(a-b) Activating RTK mutations identified a group of patients with dismal prognosis in both cohort 1 (N=89 patients, log rank p=0.005) and cohort 2 (N=34 patients, log rank p=0.009). (c) In cohort 2 (N=31), responding tumors had a higher total and clonal TMB compared to non-responding tumors (Mann Whitney U test p=0.006 and p=0.004 respectively). Progression-free survival, histology and tumor purity are shown as separate panels. Patients with a clinical response had a higher contribution of the molecular smoking signature (N=15 patients, Mann Whitney U test p=0.054). There were no differences in tumor aneuploidy between responders and non-responders (N=74 patients, Mann Whitney U test p=0.72). A significant enrichment in RTK activating mutations, including point mutations and amplifications in EGFR, amplifications in ERBB2 and MET exon 14 skipping, was found in non-responding tumors (N=89 total patients, chi-squared p=0.056). (d) A third cohort of NSCLC patients treated with ICB was obtained from CBioportal (N=240 patients) as described in detail in the Methods section; for this cohort, sequence and copy number alterations, as well as outcome information were publicly available. Patients with activating RTK mutations in EGFR, ERBB2, MET, FGFR1 and IGF1R had a significantly shorter progression-free survival (log rank p=0.035). We defined response as durable clinical benefit if complete, partial response or stable disease was achieved with a duration >6months. The median point estimate and 95% confidence intervals for survival were estimated by the Kaplan–Meier method and survival curves were compared by using the nonparametric log rank test. CNV; copy number variation. cTMB; corrected TMB, RTK; receptor tyrosine kinase, Hist; histotype, TP; tumor purity, SS; smoking signature, AI; allelic imbalance. Dots indicate hotspot mutations, and × denotes loss of heterozygosity of the wild type allele. P values are based on two-sided testing.