(P 155) LATE EVENTS OCCURRING AFTER 5 YEARS IN PEDIATRIC RHABDOMYOSARCOMA: A REPORT FROM THE CHILDREN'S ONCOLOGY GROUP

Objective: Rhabdomyosarcoma (RMS) is the most common pediatric soft tissue sarcoma with a 5-year overall survival of over 70%. With more long-term survivors, studies are needed to understand the frequency and etiology of late events, including recurrence, second malignant neoplasm (SMN), and death, occurring more than five years after initial diagnosis. Recent RMS trials have focused on lowering the burden of therapy in low-risk disease, optimizing local control approaches, and intensifying treatment in high-risk disease. Characterizing late events on these trials is critical to provide appropriate counseling and long-term surveillance of pediatric RMS survivors and to assist in efforts to refine current risk-directed therapies. The objectives of this study were to determine the prevalence of late events, identify clinicopathologic variables associated with late events, and compare these findings to older pediatric RMS IRSG cohorts.

Methods: We retrospectively reviewed the records of children with pediatric RMS treated on International Rhabdomyosarcoma Study Group (IRSG) and Children’s Oncology Group (COG) trials from 1997-2013 including D9602 (1997-2004), D9803 (1999-2005), D9802 (1999-2004), ARST 0331 (2004-2011), ARST0531 (2006-2012), ARST0431 (2006-2008), and ARST08P1 (2010-2013). We identified all patients on active follow-up and event-free at five years from study entry and evaluated late events in these patients. A late event was defined as a disease recurrence, development of a SMN, or death from any cause. Patients who experienced more than one late event were evaluated as experiencing two or more events. We analyzed the relationship between clinical and treatment variables and the occurrence of a late event.

Results: A total of 2,157 patients were enrolled between 1997 and 2013 and 1,038 were eligible for the analysis. Median follow-up was 8 years (range 5-14.75 years). A late event was identified in 30 patients (2.9%). First late event was recurrent RMS in 14, SMN in 14, and death in 2 patients. Median time to first late event was 5.7 years, 7.4 years, and 9.4 years respectively. The cumulative incidence at 10 years was 1.3% for recurrent RMS, 1.3% for SMN, and 0.19% for death. Cause of death for 1 patient treated on D9803 was lung disease and cor pulmonale. Table 1 describes the clinicopathologic features of the 30 patients with a late event.
Of the 14 patients with a late recurrence, 10 had initial Group III or IV disease. Eight had local recurrence, 3 had regional recurrence (1 with combined locoregional recurrence), and 1 had combined regional and distant recurrence. Of those patients with a late local recurrence, 50% had received radiotherapy as part of their initial treatment.
Among the 14 patients with a late SMN, there were 16 SMNs with 12 solid malignancies and 4 hematologic malignancies including myelodysplastic syndrome and acute myeloid leukemia. Solid malignancies included mucoepidermoid, thyroid, and basal cell carcinomas, osteosarcoma, undifferentiated sarcoma, anaplastic astrocytoma, pontine glioma, and Sertoli-Leydig tumor. Of the solid malignancies, 5 (41.7%) occurred in the radiation field of the primary tumor. Four of 14 patients (28.6%) had a known cancer predisposition syndrome including Li-Fraumeni, DICER1, and Gorlin syndrome.
Table 2 illustrates the factors associated with a late event on a univariate analysis. Group and stage were predictive of late recurrence with a cumulative incidence of 8.1% (p=0.0124) for patients with Group IV disease and 8.1% (p=0.0143) for patients with metastatic involvement. There was a non-significant trend towards increased rates of late SMN in patients with tumors ≥5 cm with a cumulative incidence of 4.2% (p=0.0560).

Conclusion: Late events after 5 years from diagnosis is uncommon in RMS survivors. Our data justifies discontinuation of surveillance for recurrence after 5 years from diagnosis. For patients who develop late SMN, germline evaluation should be considered if not already performed.

CTOS Abstract Table 1.pdf

CTOS Abstract Table 2.pdf