The following is a summary of the “Tumour mutational burden is overestimated by target cancer gene panels,” published in the March 2023 issue of Oncology by Fang, et al.

Tumor mutational burden (TMB) has emerged as a predictor of response to immune checkpoint inhibitors (ICI) in various tumor types. It can be computed using somatic mutations found in whole exome or targeted panel sequencing data. However, because mutations are distributed unevenly throughout the cancer genome, the clinical implications of TMB calculated using different genomic regions are unclear. Pan-cancer data from The Cancer Genome Atlas cohort were collected, and 6,831 cancer patients with either ICI or non-ICI treatment outcomes were derived from published papers. 

TMB was calculated as the number of non-synonymous mutations multiplied by the size of genomic regions. TMB from different gene panels is combined using the Dirichlet method, linear regression, and Poisson calibration models. Researchers discovered that panels based on cancer genes typically overestimate TMB compared to the whole exome, potentially misclassifying patients for ICI. Positive selection for mutations in cancer genes causes overestimation, which cannot be completely addressed by removing mutational hotspots. 

To address this disparity, researchers compared various approaches and developed a generalized statistical model capable of interconverting TMB derived from whole exome and different panel sequencing data, allowing TMB correction for patient stratification for ICI treatment. We show that when using a TMB cutoff of 10 mut/Mb in a cohort of lung cancer patients treated with ICI, our corrected TMB outperforms the original panel-based TMB.

Cancer gene-based panels typically overestimate TMB, and these findings will be useful in clinical practice for unifying TMB calculations across cancer gene panels.

Source: sciencedirect.com/science/article/pii/S2667005422000801