Multicentre evaluation of the management of children with high risk medulloblastoma: Real world performance of the SJMB03 and COG-99701 protocols

Abstract

Several treatment protocols are used for the treatment of high-risk medulloblastoma (HR-MB). In 2015, the UK Children’s Cancer and Leukaemia Group issued guidance recommending treatment as per the SJMB03 protocol, whilst also recognising that the COG-99701 protocol may be used. Patients were defined as high-risk if metastatic at presentation, large-cell/anaplastic histology, MYC amplification, significant residual disease or MYCN amplification. Recently, the latter two only define high risk if other adverse features are present.

Methods

Retrospective multi-centre service evaluation of treatment of HR-MB at five UK centres. Patients were included if treated as per SJMB03 or COG-99701. Patients were excluded if initially treated for standard-risk medulloblastoma and subsequently treated with these protocols due to upstaging or disease progression.

Results

58 patients were identified: 26 treated as per SJMB03, 32 as per COG-99701. 5-year OS was 83 % (95 %CI 73–94 %) and 5-year PFS was 65 % (53–80 %). For patients treated as per SJMB03, 5-year OS and PFS were 80 % (65–97 %) and 75 % (60–95 %) respectively; for patients treated as per COG-99701, 5-year OS and PFS were 85 % (73–100 %) and 60 % (43–83 %). There was no significant difference in outcomes between protocols. There was a higher incidence of grade 3/4 ototoxicity (44 % vs 6 %, p = 0.001) and admission to paediatric intensive care (19 % vs 0 %, p = 0.014) in patients treated as per SJMB03 compared to COG-99701.

Conclusion

These real-world outcomes are consistent with the published literature on HR-MB patients treated with these protocols within clinical trials, and provide important evidence to inform their use in routine practice.