OS4.1 MEVITEM: A European, randomized, open-label, Phase I/II study of vismodegib in combination with temozolomide versus temozolomide alone in adult patients with recurrent or refractory medulloblastoma presenting an activation of the Sonic Hedgehog (SHH) pathway

Abstract

Abstract BACKGROUND Vismodegib (V) suppresses sonic hedgehog (SHH) signaling. We postulated that vismodegib together with chemotherapy may be more efficient than chemotherapy alone in patients (pts) relapsing of a SHH-activated medulloblastoma (MB). MATERIAL AND METHODS Adult pts with recurrent SHH-MB not previously exposed to temozolomide (T) were randomly assigned (2:1 ratio) to Arm A (V daily 150mg/d, po) + T (D1-5: 150 mg/m2 for cycle 1 and 200 mg/m2 thereafter; n=up to 25pts) or Am B (T alone; n=up to 13pts). Identification of SHH activation was performed centrally by IHC (GAB1, β-catenin, filamin A, and YAP1). NGS analyses were performed to identify the mutations responsible for SHH activation. Primary objectives were to assess the incidence of severe toxicities (safety run-in based on a 3 + 3 design) and the 6-month non-progression rate (NPR-6m) according to WHO criteria and based on central read tumor assessment (Phase II). A Minimax Simon’s two-stage design was used to detect NPR-6m of 55% (p0: 30%, type I error rate of 5%, power of 80%). At first stage, ≥ 3/9 pts without progression at 6m were required for the accrual of 16 additional pts in Arm A. A 3rd independent and parallel arm with V as single agent (Arm C, n= up to15pts) was added for pts previously treated by T. RESULTS 24 SHH-MB pts were enrolled (Arm A: 10, Arm B: 5 and Arm C: 9; median age: 37 y [21–55]). At the end of the safety run-in; no major safety concerns were reported. At the end of Stage I: no objective response were reported and 2 pts among 10 were free of progression at 6m among in Arm A. According to statistical rules, the study was definitively closed to enrolment. NGS analyses showed a PTCH1 inactivating mutation in 6 pts (n=4 in arm A; n= 2 in arm B); a SMO activating mutation in 4 pts (n= 3 in Arm A; n=1 in Arm B). For 1 pt in each arm, no tumor sample was available for analysis, for 1 pt in Arm A DNA quality was insufficient, and for 1 patient in each arm no mutations of SMO, PTCH1, SUFU or SHH were found. Out of the 4 pts in Arm A with an inactivating PTCH1 mutation, only 1 was progression free at 6m. PFS and OS data will be presented at the meeting. CONCLUSION The combination of vismodegib with monthly T failed to demonstrate superior activity as compared with T alone. Further studies are warranted to refine therapeutic indication for vismodegib