Article Text
PDF
Abstract
Background High tumor mutational burden (TMB) and PD-L1 expression are associated with improved clinical outcomes in patients (pts) with non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICIs). However, how TMB performs as a predictive biomarker to ICIs in different PD-L1 expression subgroups is not well characterized.
Methods We collected clinicopathologic and genomic data from NSCLCs which underwent targeted NGS and TMB assessment at DFCI. An unbiased recursive partitioning (URP) algorithm was used to investigate an optimal TMB cut-off with respect to objective response rate (ORR) in the subset of pts treated with ICIs. This TMB cut-off was then validated in the prospective POPLAR/OAK cohort.
Results Among 3560 NSCLCs identified, median TMB was significantly higher among current smokers compared to former (P<0.0001) and never smokers (P<0.0001), and there was a significant correlation between TMB and pack-years (figure 1A-B). Pts with BRAF or KRAS mutations had the highest median TMB (10.9 and 9.8 mutations/Megabase [mut/Mb], respectively), while tumors with RET and ALK alterations had the lowest median TMB of 5.3 mut/Mb (figure 2A-B). Tumors with PD-L1 expression of ≥50% had significantly higher median TMB compared to those with a PD-L1 expression of 1–49% (P=0.002) and <1% (P<00001). Among pts treated with ICIs (N=690), URP identified an optimal grouping TMB cut-off for ORR of 19.0 mut/Mb, which corresponded to the 90th percentile. Pts with a TMB of ≥19.0 mut/Mb had a significantly higher ORR (45.2% vs 20.1%, P<0.0001) and longer median PFS (11.0 vs. 2.9 months, HR:0.49, P<0.0001) and OS (20.8 vs. 11.2 months, HR:0.59, P=0.001) compared to those with a TMB of <19.0 mut/Mb (figure 3A-C). A TMB of ≥19.0 mut/Mb was an independent predictor of improved PFS and OS at multivariable analysis (table 1). A TMB within the top 10th percentile was confirmed to correlate with improved ORR and PFS in atezolizumab arm but not in the docetaxel arm of the POPLAR/OAK trials (figure 4A-B). When TMB and PD-L1 where integrated in the URP, we identified an optimal cut-off of 19 mut/Mb among cases with a PD-L1 expression of ≤25%, and of 8.4 mut/Mb among those with a PD-L1 expression of >25%, suggesting that TMB differentially impacts response to immunotherapy among PD-L1 high versus low NSCLCs (figure 5).
(A) Correlation between TMB and smoking status. (B) Linear correlation between TMB and pack-years on a continuous scale
(A) TMB distribution across genomically defined subsets of NSCLC. (B) Pairwise comparison in TMB distribution among genomically defined subsets of NSCLC. Reported in each box are the Q-values for each comparison (FDR method of Benjamini and Hochberg)
(A) Overall response rate, (B) progression-free survival and (C) overall survival to PD-(L)1 inhibition in patients with NSCLC and a TMB of ≥19.0 vs <19.0 mut/Mb, as determined by unbiased recursive partitioning
(A) Overall response rate, progression-free survival and overall survival to atezolizumab in patients enrolled in the POPLAR/OAK trials and a bTMB of ≥25 (top 10th percentile) vs <25.0 (lower 90th percentile) mut/Mb. (B) Overall response rate, progression-free survival and overall survival to docetaxel in patients enrolled in the POPLAR/OAK trials and a bTMB of ≥25 (top 10th percentile) vs <25.0 (lower 90th percentile) mut/Mb. bTMB, blood tumor mutational burden
Unbiased regression tree algorithm recursively identifies that different TMB cut-offs impact response to immunotherapy in PD-L1 TPS high (≥25) vs low (<25%) NSCLCs. TPS, tumor proportion score
Multivariable Cox regression for PFS and OS among patients with NSCLC treated with PD-(L)1 blockade.
Conclusions The impact of TMB may vary across PD-L1 expression subgroups. Rational integration of TMB and PD-L1 expression may identify NSCLCs with the greatest likelihood of response or resistance to ICIs.
Ethics Approval Clinicopathologic data were collected from patients with advanced NSCLC who had consented to a correlative research study (DF/HCC protocol #02-180).
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.