Bidirectional Association Between Cardiovascular Disease and Lung Cancer in a Prospective Cohort Study

Abstract

Introduction: The study aimed to prospectively investigate the bidirectional association between cardiovascular disease (CVD) and lung cancer, and whether this association differs across genetic risk levels.

Methods: This study prospectively followed 455,804 participants from the United Kingdom Biobank cohort who were free of lung cancer at baseline. Cox proportional hazard models were used to estimate the hazard ratio (HR) for incident lung cancer according to CVD status. In parallel, similar approaches were used to assess the risk of incident CVD according to lung cancer status among 478,756 participants free of CVD at baseline. The bidirectional causal relations between these conditions were assessed using Mendelian randomization analysis. Besides, polygenic risk scores were estimated by integrating genome-wide association studies identified risk variants.

Results: During 4,007,477 person-years of follow-up, 2006 incident lung cancer cases were documented. Compared with participants without CVD, those with CVD had HRs (95% confidence interval [CI]) of 1.49 (1.30-1.71) for NSCLC, 1.80 (1.39-2.34) for lung squamous cell carcinoma (LUSC), and 1.25 (1.01-1.56) for lung adenocarcinoma (LUAD). After stratification by smoking status, significant associations of CVD with lung cancer risk were observed in former smokers (HR = 1.44, 95% CI: 1.20-1.74) and current smokers (HR = 1.38, 95% CI: 1.13-1.69), but not in never-smokers (HR = 0.98, 95% CI: 0.60-1.61). In addition, CVD was associated with lung cancer risk across each genetic risk level (p_{heterogeneity} = 0.336). In the second analysis, 32,974 incident CVD cases were recorded. Compared with those without lung cancer, the HRs (95% CI) for CVD were 2.33 (1.29-4.21) in NSCLC, 3.66 (1.65-8.14) in LUAD, and 1.98 (0.64-6.14) in LUSC. In particular, participants with lung cancer had a high risk of incident CVD at a high genetic risk level (HR = 3.79, 95% CI: 1.57-9.13). No causal relations between these conditions were observed in Mendelian randomization analysis.

Conclusions: CVD is associated with an increased risk of NSCLC including LUSC and LUAD. NSCLC, particularly LUAD, is associated with a higher CVD risk. Awareness of this bidirectional association may improve prevention and treatment strategies for both diseases. Future clinical demands will require a greater focus on cardiac oncology.

Keywords: Bidirectional relationship; Cardiovascular disease; Lung neoplasms; Mendelian randomization; Polygenic risk score.