not presented here. Hence,
variables with p-values no larger than 0.20 from univariate model would be selected as potential
risk predictors in the final multivariate stepwise Cox proportional-hazard model for event-free
survival failure and overall survival failure, which included categorized age at enrollment, un-
resected tumor before enrollment, and distant metastases. Finally, a Cox proportional-hazard
model among all patients with lymph node and distant metastases as independent variables
would be performed to further understand the role of lymph node and distant metastases in
overall survival and event-free survival.
Based on the final results in table 9, we conclude that the risk factor for event-free survival
among 22 lymph node metastases patients is distant metastases, indicating the risk of failure in
event-free survival is 4.11 (95% CI: 1.28, 13.24) times higher for patients with distant metastases
compared to patients without distant metastases. Furthermore, distant metastases is also a
significant risk factor for overall survival among 22 lymph node metastases patients, the risk of
failure in overall survival is 4.40 (95% CI: 1.21, 16.00) times higher for patients with distant
metastases compared to patients without metastases. Among all 551 patients, we see the risk of
failure in overall survival is 2.00 (95% CI: 1.09, 3.54) times higher for patients with lymph node
metastases compared to patients without lymph node metastases, and 7.40 (95% CI: 4.97, 11.00)
times higher for patients with distant metastases compared to patients without distant metastases
respectively. However, the risk of failure in event-free survival does not show significant
difference between patients with or without lymph node metastases (p=0.13, Table 9), which is
consistent with the results in our stratified log-rank test (Figure 4, 5).
10

DISCUSSION
The results of our study indicated that pediatric patients with NRSTS lymph node
involvement did not show any significant difference in age, gender, race, ethnicity, tumor depth
and histologic grade compared to non-lymph node NRSTS patients, but they had larger
maximum tumor diameter size and presented with more cases for presenting distant metastases,
having un-resected tumor before enrollment and treating with arm C and D.
Amongst the 22 Lymph node metastases patients, we could see only age at enrollment was
found to have significantly difference between distant metastases and non-distant metastases
patients, assuming age played an important role in the presence of distant metastases. Moreover,
as for the major clinical features among these 22 lymph node metastases patients, we found they
were mainly female with higher FNCLCC grade (3). Furthermore, most of them had larger tumor
size (10.1-15 cm, >15.1 cm), had un-resected tumor before enrollment and treatment arm
assigned C and D with invasiveness respectively. All of these findings suggesting larger tumor
size, higher FNCLCC tumor grade, invasiveness, deep tumor depth and un-resected tumor before
enrollment played important role in the presence of lymph node involvement. Lastly, synovial
sarcoma and epithelioid sarcoma were the most frequent prognostic histotypes among lymph
node NRSTS pediatric patients in this study.
Our data indicates pediatric patients had poor outcome with lymph node involvement, as
well as had more cases of distant metastases and distant recurrence. Moreover, lymph node
metastases was a strong risk factor for overall survival failure (Figure 3, Table 8), even though
with or without lymph node metastases in each strata of presence or absence of distant
11

metastases did not directly show a significant difference, it would be most likely due to small the
sample size in each strata.
Hence, our findings were consistent with the results in previous study, which suggesting the
identification of NRSTS that was associated with lymph node involvement at diagnosis would
highly correlated with the presence of distant metastases.
[16]
The reason would be for NRSTS
patients with lymph node metastases, grade 3 histologic grades (POG, FNCLLC) and larger
maximum tumor size were most frequently presented; both of these elements would mostly
likely leading to the spread of distant metastases.
[17]

CONCLUSION
In summary, our study demonstrated that pediatric patients with NRSTS lymph node
involvement had different clinical features and survival outcomes compared to patients that did
not have lymph node involvement. Moreover, lymph node metastases was a strong risk factor for
overall survival along with the presence of distant metastases. However, since our findings were
just generated from a small number of lymph node metastases patients in a non-randomized trial
during mean 5 years of follow up, we still need more evidence and information to make firm
conclusion by considering a larger randomized trial with the novel risk-based treatment arms in
order to achieve better understanding of the clinical pathologic features and survival outcome for
these patients.

12

REFERENCE
1. Sheri L. Spunt, Stephen X. Skapek, et al: Pediatric non-rhabdomyosarcoma soft tissue
sarcomas: The Oncologist. 13: 668-678, 2008
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. Pratt CB, Maurer HM, Gieser P, et al: Treatment of unresectable or metastatic pediatric soft
tissue sarcomas with surgery, irradiation, and chemotherapy: a Pediatric Oncology Group study.
Med Pediatr Oncol 30:201-9, 1998
4. Pratt CB, Pappo AS, Gieser P, et al: Role of adjuvant chemotherapy in the treatment of
surgically resected pediatric nonrhabdomyosarcomatous soft tissue sarcomas: A Pediatric
Oncology Group Study. J Clin Oncol 17:1219, 1999
5. Pappo AS, Devidas M, Jenkins J, et al: Vincristine (V), ifosfamide (I), doxorubicin (D), and
GCSF (G) for pediatric unresected metastatic non-rhabdomyosarcomatous soft tissue sarcomas
(NRSTS): a Pediatric Oncology Group study. Proc Am Soc Clin Oncol 20:378a, 2001
6. Bleyer A, Montello M, Budd T, et al: Young adults with sarcoma: Lack of clinical trial
participation and lack of survival prolongation. Proc Am Soc Clin Oncol 22:816, 2003
7. Pappo AS, Rao BN, Jenkins JJ, et al: Metastatic nonrhabdomyosarcomatous soft-tissue
sarcomas in children and adolescents: the St. Jude Children's Research Hospital experience. Med
Pediatr Oncol 33:76-82, 1999
8. Pratt CB, Meyer WH, Jenkins JJ, et al: Ifosfamide, Fanconi's syndrome, and rickets. J Clin
Oncol 9:1495-9, 1991
9. Ferrari A, Casanova M, Collini P, et al. (2005) Adult-type soft tissue sarcomas in pediatric-age
patients: Experience at the Istituto Nazionale Tumori in Milan. J Clin Oncol 23:4021–4030.
10. Sheri L. Spunt, et al: Protocol of Risk-Based Treatment for Non-Rhabdomyosarcoma Soft
Tissue Sarcomas (NRSTS) in Patients under 30 Years of Age. Children’s Oncology Group
(ARST0332).
11. Parham DM, Webber BL, Jenkins JJ, et al: Nonrhabdomyosarcomatous soft tissue sarcomas
of childhood: Formulation of a simplified system for grading. Mod Pathol 8:705-710, 1995
12. Spunt SL, Poquette CA, Hurt YS, 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. J Clin Oncol 17:3697-705, 1999
13

13. Kraybill WG, Harris JH, Spiro IJ, et al: Phase II study of neoadjuvant chemotherapy and
radiation therapy in the management of high-risk, high-grade, soft tissue sarcomas of the
extremities and body wall: Radiation Therapy Oncology Group (RTOG) 9514. J Clin Oncol
24:619-25, 2006
14. Lachin, J. M. (2000). Biostatistical Methods: The Assessment of Relative Risks. New York:
John Wiley & Sons.
15. Kalbfleisch JD and Prentice RL: The statistical analysis of failure time data. John Wiley and
Sons. New York, 2002.
16. Federica De Corti, Patrizia Dall'Igna, et al: Sentinel node biopsy in pediatric soft tissue
sarcomas of extremities. Pediatric Blood Cancer. 2009;52:51-54.
17. Ferrari, A., Brecht, I. B., Koscielniak, E., Casanova, M., Scagnellato, A., Bisogno, G.,
Alaggio, R., Cecchetto, G., Catania, S., Meazza, C., Int-Veen, C., Kirsch, S., Dantonello, T.,
Carli, M. and Treuner, J. (2005), The role of adjuvant chemotherapy in children and adolescents
with surgically resected, high-risk adult-type soft tissue sarcomas. Pediatr. Blood Cancer, 45:
128–134.

14

TABLES
Table 2.Descriptive characteristic with Mean (SD) and N (%) for patients and tumors among all
patients with and without lymph node metastases by using Two-sample t-test and Fisher’s
Exact Test
Characteristics Lymph node
Metastases
(N=22)
Non-Lymph
node
Metastases
(N=529)
p-values
Follow-up time for patients alive at last
contact (years)
4.6 (1.7) —
5.3 (1.4) 4.6 (1.7) 0.25
Age at enrollment (years) 13.4 (6.4) 13.1 (5.5) 0.79
Categorical of age enrollment (years) 0.35
0-10 8 (36.4%) 169 (32.0%)
11-18 10 (45.5%) 303 (57.3%)
19-30 4 (18.2%) 57 (10.75%)
Gender 1.00
Male 10 (45.5%) 246 (46.5%)
Female 12 (54.5%) 283 (53.5%)
Race 0.65
White 15 (68.2 %) 375 (70.9%)
Black or African American 3 (13.6 %) 80 (15.1%)
Asian — 18 (3.4%)
American Indian or Alaska — 6 (1.1%)
Unknown 4 (18.2%) 50 (9.5%)
Ethnicity 0.90
Hispanic or Latino 4 (18.2%) 79 (14.9%)
Not Hispanic or Latino 18 (81.2%) 430 (81.3%)
Unknown — 20 (3.8%)
Primary tumor size (cm) 11.2 (6.3) 7.6 (5.3) 0.002*
Categorization of primary tumor size (cm) 0.003*
0-5 4 (18.2 %) 202 (38.2%)
5.1-10 5 (22.7 %) 173 (32.7%)
10.1-15 5 (22.7 %) 102 (19.3%)
>15.1 8 (36.4%) 52 (9.8%)
Tumor POG histologic grade 0.21
1 — 59 (11.2%)
2 2 (9.1%) 91 (17.2%)
15

3 20 (90.9%) 376 (71.1%)
Indeterminate — 3 (0.5%)
Tumor FNCLCC histologic grade 0.07
1 — 79 (14.9%)
2 7 (31.8%) 215 (40.7%)
3 15 (68.2%) 227 (42.9%)
Indeterminate — 8 (1.5%)
Primary tumor depth 0.14
Superficial — 67 (12.7%)
Deep 21 (95.4%) 413 (78.1%)
Indeterminate 1 (4.6%) 49(9.3%)
Invasiveness 0.74
Non-invasive 8 (36.4%) 224 (42.4%)
Invasive 14 (63.6%) 299 (56.5%)
Indeterminate — 6 (1.1%)
Distant metastases <.0001*
Yes 12 (54.5%) 60 (11.3%)
No 10 (45.5%) 469 (88.7%)
Treatment Arm 0.0006*
A: No adjuvant treatment 1 (4.6%) 211 (39.9%)
B: Adjuvant radiotherapy — 19 (3.6%)
C: Adjuvant chemotherapy + radiotherapy 5 (22.7%) 115 (21.7 %)
D: Neoadjuvant chemo-radiotherapy 16 (72.7%) 184 (34.8%)
Un-resected Tumor 0.0003*
Yes 16 (72.7%) 176 (33.3%)
No 6 (27.3%) 353 (66.7%)

16

Table 3.Descriptive characteristics with Mean (SD) and N (%) for patients and tumors among
lymph node metastases patients with and without distant metastases by using Two-sample t-
test and Fisher’s Exact Test

Characteristics Distant
Metastases
(N=12)
No distant
Metastases
(N=10)
p-values
Age at enrollment (years) 17.0 (4.7) 9.0 (5.5) 0.001*
Categorical of age enrollment (years) 0.05
0-10 2 (16.7%) 6 (60.0%)
11-18 6 (50.0%) 4 (40.0%)
19-30 4 (33.3%) —
Gender
Male 7 (58.3%) 7 (70.0%) 0.23
Female 5 (41.7%) 3 (30.0%)
Race 0.44
White 9 (75.0%) 6 (60.0%)
Black or African American 2 (16.7%) 1 (10.0%)
Asian — —
American Indian or Alaska — —
Unknown 1 (8.3%) 3 (30.0%)
Ethnicity 1.00
Hispanic or Latino 2 (16.7%) 2 (20%)
Not Hispanic or Latino 10 (83.3%) 8 (80%)
Primary tumor size (cm) 13.4 (6.3) 8.6 (5.6) 0.08
Categorization of primary tumor size (cm) 0.45
0-5 1 (8.3%) 3 (30.0%)
5.1-10 2 (16.7%) 3 (30.0%)
10.1-15 3 (25.0%) 2 (20.0%)
>15.1 6 (50.0%) 2 (20.0%)
Tumor POG histologic grade 1.00
1 — —
2 1 (8.3%) 1 (10.0%)
3 11 (91.7) 9 (90.0%)
Tumor FNCLCC histologic grade 1.00
1 — —
2 4 (33.3%) 3 (30.0%)
3 8 (67.7%) 7 (70.0%)
Primary tumor depth 0.45
Superficial — —
Deep 12 (100.0%) 9 (90.0%)
Indeterminate — 1 (10.0%)
17

Invasiveness 1.00
Non-invasive 4 (33.3%) 4 (40.0%)
Invasive 8 (66.7%) 6 (60.0%)
Treatment Arm 0.80
A: No adjuvant treatment — 1 (10.0%)
B: Adjuvant radiotherapy — —
C: Adjuvant chemotherapy + radiotherapy 3 (25.0%) 2 (20.0%)
D: Neoadjuvant chemo-radiotherapy 9 (75.0%) 7 (70.0%)
Un-resected Tumor 1.00
Yes 3 (25.0%) 3 (30.0%)
No 9 (75.0%) 7 (70.0%)

18

Table 4.Bivariate comparison of variables among lymph node metastases patients by using
Fisher’s Exact Test with p-values equal or less than 0.05

Comparison variable 1 Comparison variable 2 p-values
Tumor FNCLCC histologic grade 0.02
2 3
Gender Male 6 (85.7%) 4 (26.7%)
Female 1 (14.3%) 11 (73.3%)
Invasiveness
Non-invasive Invasive 0.007
Categorization of
primary tumor size
(cm)
0-5 3 (37.5%) 1 (7.1%)
5.1-10 4 (50.0%) 1 (7.1%)
10.1-15 — 5 (35.7%)
>15.1 1 (12.5%) 7 (50.0%)
Un-resected Tumor Yes 3 (37.5%) 13 (92.9%) 0.01
No 5 (62.5%) 1 (7.1%)
Treatment
Arm
A: No adjuvant
treatment
1 (12.5%) — 0.01
C: Adjuvant
chemotherapy +
radiotherapy
4 (50.0%) 1 (7.1%)
D: Neoadjuvant
chemo-
radiotherapy
3 (37.5%) 13 (92.9%)

19

Table 5.Diagnosis and locations of tumor and metastases (N %)

Diagnosis and locations Lymph node
Metastases
(N=22)
Anatomic locations of distant metastatic disease
Both lungs 8 (36.4%)
Bone 2 (9.1%)
Liver 3 (13.6%)
Mesentery 1 (4.6%)
Pancreas 1(4.6%)
Patients with multiple distant sites 3 (13.6%)
Site of primary tumor
Pelvis 4(18.2%)
Shoulder 1 (4.6%)
Upper arm 2 (9.1%)
Lower arm 1 (4.6%)
Hand 1 (4.6%)
Hip 1 (4.6%)
Thigh 3 (13.6%)
Leg 1 (4.6%)
Foot 1 (4.6%)
Neck 1 (4.6%)
Chest wall - anterior 1 (4.6%)
Chest wall - posterior 1 (4.6%)
Intraperitoneal 2 (9.1%)
Retroperitoneal 2 (9.1%)
Histologic diagnosis
Angiosarcoma of soft tissue 1 (4.6%)
Parachordoma 1 (4.6%)
Synovial sarcoma 6 (27.3%)
Epithelioid sarcoma 5 (22.7%)
Clear cell sarcoma of soft tissue 1 (4.6%)
Malignant peripheral nerve sheath tumor 1 (4.6%)
Unclassified soft tissue sarcoma 3 (13.6%)
Other eligible diagnosis (High grade sarcoma) 1 (4.6%)
Undifferentiated sarcoma 3 (13.6%)
*N (%) was used to describe locations (site of primary tumor and anatomic site of distant metastases) and
diagnosis (histologic) among patients had lymph node metastases.

20

Table 6.Univariate Cox proportional hazard model for event-free survival among lymph node
metastases patients

Variables Sample
size (N)
Parameter
Estimate
(SE)
Hazard Ratio (95%
CI)
Wald χ2 df p-value
Categorized
age enrollment
(years)
Baseline (0-10) 8 — — 5.20 2 0.07*
11-19 10 1.52 (0.69) 4.65 (1.21, 17.87)
19-30 4 1.44(0.79) 4.20 (0.89, 19.85)
Gender Baseline (male) 10 — — 0.0002 1 0.99
female 12 -0.007
(0.507)
0.99 (0.37, 2.68)
Race Baseline
(White)
15 — — 0.08 2 0.96
Black or African
American
3

0.21 (0.78) 1.23 (0.26, 5.70)
Unknown 4 -0.03 (0.67) 0.97 (0.26, 3.60)
Ethnicity Baseline
(Hispanic or
Latino)
4 — — 0.88 1 0.35
Not Hispanic or
Latino
18

-0.56 (0.59) 0.57 (0.18, 1.84)
Categorized
tumor size
(cm)
Baseline (< 5) 4 — — 1.28 3 0.73
5.1-10 5 -0.57 (0.83) 0.56 (0.11, 2.86)
10.1-15 5 -0.73 (0.85) 0.48 (0.09, 2.55)
>15.1 8 -0.04 (0.71) 0.96 (0.24, 3.89)
Histologic POG
grade
Baseline (Grade
2)
2 — — 0.21 1 0.64
Grade 3 20 0.48 (1.04) 1.62 (0.21, 12.41)
Histologic
FNCLCC grade
Baseline (Grade
2)
7 — — 0.10 1 0.76
Grade 3 15 0.18 (0.59) 1.20 (0.38, 3.84)
Un-resected
tumor
Baseline
(resected)
6

— — 1.71 1 0.19*
Un-resected 16 0.86 (0.66) 2.36 (0.65, 8.53)
Distant
Metastases
Baseline (No) 10 — — 5.62 1 0.02**
Yes 12 1.41 (0.60) 4.11 (1.28, 13.24)
Invasiveness Baseline (Non-
invasive)
8

— — 0.06 1 0.81
Invasive 14 0.13 (0.55) 1.14 (0.39, 3.35)
**refer to p-value less than 0.05; * refer to p-value less than 0.20.

21

Table 7.Univariate cox proportional hazard model for overall survival among lymph node
metastases patients

Variables Sample
size (N)
Parameter
Estimate
(SE)
Hazard Ratio
(95% CI)
Wald χ2 df p-value
Categorized
age
enrollment
(years)
Baseline (0-10) 8 — — 5.92 2 0.05*
11-19 10 1.93 (0.79) 6.87 (1.45, 32.46)
19-30 4 1.62 (0.92) 5.08 (0.84, 30.80)
Gender Baseline (male) 10 — — 0.02 1 0.89
female 12 -0.078
(0.560)
0.93 (0.31, 2.77)
Race Baseline
(White)
15 — — 0.24 2 0.89
Black or African
American
3

0.25 (0.79) 1.28 (0.28, 5.97)
Unknown 4 -0.25 (0.78) 0.78 (0.17, 3.61)
Ethnicity Baseline
(Hispanic or
Latino)
4 — — 0.52 1 0.47
Not Hispanic or
Latino
18

-0.48 (0.66) 0.62 (0.17, 2.27)
Categorized
tumor size
(cm)
Baseline (< 5) 4 — — 1.31 3 0.73
5.1-10 5 -0.03 (0.93) 0.97 (0.16, 6.0)
10.1-15 5 0.07 (0.93) 1.08 (0.17, 6.73)
>15.1 8 0.65 (0.82) 1.92 (0.38, 9.59)
Histologic POG
grade
Baseline (Grade
2)
2 — — 0.11 1 0.74
Grade 3 20 0.35 (1.04) 1.42 (0.18, 10.94)
Histologic
FNCLCC grade
Baseline (Grade
2)
7 — — 0.02 1 0.87
Grade 3 15 0.10 (0.60) 1.10 (0.34, 3.59)
Un-resected
tumor
Baseline
(resected)
6

— — 2.35 1 0.13*
Un-resected 16 1.18 (0.77) 3.27 (0.72, 14.83)
Distant
Metastases
Baseline (No) 10 — — 5.05 1 0.02**
Yes 12 1.48 (0.66) 4.40 (1.21, 15.99)
Invasiveness Baseline (Non-
invasive)
8

— — 0.15 1 0.70
Invasive 14 0.21 (0.56) 1.24 (0.41, 3.71)
**refer to p-value less than 0.05; * refer to p-value less than 0.20.
22

Table 8.Univariate cox proportional hazard model for distant recurrence among lymph node
metastases patients

Variables Sample
size
Parameter
Estimate
(SE)
Hazard Ratio (95%
CI)
Wald χ2 df p-value
Categorized
age enrollment
(years)
Baseline (0-10) 8 — — 2.87 2 0.24
11-19 10 1.78 (1.08) 5.95 (0.72, 49.49)
19-30 4 1.86 (1.19) 6.44 (0.63, 66.16)
Gender Baseline (male) 10 — — 1.93 1 0.17
female 12 -0.92 (0.66) 0.40 (0.11, 1.46)
Race Baseline
(White)
15 — — 1.32 2 0.52
Black or African
American
3

0.61 (0.83) 1.84 (0.36, 9.34)
Unknown 4 -0.61 (0.80) 0.54 (0.12, 2.56)
Ethnicity Baseline
(Hispanic or
Latino)
4 — — 0.34 1 0.56
Not Hispanic or
Latino
18

0.39 (0.66) 1.48 (0.40, 5.42)
Categorized
tumor size
(cm)
Baseline (< 5) 4 1.72 3 0.63
5.1-10 5 -0.37 (0.92) 0.69 (0.11, 4.15)
10.1-15 5 — —
>15.1 8 0.29 (0.97) 1.34 (0.20, 8.99)
Histologic POG
grade
Baseline (Grade
2)
2 — — 0.02 1 0.88
Grade 3 20 -0.16 (1.06) 0.85 (0.11, 6.88)
Histologic
FNCLCC grade
Baseline (Grade
2)
7 — — 0.65 1 0.4194
Grade 3 15 -0.53 (0.65) 0.59 (0.17, 2.12)
Un-resected
tumor
Baseline
(resected)
6

— — 1.96 1 0.19*
Un-resected 16 1.09 (0.78) 2.98 (0.65, 13.77)
Distant
Metastases
Baseline (No) 10 — — 0.13 1 0.71
Yes 12 0.25 (0.67) 1.28 (0.35, 4.74)
Invasiveness Baseline (Non-
invasive)
8

— — 0.37 1 0.54
Invasive 14 0.37 (0.61) 1.45 (0.44, 4.75)
**refer to p-value less than 0.05; * refer to p-value less than 0.20.
23

Table 9.Estimate predictors among lymph node metastases patients and all patients

Study
population
Outcome Variables Parameter Estimate
(SE)
p-values Hazard Ratio
(95%CI)
1
Patients
with lymph
node
metastases
(N=22)

Event-free
survival
Distant metastases 1.41 (0.60) 0.02 4.11 (1.28, 13.24)
Overall
survival
Distant metastases 1.48 (0.66) 0.02 4.40 (1.21, 16.00)
2
All
patients
(N=551)
Event-free
survival
Distant metastases 1.86 (0.17) <.0001 6.40 (4.60, 8.92)
Lymph node
metastases
0.42 (0.28) 0.13 1.52 (0.89, 2.62)
Overall
survival
Distant metastases 2.00 (0.20) <.0001 7.40 (4.97, 11.00)
Lymph node
metastases
0.67 (0.30) 0.03 1.96 (1.09, 3.54)
1
The proportional hazards regression stepwise selection process with significant level of 0.05 for both entry and
stay chose a model with distant metastases among lymph node metastases patients(with total 22 cases within 16
events and 6 censored).
2
The proportional hazards regression used overall survival as dependent variable, distant metastases and Lymph
node metastases as independent variables among all patients, no statistically significance was found for the
interaction term of these two variables (p=0.6054).

24

FIGURES
Figure 1.Diagnosis of classifying three level risk groups and treatment arms (From ARST0332
Protocol document, Sheri L. Spunt, Study Chair)

25

Figure 2.Estimated event free survival for all patients by sites of metastases

Test χ2 df p-value
Stratified Log-Rank 2.27 1 0.13

26

Figure 3.Estimated overall survival for all patients by sites of metastases

Test χ2 df p-value
Stratified Log-Rank 4.61 1 0.03

27

Figure 4.Estimated overall survival for lymph node metastases patients without distant metastases

Test χ2 df p-value
Log-Rank 2.70 1 0.10

28

Figure 5.Estimated overall survival for lymph node metastases patients with distant metastases

Test χ2 df p-value
Log-Rank 2.62 1 0.11

Abstract (if available)

Abstract Background: Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma (NRSTS) are uncommon in pediatric patients. Previous studies suggested that the pattern of lymph node involvement in pediatric patients was similar to adult patients in NRSTS, indicating that patients with epithelioid sarcoma and clear cell sarcoma have increased incidence of lymph node involvement at initial diagnosis. Hence, in order to further understand the pattern of lymph node involvement among pediatric patients with NRSTS, we analyzed data from a large cohort study (ARST0322), which represented the largest prospectively treated cohort of pediatric NRSTS patients in the U.S. It offered us opportunity to further study the characteristics and survival outcomes for NRSTS pediatric patients. ❧ Methods: The target population consisted of 551 patients aged less than 30 years old, all of whom had met the eligibility requirements and received a novel risk-based treatment. Our study population was the patients with lymph node metastases (in total 22 people). Fisher’s exact test and two-sample t-test were used to describe demographics and tumor clinical features between lymph node metastases patients and non-lymph node metastases patients. Kaplan–Meier survival plots with log-log transformed confidence intervals and log-rank test stratified by presence or absence of distant metastases were used to evaluate event-free survival and overall survival. Finally, Cox proportional-hazard regression was performed to select potential risk factors for EFS and OS for lymph node metastases patients. ❧ Results: Synovial sarcoma (6 cases) and epithelioid sarcoma (5 cases) were the most frequent histotypes among NRSTS pediatric patients with lymph node involvement. Lymph node metastases pediatric patients had significantly larger maximum diameter tumor size [mean (SD) 11.2 (6.3) cm] compared to patients without lymph node involvement [7.6 (5.3) cm, Two-Sample T-Test: p=0.0020]. During mean 4.6 (1.7) years of follow up time, the 5-year event-free survival (EFS) and overall survival (OS) for pediatric patients without lymph node metastases and distant metastases were 75.8% ± 2.1% and 86.4% ± 1.8%, respectively. For pediatric patients with both lymph node metastases and distant metastases, the EFS and OS were both 16.7% ± 10.8%. We found presence of distant metastases at diagnosis was a strong risk factor for event-free survival and overall survival among lymph node pediatric patients, and presence of lymph node metastases at diagnosis was an independent risk factor from presence of distant metastases for overall survival among all patients. ❧ Conclusions: Pediatric patients with NRSTS lymph node involvement had different tumor size and survival outcomes compared to patients without lymph node metastases. Moreover, presence or absence of lymph node metastases only showed significant difference in overall survival among all pediatric patients. We suggest lymph node involvement play an important role in the failure of overall survival even in the presence of distant metastases.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

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

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

PDF

Infants in non-rhabdomyosarcoma soft tissue sarcoma

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

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

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

PDF

Construction of a surgical survival prediction model of stage IV NSCLC patients-based on seer database

PDF

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

Asset Metadata

Creator He, Jiayi (author)

Core Title Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 04/20/2016

Defense Date 04/19/2016

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

Tag distant metastases,event-free survival,histotype,lymph node metastases,non-rhabdomyosarcoma soft tissue sarcoma,NRSTS,OAI-PMH Harvest,overall survival,pediatric,tumor grade

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Barkauskas, Donald (committee chair), Alonzo, Todd A. (committee member), Lane, Christianne J. (committee member)

Creator Email jiayih@usc.edu,jiayihe28@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-234289

Unique identifier UC11279365

Identifier etd-HeJiayi-4306.pdf (filename),usctheses-c40-234289 (legacy record id)

Legacy Identifier etd-HeJiayi-4306.pdf

Dmrecord 234289

Document Type Thesis

Format application/pdf (imt)

Rights He, Jiayi

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