Introduction
Treatment with immune checkpoint inhibitors (ICIs) has become extremely popular owing to its high efficacy and long-term benefits. However, only a few patients benefit from ICIs because of the cost of ICIs and adverse events. Therefore, it is important to explore relevant biomarkers. Tumor mutational burden (TMB), which is defined as the number of somatic mutations per DNA mega base (Mb), is one of the emerging prognostic factors to predict the efficacy of immunotherapy, especially ICIs. Clinical studies1-3 have indicated that patients with high TMB had better response and survival after the treatment of ICIs in many cancers than those with a low TMB.
TMB is a novel prognostic biomarker of clinical benefit from ICIs independent of the PD-L1 level and is currently being prospectively explored.4,5 In clinical practice, the detection of TMB is based on whole exon sequencing or targeted next-generation sequencing of tissue or blood samples. The detected tissues may be from primary or metastatic lesions. Since the number of mutations in primary and metastatic lesions is different, it may have some limitations to use the same standard to predict the prognosis of ICIs for both lesions.
Therefore, in this study, we analyzed the difference in the TMB score between primary and metastatic lesions in pan-cancer and determined the difference in cut-offs for guiding ICIs prognosis.