Improved care of rhabdomyosarcoma in Jordan using less intensive therapy

Authors


  • Conflict of interest: Nothing to declare.

Abstract

Background

The care of rhabdomyosarcoma (RMS) is complex due to its multimodal nature. By following standard protocols with acceptable toxicity and building local expertise, better outcome should be achievable.

Procedure

A retrospective study was conducted of records of patients (n = 45; 31 males; median age 26 months) with RMS treated at King Hussein Cancer Center in Jordan from January 2004 to December 2008. Patient demographics, tumor characteristics, risk stratification, treatment plan, and outcomes were studied. In June 2006, the cyclophosphamide dose was lowered from 2.2 g/m2 to 1.2 g/m2 per cycle because of the significant toxicity with higher dose. Survival rates, hematological toxicities, period of hospitalization due to febrile neutropenia (FN), and response rate at week 12 of treatment were compared between low- and high-dose cyclophosphamide groups.

Results

Four-year progression-free survival (PFS) and overall survival (OS) rates were 61% ± 7.5% and 72% ± 6.9%, respectively. There was a significant difference in outcome by risk group in 4-year PFS (low-risk, 88% ± 12%; intermediate-risk 63% ± 9.3%; high-risk, 14% ± 13%; P = 0.0001) and OS (low-risk, 88% ± 12%; intermediate-risk 79% ± 7.5%; high-risk, 17% ± 15%; P = 0.0011). There was significant reduction in hematological toxicities, incidence of FN, and period of hospitalization for FN in patients given low-dose cyclophosphamide but no significant difference in PFS between low- and high-dose cyclophosphamide groups.

Conclusions

Survival rates of patients with RMS in some developing countries can be improved by following or modifying evidence-based approaches successful in developed countries and establishing multidisciplinary strategies. Therapy intensity should be increased in developing countries only when evidence supports its utility. Pediatr Blood Cancer 2013; 60: 53–58. © 2012 Wiley Periodicals, Inc.

INTRODUCTION

Rhabdomyosarcoma (RMS), the most common soft tissue sarcoma in children, originates from mesenchymal tissues at various anatomic locations 1. Overall survival (OS) rates for children with RMS have improved significantly over the last 30 years 2, primarily due to advances in multimodal therapy. All treatment strategies incorporate systematic multiagent chemotherapy and local therapy 3, 4. Chemotherapy for RMS typically comprises alkylating agents (ifosfamide or cyclophosphamide) at varying dosages, depending on regimen. Although surgical resection, either as a primary procedure or a secondary strategy, is important for local control, complete resection is not feasible for many tumors (e.g., parameningeal RMS) 4. Radiotherapy (RT) plays a major role in the management of many patients with RMS although its late sequelae can be significant 5, 6.

Despite improved outcomes of children with RMS in developed countries, survival rates of patients in limited-resource countries continue to remain poor 7. Obstacles to the management of RMS in limited-resource nations include complexity of treatment, cost of therapy, and families' compliance 7. At the King Hussein Cancer Center (KHCC), children with RMS are managed by a multidisciplinary team (comprising of pediatric oncologists, pediatric surgeons, radiation oncologists, radiologists, pathologists, and psycho-social team). Patients are treated according to internationally accepted protocols (Children's Oncology Group COG protocol), their progress is reviewed, and strategies constantly modified to achieve improved outcomes. Of note, KHCC receives approximately 70% of children with cancer registered in Jordan cancer registry. It is the only comprehensive cancer center in Jordan. In this study, we conducted a 5-year retrospective chart review of the clinical characteristics, multiagent treatment plan, and outcomes of children with RMS treated at KHCC to evaluate how the change in protocol affected outcomes/toxicity and how a multimodal approach worked well for a center in a developing nation.

MATERIALS AND METHODS

Patients

This study was approved by the institutional review board at KHCC. A retrospective review of the medical records of children with RMS (January 2004 to December 2008) who were younger than 18 years at the time of diagnosis was conducted. Data were retrieved from medical records and pathology databases.

Patients were classified by using a risk stratification system, which was adopted from that used by the Children Oncology Group (COG); it integrates post surgical grouping classification developed by the North American Intergroup Rhabdomyosarcoma Study Group IRSG (now known as the Soft Tissue Sarcoma Committee of the COG) 8, IRSG pretreatment staging classification (modified from a tumor-node-metastasis system) 9, and histology 3. According to the risk stratification system we used, patients were stratified into a low-risk (LR) group including those with nonmetastatic embryonal tumors (i.e., embryonal histology, stage I, post surgical group I, II, or III; or embryonal histology, stage I or II, post surgical group I or II); intermediate-risk (IR) group if they had nonmetastatic alveolar histology tumors (i.e., stage I, II, or III, post surgical group I, II, or III) or an unresected embryonal tumor at unfavorable sites (i.e., stage II or III, post surgical group III). Patients with high-risk (HR) disease had metastatic tumors (i.e., stage IV, post surgical group IV).

We reviewed patients' characteristics (demographics, tumor location, tumor size, regional lymph node involvement, pathology, grouping, staging, and risk stratification), treatment plan (chemotherapy agents, initial and second-look surgery, and RT), and outcomes 4-year progression-free survival (PFS) and OS.

Chemotherapy Protocol

The chemotherapy regimen was similar to that used in the Intergroup Rhabdomyosarcoma Study (IRS) and the Soft Tissue Sarcoma Subcommittee of the Children's Oncology Group (COG) studies. Before 2006, patients with RMS were treated according to the IRS IV protocol 3. In June 2006, a major modification was introduced to the protocol at KHCC—the dose of cyclophosphamide was lowered to 1.2 g/m2 because of the high toxicity associated with the VAC regimen at a cyclophosphamide dose of 2.2 g/m2/cycle. Grade III and IV hematological toxicities, incidences of febrile neutropenia (FN), and period of hospitalization during episodes of FN were compared in patients receiving 2.2 g/m2 or 1.2 g/m2 cyclophosphamide. The response rate at week 12 of treatment was also evaluated. Survival rates of both groups were determined.

Radiotherapy

Patients received RT according to their post surgical grouping classification, regional lymph node involvement, and initial histology. Patients with embryonal histology who had post surgical group I tumors received no RT. Patients with residual microscopic disease (post surgical group II) received 36 Gy if there was no regional lymph node involvement (N0) or 41.4 Gy when there was regional lymph node involvement (N1). Patients with post surgical group III tumors (N0 or N1) received 50.4 Gy to the non-orbital tumor bed and 45 Gy to the orbital tumor site. In patients with alveolar histology, those with post surgical group I or group II tumors with N0 received 36 Gy and those with post surgical group II tumors and N1 received 41.4 Gy. Patients with post surgical group III tumors received 45 Gy (for orbital tumor sites) or 50.4 Gy (for non-orbital tumor sites).

Prior to simulation and treatment, informed consent must be obtained and documented. An initial CT simulation was performed with a Philips Big Bore CT simulator (Philips Medical Systems, Andover, MA). Treatment planning was performed with a Pinnacle V9 (Philips Medical Systems). Treatment planning and delivery methods consisted of 3D-CRT with multiple static fields. Manual forward planning with multiple segments was intended to provide better coverage and dose homogeneity. RT was performed on Elekta Synergy linear accelerators using photon energies of 4, 6, 10, and 15 MV. In vivo dosimetry was done during the first session. Portal imaging of radiation fields was performed at the beginning of treatment and every 5–10 treatments.

Surgery

The goal of surgery was to achieve a wide and complete resection of the primary tumor with a surrounding envelope of normal tissue. Surgical resection was performed at the initial and/or subsequent (second-look surgery, usually after week 12 of treatment) operations wherever possible and reasonable. Complete resection was generally more applicable to sites in the extremities or trunk than in the head and neck and parameningeal areas. In sites where complete resection was not applicable or feasible, biopsy or subtotal resection (STR) was performed.

Radiology

All through the study period, patients were routinely imaged using a high quality MRI (AVANTO 1.5 and INGENIA 3/TESLA) and/or CT scan (VOLUME ZOOM 8 and BRILLIANCE 64/PHILLIPS). Radiologic images were interpreted by a pediatric radiologist who routinely attended the multidisciplinary clinic.

Pathology

All samples were reviewed in the center by a board certified pathologist. The following immunostains were used to diagnose RMS: desmin (clone DE-R-11, antigen retrieval by protease 1 for 8 minutes, Ventana, USA) and myogenin (clone F5D, ready to use, antigen retrieval by heat (CC1) 60 minutes, Ventana, USA). Diffuse staining for myogenin in >50% of tumor cells was supportive of the diagnosis of alveolar RMS. Molecular studies for t (2; 13) or t (1; 13) (Vysis LSI FOXO1 Dual Color Break Apart Probe) were used in cases in which staining was not conclusive for the diagnosis of alveolar RMS. The Vysis FISH probe used was a Break Apart probe that is directed at the FOXO1 gene on chromosome 13q14. This means that the hybridization will identify a chromosomal rearrangement at the FOXO1 gene but not a specific gene-fusion partner, which could be either PAX3 at chromosome 2 or PAX7 at chromosome 1.

Staging

For staging, an MRI and/or CT scan of the primary tumor, CT chest, bone scan, and bilateral bone marrow aspirates and biopsies were done. Assessment of the response was usually done at week 12 of chemotherapy.

Statistical Analyses

PFS was defined as the time from the date of diagnosis to disease recurrence or death as a first event. OS was defined as the time from date of diagnosis to death from any cause or lost to follow-up. The Kaplan–Meier method was used to estimate PFS and OS distributions. Differences between survival curves in stratified risk groups and in high- and low-dose cyclophosphamide groups were analyzed by the log-rank test 10. The statistical level with a P-value of ≤0.05 was considered significant.

RESULTS

Patients' Characteristics

Demographics

From January 2004 to December 2008, 45 patients with RMS were referred to KHCC. Another two patients were referred after progression, and were therefore excluded from analysis. There were 31 males (69%) and 14 females (31%) with a median age of 26 months (range 3–96 months). The median follow up was 39 months and the range was between 3 and 95 months.

Tumor characteristics

Of the 45 patients, 37 (82%) had embryonal histology, 7 (16%) had alveolar histology, and 1 (2%) had embryonal RMS with anaplasia. Eight (18%) patients had primary tumors at favorable sites. Tumor size was <5 cm in 18 (40%) patients. Only 6 (13%) patients had regional lymph node involvement. Seven (16%) patients had distant metastases at the time of diagnosis. Common sites of metastases were lung (n = 3), bone (n = 3), lymph nodes (n = 1), and brain (n = 1; Table I).

Table I. Patient Characteristics
 AllLow riskIntermediate riskHigh risk
Gender
 Male317231
 Female14176
Age (years)
 Median4.47.93.854.5
 Range0.3–17.81.4–13.40.5–17.80.3–12.7
Tumor site
 Favorable8620
 Un favorable372287
Tumor size
 <5 cm18792
 >5 cm271215
LN involvement
 Yes6132
 No397275
Distant metastases
 Yes7007
 No388300
Histology
 Embryonal378236
 Alveolar7061
Others1010
Post surgical group
 I2200
 II6420
 III302280
 IV7007
Staging
 18620
 28170
 3221210
 47007

Grouping and risk stratification

The majority of patients (n = 30; 67%) were classified as post surgical group III, and 8 as post surgical group I and II, because of grossly resected tumors (Table I). According to our risk stratification system, 8 (18%) patients were stratified as the low-risk (LR) group, with having nonmetastatic embryonal tumors. All the 30 (67%) intermediate-risk (IR) patients had nonmetastatic alveolar histology tumors and/or a grossly residual mass at unfavorable sites. Seven (16%) high-risk (HR) patients had metastatic tumors (Table I).

Treatment plan

A 3-drug regimen [vincristine, actinomycin, and cyclophosphamide (VAC)] was used in 44 (98%) patients and 2-drug chemotherapy [vincristine, actinomycin (VA)] was used in 1 (2%) patient.

Of the 40 (89%) patients who received RT, 38 received it to the local site of tumor, 1 to the regional lymph nodes after total resection of the primary tumor, and 1 received it both locally and to the regional lymph nodes. The remaining 5 (11%) patients did not receive RT (2 were in post surgical group I, families of 2 patients declined RT, and 1 died before receiving RT because of chemotherapy-related complications).

With regard to surgery, 38 (84%) patients underwent biopsy/STR, 3 (7%) had gross total resection (GTR) with histologically proven free margins, and 4 (9%) underwent GTR with microscopic involvement (i.e., positive margins). Second-look surgery was done in nine patients: one had debulking surgery, five had GTR, and three patients underwent only biopsy (one with orbital tumor that showed no viable tumor; one with retroperitoneal mass which revealed fibrosis; and one with bladder/prostate tumor, with histology revealing mature rhabdomyoblasts).

Outcomes: Total and by Risk Stratification

The 4-year PFS rate for patients was 61% ± 7.5% (Fig. 1A), and the 4-year OS rate was 72% ± 6.9% (Fig. 1B). The 4-year PFS rate according to risk stratification was as follows: LR group, 88% ± 12%; IR group, 63% ± 9.3%; and HR group, 14% ± 13%; this difference was significant (P = 0.0001; Fig. 1C). The 4-year OS according to risk stratification was as follows: LR group, 88% ± 12%; IR group 79% ± 7.5%; and HR group, 17% ± 15%; this difference was significant (P = 0.0011; Fig. 1D).

Figure 1.

Kaplan–Meier curves showing total outcomes and outcomes by risk stratification of patients (n = 45) with RMS. A: 4-year PFS (61% ± 7.5%); (B) 4-year OS (72% ± 6.9%); (C) 4-year PFS by risk stratification (P = 0.0001); (D) 4-year OS by risk stratification (P = 0.0011); log-rank test was used for analyzing the differences between survival curves. [Color figure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/journal/pbc]

Hematological Toxicity and Cyclophosphamide Dose–Related Outcomes

The dose of cyclophosphamide was reduced to 1.2 g/m2 for patients treated at our center from June 2006. We compared the hematological toxicities of patients in the IR and HR groups who received the high (2.2 g/m2) and low (1.2 g/m2) doses of cyclophosphamide. There was a significant difference in the incidence of grade IV neutropenia, grade III and IV thrombocytopenia, episodes of FN, and total days of hospitalization for FN between the groups (Table II). The 4-year PFS rates were similar for both groups (P = 0.085; Fig. 2A). The difference between groups for response to chemotherapy at week 12 was not significant (P = 0.36; Fig. 2B).

Table II. Comparing Patient Characteristics and Chemotherapy Toxicity of Patients Who Received 1.2 g/m2 Cyclophosphamide Versus Those Who Received 2.2 g/m2
 1.2 g/m22.2 g/m2P
  • a

    All low risk patients (N = 8) were excluded as they received no cyclophosphamide (N = 1) or only 4 cycles (N = 7);

  • b

    Median values are presented and compared using Mann–Whitney test.

Number of patientsa1720 
Age (years)b5.13.30.29
Follow-up (months)b34630.0052
 (Range 17–51)(Range 3–95) 
Tumor site (unfavorable)15200.39
Tumor size (>5 cm)11150.75
Histology (alveolar)430.41
Post surgical grouping   
 II110.59
 III1414 
 IV25 
Staging   
 1200.30
 243 
 3912 
 4 (metastatic)25 
Toxicity   
 Grade III neutropeniab110.73
 Grade IV neutropeniab250.0001
 Grade III thrombocytopeniab010.0005
 Grade IV thrombocytopeniab170.0035
 FN episodesb25<0.0001
 Total days of admissionb523<0.0001
Figure 2.

A: Kaplan–Meier curves comparing 4-year PFS of patients given 1.2 g/m2 or 2.2 g/m2 cyclophosphamide; (B) response evaluation at week 12 of chemotherapy for patients receiving 1.2 g/m2 or 2.2 g/m2 cyclophosphamide (P = 0.36); CR, complete response; VGPR, very good partial response; PR, partial response; SD, stable disease; PD, progressive disease; log-rank test was used for analyzing the differences between survival curves. [Color figure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/journal/pbc]

DISCUSSION

The treatment of children with cancer in developing countries faces many obstacles, such as patient and family compliance, cost, delayed diagnosis, low awareness of disease, and poor access to multimodal care 7. In this study, we report the outcomes of children with RMS in a single institution in Jordan. At KHCC, 70% of patients are Jordanians who get full financial coverage through the Ministry of Health. The non-Jordanian patients are either supported by referring hospitals outside Jordan or receive donations from the charity foundation (King Hussein Cancer Foundation). Although there was a higher male/female ratio in this study, this does not reflect the referral pattern at our institution. None of the patients with RMS treated at KHCC reported abandonment of care, which is similar to what has been reported for patients treated for other malignancies at the center 11–13. A multidisciplinary team comprising of pediatric oncologists, pediatric surgeons, radiation oncologists, radiologists, pathologists, and psycho-social team was formed in 2006 and meets biweekly to discuss the care of children with solid tumors and sarcoma.

In this study, the 4-year PFS (61% ± 7.5%) and OS (72% ± 6.9%) rates were similar to those reported previously for RMS 3, 9, 14–17. Moreover, there was significant association between risk groups and 4-year PFS and OS rates: patients with LR and IR group had better outcomes than those with HR diseases, which is consistent with results reported by collaborative groups on RMS 9, 15, 18–24. Seven (16%) patients had alveolar histology, which was lower than expected as population-based estimates of alveolar histology are around one third of all tumors; however, our study sample is small and it is hard to draw any conclusion from this finding.

Because of the high toxicity associated with the VAC regimen at a cyclophosphamide dose of 2.2 g/m2/cycle, we reduced the dose of cyclophosphamide to 1.2 g/m2 for patients treated after 2006.

The justification for this reduction was based on the published results of IRS-IV study. In this study, cyclophosphamide was given at a single dose of 2.2 g/m2/course (considered to be quite equivalent to 9 g/m2/course of ifosfamide used by the SIOP-MMT group, giving an ifosfamide to cyclophosphamide ratio of 4.3) 25; when compared with the results of the previous study (cyclophosphamide at 10 mg/kg/day for 3 days, approximately 1 g/m2/course), this intensification proved beneficial only in certain subgroups of intermediate-risk patients (i.e., Group III embryonal RMS of the head and neck region, initially-resected embryonal RMS of favorable sites), while it did not improve survival in unresectable embryonal RMS occurring at unfavorable sites and in alveolar RMS 9, 26. Our results showed similar survival rates in both groups but significantly less hematologic toxicity in patients who received low-dose cyclophosphamide. Also the similar response to induction chemotherapy at week 12 suggested that increasing the dose of cyclophosphamide does not significantly improve response, which is consistent with results of a previous study on cyclophosphamide dose intensification in pediatric RMS 27. Due to the lack of evidence for better results of higher doses of cyclophosphamide, we reduced the doses used in our protocol. With less toxicity, we also observed slight improvement in our results after using low-dose cyclophosphamide (although not statistically significant); this improvement may reflect the effect of multidisciplinary approach in the management of RMS practiced in our center.

Although the results of the Children Oncology Group (COG) set of studies using low dose of cyclophosphamide are not yet published 28, our results suggest that adoption of low-dose cyclophosphamide is advisable in developing countries where careful balance between treatment intensity and toxicity determines the ultimate outcome of patients. In these settings, a careful shift towards less toxic regimens should be implied.

In summary, the survival rates of children with RMS in developing countries can be improved if evidence-based approaches are adapted from developed countries and multidisciplinary strategies are established. Our results indicate that high-dose cyclophosphamide may not improve survival rates and is associated with a significant increase of toxicity in pediatric RMS. Increasing therapy intensity should be practiced with caution in developing countries unless clear evidence exists of its utility.

Acknowledgements

We acknowledge the assistance of Vani Shanker with preparation of the manuscript.

Ancillary