Premium
Size matters: Dissecting key parameters for panel‐based tumor mutational burden analysis
Author(s) -
Buchhalter Ivo,
Rempel Eugen,
Endris Volker,
Allgäuer Michael,
Neumann Olaf,
Volckmar AnnaLena,
Kirchner Martina,
Leichsenring Jonas,
Lier Amelie,
Winterfeld Moritz,
Penzel Roland,
Christopoulos Petros,
Thomas Michael,
Fröhling Stefan,
Schirmacher Peter,
Budczies Jan,
Stenzinger Albrecht
Publication year - 2018
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.31878
Subject(s) - cutoff , exome , computational biology , confidence interval , computer science , turnaround time , exome sequencing , oncology , medicine , biology , mutation , genetics , gene , physics , quantum mechanics , operating system
Tumor mutational burden (TMB) represents a new determinant of clinical benefit from immune checkpoint blockade that identifies responders independent of PD‐L1 expression levels and is currently being explored in clinical trials. Although TMB can be measured directly by comprehensive genomic approaches such as whole‐genome and exome sequencing, broad availability, short turnaround times, costs and amenability to formalin‐fixed and paraffin‐embedded tissue support the use of gene panel sequencing for approximating TMB in routine diagnostics. However, data on the parameters influencing panel‐based TMB estimation are limited. Here, we report an extensive in silico analysis of the TCGA data set that simulates various panel sizes and compositions. We demonstrate that panel size is a critical parameter that influences confidence intervals (CIs) and cutoff values as well as important test parameters including sensitivity, specificity, and positive predictive value. Moreover, we evaluate the Illumina TSO500 panel, which will be made available for TMB estimation, and propose dynamic, entity‐specific cutoff values based on current clinical trial data. Optimizing the cost–benefit ratio, our data suggest that panels between 1.5 and 3 Mbp are ideally suited to estimate TMB with small CIs, whereas smaller panels tend to deliver imprecise TMB estimates for low to moderate TMB (0–30 muts/Mbp), connected with insufficient separation of hypermutated tumors from non‐hypermutated tumors.