Checkpoint Blockade Combinations in High Tumor Mutational Burden/Load Tumors: Insights from the Atezolizumab + Bevacizumab and Nivolumab + Ipilimumab Cohorts in DRUP

Abstract Purpose: To evaluate the efficacy of atezolizumab plus bevacizumab (atezo+beva) in tumors with high tumor mutational burden (TMB; number of mutations per megabase), and nivolumab plus ipilimumab (nivo+ipi) in tumors with high TMB or tumor mutational load (TML; total number of non-synonymous mutations across the genome). Patients and methods: Patients with treatment-refractory, solid tumors were treated in the Drug Rediscovery Protocol (2023-509152-33-00). Patients with microsatellite stable tumors harboring a TML of 200-1000, or TMB of 11-24 (Oncomine) or 15-39 (TSO500) were eligible for nivo+ipi. Similar patients with a panel-independent TMB ≥16 received atezo+beva. Clinical benefit (CB; confirmed objective response or stable disease ≥16 weeks) was the primary endpoint. Whole-genome and RNA-sequencing were performed on pre-treatment tumor biopsies. Results: Among 25 evaluable patients with 14 different tumor types treated with atezo+beva, the CB-rate (CBR) was 60% (95% confidence interval CI, 39-79), with an objective response rate (ORR) of 24% (95% CI, 9-45) and median duration of response (mDoR) of 25.0 months (95% CI, 13.8-NA). In the nivo+ipi cohort the CBR was 50% (95% CI, 29-71) and ORR 37.5% (95% CI, 19-59) among 24 evaluable patients with 13 distinct tumor types. The mDoR was not reached after a median follow-up of 36 months. In both cohorts, responses were only observed in patients with TMB 20, and TMB and (clonal) TML were significantly correlated with response. Various markers of adaptive immune infiltration were associated with longer progression-free survival. Conclusions: Atezo+beva and nivo+ipi showed durable responses in patients with TMB 20, underscoring their tumor-agnostic efficacy in this patient population.