Are molecular subgroups of medulloblastomas really prognostic?

PURPOSE OF REVIEW: Medulloblastoma is no more a unique disease. Clinical and biologic classification used so far are challenged by molecular classification(s). Following the consensus article that described four molecular groups of medulloblastoma in 2012, several articles in 2017 provided more relevant classifications that may impact on further clinical trial design. RECENT FINDINGS: Though wingless (WNT) and sonic hedgehog (SHH) are defined by the activation of their respective pathways, the age and type of activation define various subgroups with specific features and outcome. Groups 3 and 4 remain ill defined. The whole population of medulloblastoma may be divided in 12 subgroups: WNTαß, SHHαßγδ, group 3αßγ and group 4αßγ. The paediatric population may be divided in seven subgroups: WNT, SHH of infants and children, and low-risk and high-risk groups 3 and 4. SHH of infants may be divided as iSHH-I vs. iSHH-II that have different prognosis. Moreover, specific drivers of groups 3 and 4 were reported. SUMMARY: These findings have and will have direct implications on the conception of clinical trials. Low-risk groups will benefit from less toxic therapies, and high-risk groups will benefit from targeted therapies.

Medulloblastomas in adults: prognostic factors and lessons from paediatrics.

PURPOSE OF REVIEW: Medulloblastomas are very rare in adults. Usual treatment consists of craniospinal radiation with or without chemotherapy. Current efforts focus on a better understanding of tumour biology, stratifying patients into risk groups and adapting treatment accordingly. This review discusses clinical and new molecular risk factors that will help to optimize treatment in adult medulloblastoma patients. RECENT FINDINGS: The clinical risk stratification should be complemented with new molecular prognostic markers. Gene-expression profiling has permitted identification of four to six molecular medulloblastoma subgroups. The WNT subgroup shows overexpression of genes of the WNT/wingless signalling pathway with frequent mutations of the CNNTB1 gene, loss of chromosome 6 and accumulation of nuclear ß-catenin, and is most often seen in children with medulloblastomas of classical histology. This variant has a good prognosis. Activation of the sonic hedgehog pathway with frequent mutations of the PTCH and SUFU genes, loss of 9q, and positivity for GLI1 and SFRP1 is more frequent in children less than 3 years old and in adults, commonly associated with desmoplastic histology. Other subgroups are not so well defined and have overlapping characteristics, but MYC/MYCN amplification, 17q gain and, large cell/anaplastic histology are factors of poor prognosis. SUMMARY: New molecular subgroups will help tailor treatment and further develop new targeted therapies. Prospective and ideally randomized trials should be performed in adults, including risk stratification by molecular markers, to identify optimal treatment for each risk group.

Postoperative Radiation Therapy in Patients with Extracranial Chondrosarcoma: A Joint Study of the French Sarcoma Group and Rare Cancer Network.

PURPOSE: Postoperative radiation therapy (poRT) of intracranial/skull base chondrosarcomas (CHSs) is standard treatment. However, consensus is lacking for poRT in extracranial CHS (eCHS) owing to their easier resectability and intrinsic radioresistance. We assessed the practice and efficacy of poRT in eCHS. METHODS AND MATERIALS: This multicentric retrospective study of the French Sarcoma Group/Rare Cancer Network included patients with eCHS who were operated on between 1985 and 2015. Inverse propensity score weighting (IPTW) was used to minimize poRT allocation biases. RESULTS: Of 182 patients, 60.4% had bone and 39.6% had soft-tissue eCHS. eCHS were of conventional (31.9%), myxoid (28.6%; 41 extraskeletal, 11 skeletal), mesenchymal (9.9%), or other subtypes. En-bloc surgery with complete resection was performed in 52.6% and poRT in 36.8% of patients (median dose, 54 Gy). Irradiated patients had unfavorable initial characteristics, with higher grade and incomplete resection. Median follow-up time was 61 months. Five-year incidence of local relapse was 10% with poRT versus 21.6% without (P = .050). Using the IPTW method, poRT reduced the local relapse risk (hazard ratio, 0.27; 95% confidence interval, 0.14-0.52; P < .001). Five-year disease-free survival (DFS) was 71.8% with poRT and 64.2% without (P = .680). Using the IPTW method, poRT improved DFS (hazard ratio, 0.51; 95% confidence interval, 0.30-0.85; P = .010). The benefit of poRT on local relapse and DFS was confirmed after exclusion of the extraskeletal subtype. There was no difference in overall survival. Prognostic factors of poorer DFS in multivariate analysis were deeper location, higher grade, incomplete resection, and no poRT. CONCLUSIONS: poRT should be offered in patients with eCHS and high-grade or incomplete resection, regardless of the histologic subtype.