Abstract
BACKGROUND
New therapies are especially needed for uMGMT nGBM. NRG Oncology BN002 (phase I) demonstrated safety and preliminary efficacy of Ipi+Nivo for nGBM, leading to BN007 as phase II/III.
METHODS
Adults with uMGMT nGBM (WHO 2016) and KPS ≥ 70 were randomized to RT and concurrent and adjuvant Ipi+Nivo or TMZ, stratified by glioma-recursive partitioning analysis class (RPA), intent to use tumor-treating fields (NCT04396860). Phase II primary endpoint was progression-free survival (PFS), powered 95% to detect a hazard ratio (HR) ≤0.58 with 1-sided 0.15 significance after 100 events. If p < 0.15, then phase III would accrue with overall survival (OS) as primary endpoint. Corticosteroids were disallowed at Ipi+Nivo start. Histology, biomarkers, and PFS were assessed centrally.
RESULTS
Among 374 patients screened, 159 were randomized in phase II (79 Ipi+Nivo, 80 TMZ). Demographics (age median 60 years, range 28-79; male 66%, white 88%, not Hispanic/Latino 89%), KPS (90-100 61%), extent of resection (gross total 65% per local PI), and RPA (III 10%, IV 73%, V 17%) were well-balanced between arms. After 100 PFS events, the pre-planned phase II analysis demonstrated no improvement for Ipi+Nivo vs. TMZ [median 7.7 (95% CI: 6.5, 8.5) vs. 8.5 (95% CI: 7.1, 10.4) months; HR 1.47 (95% CI: 0.98-2.2); 1-sided p = 0.96]. OS is immature ( >50% alive, 13.7 months median follow-up) but without observed difference between arms [median ~13 months each; HR 0.95 (95% CI: 0.61-1.49); 1-sided p = 0.36]. On the Ipi+Nivo arm, treatment-related grade 3, 4, and 5 events were reported in 26 (33.3%), 4 (5.1%), and 2 (2.6%; colitis and autoimmune disorder, n = 1 each) patients.
CONCLUSIONS
Ipi+Nivo did not improve PFS for uMGMT nGBM. Accrual permanently closed after phase II. No new safety signals were identified. Molecular analyses and survival follow-up are ongoing. FUNDING: NCI Grants U10CA180868, U10CA180822, U24CA196067 (NRG Oncology Operations, SDMC, Specimen Bank).