You can manage bookmarks using lists, please log in to your user account for this.
Media type:
E-Article
Title:
Abstract 4125: Precise and reproducible measurement of tumor mutational burden using panel sequencing: the role of panel size and of mutation types
Description:
Abstract Tumor mutational burden (TMB) is an emerging biomarker for patient stratification for immune checkpoint inhibition. Implementing TMB measurement by panel sequencing, experimental setup and bioinformatics pipelines need to be optimized to minimize sources of error and variance. To gain insight into contributors to TMB variance, combinatorial calculations were performed based on the hypothesis that each base in the coding sequence was mutated with the same probability. Simulations of sequencing panels in the TCGA data were performed to check for additional contributors to the variance. A narrow definition of TMB including only missense mutations was compared to a broader definition additionally including other mutation types. The quantity under investigation was TMB measured in relatively to the size of the panel (per Mbp). The combinatorial contribution to the coefficient of variation (CV) of TMB turned out to be inversely proportional to the square root of the panel size and inversely proportional to the square root of the TMB. For a tumor with TMB of 10 muts/Mbp, the combinatorial CV increased drastically for panels sizes smaller than 1.5 Mbp. The combinatorial variance represents a persistent contribution to the error of TMB calculation that cannot be avoided. Simulations in WES data confirmed the combinatorial variance as main contributor to the imprecision of TMB estimation. In many instances, the fold change between the number of all mutations in the coding sequence and the number of missense mutations was similar for tumors of the same cancer type. However, the fold changes were different in specific cancer types for example significantly higher in melanoma (1.66) compared lung adenocarcinoma (1.48). In particular, the former tumors were observed to include a higher proportion of synonymous mutations than the latter. The combinatorial source of error implies that a panel of least 1.5 Mbp is needed to precisely determine TMB for tumors ranging around typical currently used diagnostic thresholds (10 muts/Mbp). Additional sources of error might even favor larger panel sizes such as 3 Mbp. The combinatorial error can be reduced by including all mutations types instead of only missense mutations and simultaneously adjusting TMB with a cancer type specific factor. Citation Format: Jan Budczies, Eugen Rempel, Michael Allgäuer, Daniel Kazdal, Volker Endris, Peter Schirmacher, Albrecht Stenzinger. Precise and reproducible measurement of tumor mutational burden using panel sequencing: the role of panel size and of mutation types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4125.