PurposeRadiation oncologists utilize radiation variably for children with metastatic rhabdomyosarcoma (RMS). Data from the XXXX study was retrospectively analyzed to validate the prior observation that the use of radiation is associated with improved outcomes, and guide future recommendations on radiation use in this patient group. Methods and MaterialsThe radiation delivered to 216 patients aged 0-21 years with metastatic RMS was retrospectively reviewed and classified as radical (all sites of disease irradiated within the protocol parameters), partial (some sites irradiated within the protocol parameters) and none (no radiation or delivered outside the protocol parameters). Landmark analysis excluded those with an event prior to day 221. Overall survival (OS) and progression free survival (PFS) were modelled using the Kaplan-Meier method to investigate the impact of radiation. The joint effect of treatment and known prognostic factors was examined using the Cox regression model. ResultsOverall 56 patients received radical, 104 partial and 56 no radiation per protocol. Due to non-randomised data, the groups were heterogeneous, particularly fewer sites of metatatic disease and less with bone metatases in those receiving radical radiation. 3-year PFS was 62.0% (95%CI 47.9-73.4) v 39.5% (29.8-49.1) v 30.1% (18.7-42.3)(p=0.002) for radical v partial v no radiotherapy respectively; 3-year OS was 70.1 % (55.8-80.6) v 53.1% (42.6-62.5) v 52.3% (38.3-64.5)(p=0.019) respectively. Multivariable analysis confirmed incremental improvement in OS with additional radiation with hazard ratio (HR) 1 v 1.8 v 2.4 (p=0.022) for radical v partial v no per protocol radiation. ConclusionsRadiation to all sites of disease seems to improve outcomes for children with metastatic RMS and should be considered when feasible. If not feasible, radiation is still recommended to the primary site and involved regional lymphadenopathy. Randomized clinical trials are required to confirm these findings, given the heterogeneity between the groups and potential confounding factors in this analysis.
Read full abstract