There is limited data on the outcome of this disease from developing countries, including India [4-6]. The multidisciplinary approach mandated for managing RMS is often a hindrance for orchestrated treatment, as well as data collection in pediatric oncology units in developing countries. Selected studies from India have reported sub-optimal survival (13-25%) [4,5]. The factors contributing to poor survival include, treatment abandonment, infection related mortality, lack of surgical expertise for local disease-control, and failure to deliver chemotherapy in a timely manner due to poor compliance [4-6]. The aim was to analyze the profile and outcome of children with RMS from a pediatric oncology unit. The purpose was to characterize the disease profile in our population and to identify hindrances for improving survival.

The study was performed in a government-run, tertiary-care, university hospital in Northern India. A retrospective analysis of case-records of children (age <12 years at diagnosis) with RMS presenting to the pediatric-oncology clinic over a 23-year (1990-2012) period was performed after Institutional Ethical Clearance. The patients were classified by the ‘Intergroup Rhabdomyosarcoma Study (IRS) Group’ staging and grouping system as low, intermediate or high-risk [7]. The IRS ‘group’ was determined after the initial surgical procedure, prior to systemic therapy and was based on the extent of residual tumor after surgery with consideration of regional lymph node involvement. The IRS ‘stage’ was based on tumor size, invasiveness, nodal status, and the site of primary tumor. The group and stage were both taken into consideration for the final risk stratification. The favorable sites included orbit, superficial head and neck, biliary tree, vagina, and para-testis; the remaining sites are considered unfavorable [7,8]. The tumor size was based on the largest dimension of the primary tumor reported in the pretreatment imaging. Nodal involvement was based on physical examination, imaging studies and/or node sampling at surgery. Pathology was classified as ‘embryonal,’ ‘alveolar,’ ‘pleomorphic’ or ‘not otherwise specified (NOS).’ Molecular tests were not available. Staging investigations included a computed tomography (CT) of the chest and a bone scan. Cerebrospinal fluid (CSF) was examined for para-meningeal tumors. A bone marrow examination was performed for noticeable hematological abnormality, or in the presence of other sites of metastasis. Congenital RMS was defined as presentation within the first month of life with history of onset of symptoms since birth [9].

Treatment included neo-adjuvant chemotherapy, followed by surgical excision (wherever feasible), radiotherapy (for residual disease, if any, and, for inoperable tumors), and adjuvant chemotherapy, based on the Children’s Oncology Group (COG) protocols [7]. Low-risk patients received vincristine and actinomycin-D for 22 weeks, while intermediate and high-risk groups were treated with vincristine, actinomycin-D, and cyclophos-phamide for the duration of 43-weeks. The specific high-risk protocol was not administered due to apprehension regarding toxicity. The response was evaluated by radiological imaging (CT/Magnetic Resonance Imaging) following week 9-12 of chemotherapy. Survival estimates were calculated using the Kaplan-Meier method. Analysis was performed with SPSS v20.0 (IBM).

The data of 159 patients was collected. The median age at presentation was 4 years (Interquartile range, IQR 2;7). There were 20 (13%) infants; 5 (3%) had congenital rhabdomyosarcoma. Nine (6%) children were >10 years at presentation. The male:female ratio was 2.7:1. The median symptom interval was 2 months (IQR 1;5). Head and neck region was the most frequent site (44%), followed by tumors in the extremity (15.7%) (Table I). Majority (67%) of the tumors were located at unfavorable sites. Size was documented in 101 patients; 69 (68%) were >5 cm. Nodal status was documented in 76 and was positive in 36 (47%). Thirteen (8%) had evidence of metastasis at diagnosis. The most frequent (58%) pathological subtype was embryonal. Information required for risk-stratification was available for 126; low-risk: 42 (33%), intermediate-risk: 70 (56%) and high-risk: 14 (11%) [7]. Of the 129 patients in whom treatment was initiated, surgery was performed in 63 (49%), while 54 (42%) received radiotherapy; 31 (24%) received both.

TABLE I	Details of Patients (N=159) with Rhabdomyosarcoma 

Caregivers of 28 (18%) patients refused treatment at the outset. In addition, 52 (33%) abandoned treatment at various phases of therapy. The remaining had a mean (SD) follow-up duration of 4.4 (2.6) years (range: 0.5-10), including two patients with ongoing treatment.

Among the 77 patients who completed treatment, there were 36 survivors, 7 with progressive disease, 5 deaths due to febrile neutropenia, and 29 relapses at a median duration of 11 months (IQR 7;15) following diagnosis. The latter received palliative care. Median ± SE five-year event-free survival (EFS) after excluding patients with treatment refusal/abandonment was 43.6 ± 6%. When abandonment/refusal were included as events, the 5-year EFS dropped to 21.3 ± 3.4% (Fig. 1). The last two years (2011-12) demonstrated a similar rate of refusal/abandonment (54% for 2011-12 versus 50% for 1990-2010). The survival was site-dependent; favorable sites, including orbit (EFS 57.1 ± 1.8%) demonstrated superior survival as compared to the following primary sites: parameningeal (35.2 ± 1.5%) extremities (36 ± 1.6%) and bladder-prostate (22.2 ± 1.2 %) (P=0.03).

Fig. 1 Five-year event-free survival in childhood rhabdomyosarcoma: (a) Excluding treatment refusal/abandonment; (b) Including treatment refusal/abandonment as events.

The study reports 159 children with RMS over a 23-year (1990-2012) period. Several tumors were >5 cm in size at presentation. The majority were located at unfavorable sites. Treatment refusal and abandonment were major concerns. Overall, a 5-year EFS of 43.6 ± 6% was documented. Orbit as the primary site demonstrated superior survival as compared to parameningeal, extremities and bladder-prostrate primaries.

The limitations of the study include the extended duration of enrollment, precluding uniformity in diagnostic and therapeutic modalities. In addition, high-risk protocol was not administered for likelihood of increasing treatment related mortality. Treatment refusal and abandonment were major predicaments.

In conclusion, in comparison to the West, children with RMS from India had an inferior survival despite similar disease characteristics. Treatment refusal/abandonment were recognized as major impediments to improving outcome. The measures incorporated in the unit in the recent years to reduce abandonment include, repetitive counseling sessions with the extended family as stakeholders, active collection of government funds with the help of social workers provided by non-government organizations, support for nutrition, as well as tracking of defaulters. Incorporation of measures to improve survival need to be studied by future researchers.

Funding: None. Competing interests: None stated.