Toward a Better Understanding of the Use of Targeted Therapies in Pediatric Sarcoma Patients

Abstract

Sarcomas account for one-seventh of all pediatric cancers, and patients whose tumors recur following initial therapy often do not survive. Although there are occasional extraordinary responders, care for most patients with metastatic recurrence becomes palliative at some point. In the search for new strategies for children with relapsed sarcoma, targeted therapies are attractive, as they offer the hope of exploiting specific metabolic pathways, important for tumor growth. These targeted therapies are usually outpatient medicines that have less myelosuppression in heavily pretreated patients when compared with conventional cytotoxic agents.1 For these and other reasons, many pediatric oncologists prescribe off-label targeted therapies for relapsed sarcoma patients, although there is little in the medical literature about this practice. In this issue of Journal of Pediatric Biochemistry, Garnier et al report data from a French registry of prospectively treated patients with relapsed pediatric sarcoma who received targeted therapies at the discretion of the treating physician.2 This cohort of 34 patients was not eligible for other clinical trials, and treatment decisions were made without genetic evaluation of tumor tissue. The majority of patients received a targeted agent in combination with standard chemotherapy. Although there are methodological issues that limit firm conclusions from this study, the findings do provide insight into practice patterns and potential benefits and toxicities of these agents in recurrent sarcoma patients, and represent a good attempt at prospectively collecting data on patients treated outside of the typical clinical trial. While a few childhood tumors such as anaplastic large cell lymphoma or subependymal giant cell astrocytoma have had dramatic and consistent responses to single-agent targeted therapies,3,4 this has not been the case so far with pediatric sarcoma. Instead, the most likely path forward will be to combine targeted agents with conventional cytotoxics, as evidenced in Garnier et al’s study in which half of the responders were also receiving standard chemotherapy drugs. This approach has already been shown useful for neuroblastoma, in which combining the anti-GD2 antibody dinutuximab with the chemotherapy backbone of temozolomide plus irinotecan has shown very encouraging activity in patients with relapsed or refractory high-risk disease.5 This strategy of coupling targeted agents with conventional cytotoxics is now being tested in cooperative group studies for metastatic Ewing sarcoma (clinical trials.gov identifier NCT02306161), rhabdomyosarcoma (NCT01871766), and other soft tissue sarcomas (NCT02180867). For osteosarcoma, the search for the best targeted therapyagent for combination continues, with single-agent trials being conducted in phase II studies of patients with relapsed disease (NCT02470091, NCT02484443, NCT02487979). As oncologists await these results, the question remains how best to treat our patients with relapsed disease. Caution is warranted when using targeted therapies in heavily pretreated patients. Although there is less myelosuppression, the incidence of grade 3 to 4 nonhematologic toxicity and toxic death may be similar to those receiving conventional cytotoxics.1 Further, dosing in younger patients is often less established, particularly in combination with conventional cytotoxics, and tolerance of these agents may be different in children compared with adults.6 Genetic testing of tumor tissue may make selection of particular targeted therapies more rational, although actionable mutations are only present in a minority,7 and the benefit of using appropriately matched targeted therapy has not been firmly established for pediatric sarcoma patients.8 Given these limitations, it is incumbent on treating physicians to strongly consider clinical trial enrollment for relapsed sarcoma patients, either on novel regimens or on registries as done by Garnier et al. Such prospective collection of data will be the only way to identify biomarkers and treatment strategies that have the potential to help both our current and future patients.