© Journal of Thoracic Disease. All rights reserved. J Thorac Dis 2019; 11 (Suppl 15): S1906-S1908| http://dx. doi. org/10.21037/jtd. 2019.07. 60 gene panels include Memorial Sloan Kettering Cancer Center’s Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) and Foundation One CDx (F1CDx). Commercial gene panels also are available such as TrusSight Oncology 500 from Illumina and Oncomine Tumor Mutation Load Assay from ThermoFisher Scientific. While a variety of such gene panels of differing sizes are offered, there has been intense debate on the ideal sizes or methods of calculating TMB. In a recent study published in JAMA oncology, Wang and colleagues (10) used WES data of 9205 NSCLC cancer from The Cancer Genome Atlas (TCGA) to determine the minimum gene panel size and to optimize the cost-benefit ratio. As expected, a larger panel size increases the correlation between the panel-and WES-based TMB along with a decreasing Standard Deviation. This in silico analysis showed that a minimum of 150 genes should be covered. Interestingly and as previously reported, the inclusion of synonymous mutation strongly increases the sensitivity of the test when using small (< 150 genes) panel size. Panel composition should also be carefully selected to obtain sufficient separation of hypermutated tumors from non-hypermutated tumors. The authors designed a cancer gene panel covered whole exon regions of 150 selected cancer-related genes and called this panel NCCGP150. The panel was virtually validated using TCGA database. To further test the practicability of NCC-GP150, the authors used a public dataset including 34 patients with NSCLC treated with PD-1 inhibitor pembrolizumab. The progression-free survival was significantly longer in patients with high TMB than in patients with low TMB. The genetic profiling of tumors involves necessarily the use of tissue biopsies. Nevertheless, the availability of adequate tissue can be a limiting factor, especially for NSCLC patients (11). This is highlighted by the reduced number of patients for whom TMB evaluation on tissue samples was possible in clinical trials [59% in CheckMate 026 (7) and 58% in CheckMate 227 (8)]. In some instances, liquid biopsies are used alternatively to tissue biopsies, and they may be even preferred due to their noninvasive nature. However, the amount of circulating free DNA (cfDNA) varies greatly depending on diverse pathological conditions, such as the type of tumor, progression status, proliferative rate, and therapy regimen of the patient (12). In addition, the fraction of circulating tumor DNA (ctDNA) in total cfDNA is usually low. Therefore, a highly sensitive method would be needed to analyze TMB in cfDNA samples. The ctDNA isolated from blood can be analyzed by different technologies, including allele-specific PCR, digital droplet PCR and panel based-NGS. For TMB, given that the sensitivity of NGS-based technologies is inversely proportional to the number of loci analyzed, the use of gene panels that allow deeper coverage than WES would be more appropriate. Wang et al.(10) found that blood TMB (bTMB) can be reliably evaluated by their NGS panel (NCC-GP150) which was showed to have a satisfactory performance as compared to WES. Indeed, bTMB estimates via their panel correlated well with tissue TMB estimates via WES (Spearman correlation, 0.62). Moreover, high bTMB was associated with superior progression-free survival and objective response rates to ICIs. The authors validated their panel not only analytically, but also clinically. TMB evaluation on ctDNA is thus a very attractive tool, as it is non (less …