Temporal trends in treatment-related cardiopulmonary disease-specific mortality in NSCLC based on pathological subtypes: a retrospective population-based cohort study.

Background: While cancer-specific mortality has decreased with therapeutic advances, a growing proportion of survivors are at risk of treatment-related adverse effects (AEs).

Objectives: To assess cardiopulmonary disease-specific mortality risks in patients with non-small-cell lung cancer (NSCLC) based on pathological subtypes.

Design: This was a retrospective cohort study. Mortality data from the Surveillance, Epidemiology, and End Results Program database and AE data from the FDA Adverse Event Reporting System were analyzed in NSCLC patients. In addition, an independent validation cohort comprising 161 NSCLC patients was enrolled.

Methods: The piecewise regression was established using the Joinpoint software to characterize temporal trends. Univariate and multivariate analyses for specific mortality were performed using the Cox risk regression model.

Results: From 2012 to 2016, patients with lung adenocarcinoma (LADC) showed a 2.06% annual reduction in mortality. Conversely, this population exhibited a 3.82% yearly increase in cardiovascular disease (CVD)-specific mortality and a 6.54% rise in pulmonary disease-specific mortality during the same period. Patients with lung squamous cell carcinoma (LSCC) may have benefited more from the initiation of immunotherapy, with a gradual decrease in incidence (2016-2019, -5.19%). The incidence of large-cell lung cancer (LCLC) declined significantly (2008-2012, -14.52%; 2012-2016, -7.93%; 2016-2019, -6.42%), with decrease in overall mortality (2008-2012, -10.63%; 2012-2016, -8.76%; 2016-2019, -10.27%) and cancer-specific mortality (2008-2012, -12.17%; 2012-2016, -10.35%; 2016-2019, -10.78%). LCLC showed no significant rise in CVD-specific mortality (2012-2016: CVD-specific mortality, -4.83%; pulmonary disease-specific mortality, -0.82%. 2016-2019: CVD-specific mortality, -21.82%; pulmonary disease-specific mortality, -12.78%) compared to LSCC and LADC. The reporting odds ratio (ROR = 47.80, 95% CI, 24.44-93.50) of endocarditis of LADC was the top-reported AE. Multivariate analysis showed that immune checkpoint blockade was associated with increased CVD death (HR = 4.602, 95% CI, 1.154-18.359, p = 0.031).

Conclusion: Our analyses revealed a temporal increase in cardiopulmonary disease-specific mortality among NSCLC patients during the study period. This trend coincided with the introduction of targeted therapies and immunotherapies, particularly in LADC patients. CVD-specific mortality risk requires extensive attention during anticancer therapy, regardless of the geographical population.