Long-term outcome of the Milano-hyperfractionated accelerated radiotherapy strategy for high-risk medulloblastoma, including the impact of molecular subtype.

Background: We applied the strategy for M+ medulloblastoma across all high-risk subgroups, including LC/A histology, TP53 mutations, and MYC/MYCN amplification.

Methods: Patients over 3 years old received, after surgery, staging and histo-biological analysis, sequential high-dose-methotrexate(HD-MTX), high-dose-etoposide(HD-VP16), high-dose-cyclophosphamide(HD-Cyclo), and high-dose-carboplatin(HD-Carbo). Hyperfractionated-accelerated-radiotherapy-craniospinal(HART-CSI), administered twice daily 1.3 Gy-fractions reached a total dose tailored to the patients' age and pre-radiation response to chemotherapy(CT): 31.2 Gy if under 10-years-old and complete response(CR) or partial response(PR) obtained or absence of metastatic disease, 39 Gy in other/older patients. Boosts to posterior fossa/residual metastatic(M+) deposits were given up to a total dose of 60 Gy/9 Gy, respectively, but avoided if metastatic nodules were very big or patients were very young. Two courses of high-dose-thiotepa were delivered in case of not CR/PR after the pre-radiotherapy (RT) phase and in all M0 patients either-pre/post-HART. Subgrouping was performed where the tissue was available.

Results: Eighty-nine patients were enrolled, with a median age of 8.8 years, and a median follow-up of 136 months. Overall survival (OS) and event-free survival (EFS) at 5/15 years were 75.9/66.5% and 68.2/65.3%, respectively; 5/28 fatal events were not related to relapse(3 developed secondary malignancies). Sex, age less than 10 years, histological subtype, presence of MYC/MYCN amplification, reduction in CSI dose, omission of RT-boosts, implementation of myeloablative therapy, presence-absence of metastases did not impact prognosis.Patients progressing after pre-HART CT(14/89) and stable-disease(SD)+PD after HART(10/89) negatively affected outcome(P < .001).Subgrouping in 66/89 patients' samples demonstrated a significantly worse EFS for patients with Sonic Hedgehog(SHH)-tumors(#15, 2 with constitutional TP53-mutations) versus groups 3 and 4(15 and 29 patients, respectively, group3/4 in 7).Patients younger than 10 received lower CSI doses if stratified according to CT response.

Conclusions: This strategy, partly adopted in the ongoing SIOPE protocol, confirmed improved EFS and OS over previously reported outcomes in all high-risk categories; SHH tumors appeared the most aggressive.