Clinical and genomic determinants of early treatment discontinuation in non-small-cell lung cancer: an analysis of the AACR Project GENIE Biopharma Collaborative NSCLC dataset.

Background: Early treatment discontinuation (ETD) is common in cancer treatment but understudied. We sought to identify clinical and treatment-related factors associated with ETD in a real-world cohort of patients with non-small-cell lung cancer (NSCLC). Patients and methods: We analyzed lines of therapy (LoTs) including drugs intended for an indefinite treatment sourced from the American Association for Cancer Research Project Genomics Evidence Neoplasia Information Exchange Biopharma Collaborative NSCLC cohort. Associations between treatment exposure and ETD were evaluated using multivariable logistic regression adjusting for age at LoT initiation, sex, stage, and context (first-line, second-line, third-line, or later). Sensitivity analyses examined mutation-specific targeted therapy exposures and ETD within 14 days of death. Results: ETD occurred in 883 of 2098 eligible LoTs (42%). Programmed cell death protein 1/programmed death-ligand 1 inhibitor exposure was associated with higher odds of ETD [odds ratio (OR) = 2.07, 95% confidence interval (CI) 1.27-3.38], whereas small-molecule inhibitor exposure was associated with lower odds (OR = 0.43, 95% CI 0.26-0.70). Context showed a graded increase in ETD risk: second-line versus first-line (OR = 1.46, 95% CI 1.14-1.87) and third-line or later versus first-line (OR = 2.63, 95% CI 2.09-3.32). In 302 LoTs with an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) in patients with Conclusions: ETD in NSCLC is associated with treatment modality, timing, and biology. The drivers of ETD are complex and span medical, social, and psychological domains. This study highlights ETD as a measurable quantity and provides a framework for future studies.