Total number of circulating tumor DNA (ctDNA) genomic alterations and outcome after checkpoint inhibitor-based immunotherapy.

Abstract

e14504 Background: High tumor mutational burden on tissue biopsy correlates with efficacy of checkpoint immunotherapy in cancers, such as non-small cell lung cancer and melanoma. However, the relationship between mutational burden detected in non-invasive surrogates, such as blood-derived ctDNA genomic alterations, and outcome after immunotherapy, has not been evaluated. Methods: We analyzed 69 patients with diverse malignancies whose plasma-derived ctDNA had undergone next generation sequencing (NGS) (54 to 70 genes: Guardant Health) and who had received checkpoint inhibitor-based immunotherapy. Data was assessed to correlate total number of alterations (characterized alterations and variants of unknown significance, VUS) in ctDNA and outcome: stable disease (SD) >6 months, complete or partial remission (CR or PR), progression-free and overall survival (PFS and OS), from date of immunotherapy initiation. Results: The69 patients had 23 different diagnoses. All individuals received checkpoint inhibitor-based immunotherapy, with the majority receiving anti-PD1/PD-L1 monotherapy. Evaluable patients with ≥6 vs. <6 total ctDNA alterations had significantly higher rates of SD≥6 months/CR/PR: 40.9% (9/22) vs. 15.9% (7/44) (p=0.035) (Table). Median PFS was 2.85 months vs. 2.1 months (p=0.046). OS was not statistically different. A landmark analysis at 2 months showed that median PFS in the ≥6 alterations group was 23.2 months for responders (CR/PR) vs. 2.3 months for non-responders (p=0.0006); for the group with <6 alterations, the median PFS was 12.6 vs. 3.3 months (p = 0.0006). Conclusions: Total number of ctDNA genomic alterations merits further exploration as a potential predictive biomarker for outcome with checkpoint inhibitor-based immunotherapy. Achieving response to checkpoint-based immunotherapy strongly predicts PFS from the 2-month mark. [Table: see text]