Abstract LB-261: Association between tumor mutational burden (TMB) assessed by whole-exome sequencing (WES) and outcomes of pembrolizumab (pembro) monotherapy

Abstract Background: We retrospectively assessed the relationship between TMB and outcomes in patients (pts) with previously treated advanced solid tumors enrolled in 12 trials that evaluated pembro monotherapy. Methods: WES was performed on DNA extracted from pretreatment tumor samples and matched normal DNA from pts in KEYNOTE-001, 002, 010, 012, 028, 045, 055, 059, 061, 086, 100, and 199 who received ≥1 prior therapy and ≥1 dose of study therapy. ORR, DOR, PFS, and OS were assessed for pts with TMB ≥175 mut/exome and 175 mut/exome. Exploratory regression models were performed and nominal two-sided P values (not adjusted for multiplicity) calculated to evaluate the association between continuous, log10-transformed TMB score and outcomes for pembro and chemo in the 3 randomized studies (KEYNOTE-010, 045, and 061). Results: 2234 pts across 24 tumor types had evaluable TMB. Among the 1772 pembro-treated pts, 433 had TMB ≥175 mut/exome and 1339 had TMB 175 mut/exome. Median follow-up was 13.7 and 9.6 mo, respectively. In the pooled pembro population, clinically meaningful enrichment in efficacy was seen for TMB ≥175 mut/exome (Table). Enrichment was seen in the 8 tumor types that included 10 pts, in a pooled set of tumor types with ≤10 pts, and regardless of PD-L1 expression assessed by IHC. In the randomized studies, pembro improved outcomes vs chemo in pts with TMB ≥175 mut/exome (Table). TMB was consistently associated with ORR (P ≤ 0.016), PFS (P ≤ 0.005), and OS (P ≤ 0.029) for pembro but not for chemo (P ≥ 0.340, ≥0.643, and ≥0.174, respectively). Conclusions: TMB ≥175 mut/exome assessed by WES was associated with clinically meaningful improvement in efficacy of pembro monotherapy and improved outcomes for pembro vs chemo in pts with previously treated advanced solid tumors. TMB was associated with outcomes of pembro but not of chemo. These data suggest TMB may predict efficacy of pembro monotherapy, regardless of tumor type. Table.Efficacy by TMB ScoreTMB ≥175 mut/exomeTMB 175 mut/exomePooled pembro studiesN = 433N = 1339ORR, % (95% CI)31.4 (27.1-36.0)9.5 (8.0-11.2)DOR, mo, median (range)Not reached (1.6+ to 51.5+)12.5 (1.2+ to 49.7+)PFS, mo, median (95% CI)4.2 (3.6-5.8)2.1 (2.0-2.1)OS, mo, median (95% CI)15.5 (13.1-17.7)10.1 (9.4-10.9)Pembro vs chemo in KEYNOTE-010n = 132n = 121ORR, odds ratio (95% CI)2.82 (0.98-8.11)0.76 (0.29-2.00)PFS, HR (95% CI)0.59 (0.40-0.87)1.09 (0.72-1.63)OS, HR (95% CI)0.56 (0.38-0.83)0.85 (0.56-1.30)Pembro vs chemo in KEYNOTE-045n = 107n = 246ORR, odds ratio (95% CI)3.05 (1.20-7.79)1.10 (0.55-2.21)PFS, HR (95% CI)0.62 (0.40-0.96)1.13 (0.87-1.47)OS, HR (95% CI)0.63 (0.40-1.00)0.71 (0.54-0.94)Pembro vs chemo in KEYNOTE-061n = 76n = 348ORR, odds ratio (95% CI)3.43 (0.99-11.85)0.61 (0.31-1.22)PFS, HR (95% CI)0.73 (0.44-1.22)1.78 (1.43-2.22)OS, HR (95% CI)0.46 (0.27-0.81)1.12 (0.90-1.41) Citation Format: Razvan Cristescu, Deepti Aurora-Garg, Andrew Albright, Lei Xu, Xiao Qiao Liu, Andrey Loboda, Lixin Lang, Fan Jin, Alexandra Snyder, Jared Lunceford. Association between tumor mutational burden (TMB) assessed by whole-exome sequencing (WES) and outcomes of pembrolizumab (pembro) monotherapy abstract. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr LB-261.