The prognostic impact of tissue tumour mutational burden (TMB) in the first-line treatment of advanced non-oncogene addicted non-small cell lung cancer (NSCLC): A systematic review and meta-analysis of randomized controlled trials

Although PD-L1 expression and Tumor Mutation Burden (TMB) revealed to be the most effective predictors for selecting Non-Small Cell Lung Cancer (NSCLC) patients who might benefit from immune checkpoint inhibitors (ICIs), the role of these biomarkers into clinic is still debated. In addition, tissue TMB failed to predict benefit in overall survival (OS) among NSCLC patients treated with ICIs. The purpose of this meta-analysis, including seven randomized phase II and III studies (KEYNOTE-042, CheckMate-227, KEYNOTE-189, KEYNOTE-407, KEYNOTE-021 cohort G, CheckMate-026, MYSTIC), was to compare efficacy outcomes among first-line Immune-Oncology (IO) treatment strategies (single agent ICI, ICIs plus chemotherapy and combination ICIs) within two subgroups (tissue TMB-low and -high) versus standard platinum-based chemotherapy (CT). The pooled hazard ratio (HR) and relative risk (RR) for the indirect comparisons have been estimated. Indirect comparisons for efficacy outcomes showed that ICIs were associated with a statistically significant advantage over CT alone in terms of response rate (RR 1.37, 95% CI 0.94 - 1.99), progression-free survival (PFS; HR 0.57, 95% CI 0.48 - 0.67) and OS (HR 0.72, 95% CI 0.60 - 0.86), supporting the use of I-O regimens strategy within the TMB-high subgroup (1255 patients). PFS pooled results underline a strong trend for significance in favor of CT over I-O regimens (HR 1.16,95% CI 0.89 - 1.51) in the TMB-low subgroup (1664 patients), thus suggesting a possible predictive role of high TMB for I-O regimens. This is the first scientific effort that proved a strong benefit in OS in favor of ICIs over CT alone in the TMB-high population. Strikingly, the pooled results of PFS in the TMB-low subgroup were associated with an unprecedented trend towards CT over ICIs. Although more prospective data are warranted, we inferred that monitoring TMB, in addition to the existing PD-L1 expression level, could represent the preferable option for the first-line management of advanced non-oncogene addicted NSCLC.