Real-World Costs of Adverse Events in First-Line Treatment of Metastatic Non-Small Cell Lung Cancer.

Abstract

Non-small cell lung cancer (NSCLC) is the most common form of lung cancer in the United States. Immunotherapies and cytotoxic chemotherapies used to treat advanced NSCLC carry a substantial risk of adverse events (AEs), but real-world data on the incidence and costs associated with the unique AE profiles of these treatments are sparse. To examine the AE incidence and costs among patients initiating non-driver mutation-targeted first-line therapy for metastatic NSCLC (mNSCLC) in clinical practice. This was a retrospective administrative claims study conducted among commercial and Medicare Advantage health plan members who initiated first-line, nontargeted systemic anti-NSCLC therapy between January 1, 2008, and February 28, 2018. Patients were assigned to mutually exclusive treatment cohorts (cytotoxic chemotherapy [CHEM], immuno-oncology agents [IO], or immuno-oncology + cytotoxic chemotherapy [IO-CHEM]) and were observed from the index date (start of first-line therapy) through the earliest of health plan disenrollment, death, or March 31, 2018. AE incidence rates and associated health care costs were measured from the index date through the earliest of the start of a new therapy, 180 days after the end of first-line therapy, or the end of the study period. The factors influencing whether patients incurred high AE-related health care costs were assessed using multivariable models adjusted for patient demographic and clinical characteristics. The final study population (mean [SD] age 68.6 [9.5] years, 53.9% male) included 8,818 in the CHEM cohort, 482 in the IO cohort, and 412 in the IO-CHEM cohort. Overall, 74.4% had at least 1 AE during follow-up. The AE incidence rate was lowest for the IO cohort, with incidence rate ratios (95% CI) of 1.4 (1.3-1.6) for the CHEM cohort and 1.4 (1.2-1.6) for the IO-CHEM cohort. Mean AE-related costs were lowest for the IO cohort ($16,319) and highest for the CHEM cohort ($23,009; P < 0.001). In the multivariable analysis, the odds of incurring any AE costs were similar for the IO and IO-CHEM cohorts compared with the CHEM cohort (OR = 0.82; P = 0.135 and OR = 0.98; P = 0.888, respectively). Among patients who incurred AE costs, those in the IO cohort were less likely than those in the CHEM cohort to have high costs (OR = 0.60; P = 0.030); the difference between the IO-CHEM and CHEM cohorts was not statistically significant. Among real-world patients initiating nontargeted first-line therapy for mNSCLC, those receiving immunotherapy experienced fewer AEs and had lower total AE-related costs than those treated with cytotoxic chemotherapy. Immunotherapy-treated patients were no more likely than chemotherapy-treated patients to incur AE-related costs and were less likely to have high AE costs if they incurred any at all. These findings indicate that immunotherapy-related AEs are not a differentiating factor in cost of care for this patient population in clinical practice. This study was sponsored by AstraZeneca. Ryan is an employee of AstraZeneca. Engel-Nitz, Johnson, and Bunner are employees of Optum, which was contracted by AstraZeneca to conduct this study, and shareholders in UnitedHealth Group. Engel-Nitz has also worked on cancer-related studies for which Optum received funding from Bayer AG, Clovis Oncology, Eli Lilly, EMD Serono, Exact Sciences, Janssen, and Novartis. Johnson worked on cancer-related studies for which Optum received funding from Eli Lilly, Medtronic, Sanofi, and UnitedHealthcare. Bunner has worked on cancer-related studies for which Optum received funding from Celgene and Incyte.