Bioinformatic Methods and Bridging of Assay Results for Reliable Tumor Mutational Burden Assessment in Non-Small-Cell Lung Cancer

Abstract

Introduction: Tumor mutational burden (TMB) has emerged as a clinically relevant biomarker that may be associated with immune checkpoint inhibitor efficacy. Standardization of TMB measurement is essential for implementing diagnostic tools to guide treatment.

Objective: Here we describe the in-depth evaluation of bioinformatic TMB analysis by whole exome sequencing (WES) in formalin-fixed, paraffin-embedded samples from a phase III clinical trial.

Methods: In the CheckMate 026 clinical trial, TMB was retrospectively assessed in 312 patients with non-small-cell lung cancer (58% of the intent-to-treat population) who received first-line nivolumab treatment or standard-of-care chemotherapy. We examined the sensitivity of TMB assessment to bioinformatic filtering methods and assessed concordance between TMB data derived by WES and the FoundationOne^® CDx assay.

Results: TMB scores comprising synonymous, indel, frameshift, and nonsense mutations (all mutations) were 3.1-fold higher than data including missense mutations only, but values were highly correlated (Spearman's r = 0.99). Scores from CheckMate 026 samples including missense mutations only were similar to those generated from data in The Cancer Genome Atlas, but those including all mutations were generally higher. Using databases for germline subtraction (instead of matched controls) showed a trend for race-dependent increases in TMB scores. WES and FoundationOne CDx outputs were highly correlated (Spearman's r = 0.90).

Conclusions: Parameter variation can impact TMB calculations, highlighting the need for standardization. Encouragingly, differences between assays could be accounted for by empirical calibration, suggesting that reliable TMB assessment across assays, platforms, and centers is achievable.

Fig. 1

Workflow for TMB assessment by WES in this study. Solid black arrows denote steps that were included in all WES analyses. Dashed gray arrows show which steps were investigated for effects on TMB output in this study. COSMIC Catalogue of Somatic Mutations in Cancer, ExAC Exome Aggregation Consortium, indel short insertion/deletion, NGS next-generation sequencing, SNV single nucleotide variant, TMB tumor mutational burden, WES whole exome sequencing

Fig. 2

Analysis of TMB by WES in CheckMate 026 samples. Data show TMB scores including missense mutations only vs “all mutations” (including synonymous, indel, frameshift, missense, and nonsense mutations). The linear regression line is shown in blue. Indel short insertion/deletion, TMB tumor mutational burden, WES whole exome sequencing

Fig. 3

Comparison of TMB scores determined by WES in CheckMate 026 samples with those obtained from publicly available TCGA data. a TMB assessment includes missense mutations only. b TMB assessment includes “all mutations” (including synonymous, indel, frameshift, missense, and nonsense mutations). Dark blue lines show the median value and interquartile ranges for each subset. Indel short insertion/deletion, non-SQ nonsquamous non-small-cell lung cancer, SQ squamous non-small-cell lung cancer, TCGA The Cancer Genome Atlas, TMB tumor mutational burden, WES whole exome sequencing

Fig. 4

a The effect of germline subtraction by different methods on TMB values. TMB was derived from 312 patient samples by WES. The horizontal axis (tumor/normal TMB) shows TMB values calculated using patient-matched blood samples for germline subtraction. The vertical axis (tumor-only TMB) shows TMB values derived from the same sample, using public databases for in silico filtering. Black line shows linear regression across all patients. Equal values across the two datasets (X = Y) are represented by a gray dotted line. b Data from part (a) colored by patient race. Solid lines show linear regression analyses for subsets of patients grouped by race. Numbers in parentheses denote the sample numbers in each subgroup. Linear regression for the entire dataset is shown by a red dotted line. Inset shows the data magnified across tumor/normal TMB values from 0 to 350 mutations to highlight linear correlations occurring around TMB cutoff values that may be considered to be “clinically relevant”. TMB tumor mutational burden, WES whole exome sequencing

Fig. 5

Workflow for TMB assessment using the FoundationOne CDx assay [63]. ExAC Exome Aggregation Consortium, indel short insertion/deletion, NGS next-generation sequencing, SGZ somatic-germline zygosity, TMB tumor mutational burden

Fig. 6

a Correlation of TMB assessed by WES and the FoundationOne CDx assay. Nonparametric linear regression is shown with a blue line. Blue shaded area shows 0.95–confidence bounds for the linear regression, calculated with a bootstrap (quantile) method. Cutoffs chosen for grouping of sample data are shown with red and green lines. b Grouping of TMB data into categories defined by cutoff values (10 mut/Mb by the FoundationOne CDx assay or 199 missense mutations by WES). F1 CDx FoundationOne CDx, mut/Mb mutations per megabase, TMB tumor mutational burden, WES whole exome sequencing