The genomic characteristics of different progression patterns in advanced non-small cell lung cancer patients treated with immune checkpoint inhibitors

Background: Fast progression (FP), hyperprogressive disease (HPD), and early death (ED) are the newly reported cancer progression patterns in response to immune checkpoint inhibitor (ICI) treatment. This study aimed to investigate the clinical and genomic characteristics of FP, HPD, and ED following the ICI treatment of advanced non-small cell lung cancer (NSCLC). Methods: We retrospectively reviewed 117 patients with advanced NSCLC who were treated with ICIs from March 2017 to October 2019. FP was defined as (I) time to treatment failure (TTF) <1.5 months; and (II) >= 50% increase in the sum of the longest diameter (SLD) of target lesions. HPD was defined as (I) TTF <2 months; and (II) >= 50% change in tumor growth rate compared with before ICI initiation. ED was defined as overall survival (OS) <3 months. Tissue samples from 18 FP/HPD/ED patients and 5 partial response (PR) patients were subjected to genomic profiling. Genomic data from 693 tumor mutational burden- and histology-matched lung cancer samples were retrieved from an internal database as a control. Results: FP, HPD, and ED occurred in 7.21%, 9.38%, and 11.97% patients, respectively. The progression-free survival was comparable among the 3 groups. The median overall survival for FP, HPD, and ED were 3.19, 11.2, and 1.84 months, respectively. The genomic landscape revealed 1 EGFR amplification, 1 ALK fusion, 6 KRAS mutations, 1 ERBB2 amplification, 1 MET amplification, and 1 RET fusion among the 18 patients with FP/HPD/ED. Compared with the Control group, ED patients showed higher mutation frequencies for KRAS (P<0.01), CDKN1B (P<0.01), and NTRK1 (P=0.04). Mutations in RAD54L (P=0.018) and MYC (P=0.04) were more common in FP patients; HPD patients showed more frequent RAD54L mutations (P<0.001). Conclusions: We demonstrated different genomic characteristics across different progression patterns following ICI treatment, which might assist clinicians in the prediction of a patient's response, identifying candidates for more effective ICI therapy.