Amiodarone therapy and risk of primary lung cancer; insights from a real-life contemporary population study

Abstract

Abstract Background Amiodarone is the most effective antiarrhythmic therapy to maintain sinus rhythm in atrial fibrillation (AF) patients. There is a controversy regarding the association of amiodarone therapy with an increased risk of lung cancer. We aimed to assess the relationship between amiodarone treatment and primary lung cancer (PLC) incidence in a contemporary population of AF patients. Methods This is a retrospective study among patients from the Southern district of Clalit Health Services in Israel between 1st January 1999 and 1st July 2021. The study population comprised solely of patients with atrial fibrillation since this is the most prevalent condition for prescribing amiodarone. Patients treated with amiodarone were matched in a 1:1 manner with controls not exposed to amiodarone based on age, sex, living area, and atrial fibrillation diagnosis duration. To avoid immortal time bias, follow-up times of the matched controls were left adjudicated to truncate the amiodarone-free period of their matched cases. Patients with prior PLC were excluded. The primary outcome was the incidence of any PLC. We performed survival analyses using multivariable Cox regression controlling for potential confounders. Results During the study period, 7,048 patients who initiated amiodarone treatment were identified, of whom 6,449 were matched with controls. The mean age was 71.5 years, and 46.4% were women among the matched population. The amiodarone-treatment group had higher rates of smoking (30.3% vs. 26.8%), prior non-lung-related malignancy (14.9% vs. 13.5%), rheumatic disease (4.5% vs. 3.5%), diabetes (38.3% vs. 32.6%), history of myocardial infarction (316% vs. 19.6%), renal failure (17.2% vs. 10.8%) and congestive heart failure (33.8% vs. 20.7%); p<0.05 for all. During the study, with a mean follow-up time of 5.7 years, PLC occurred in 94 (1.5%) patients in the amiodarone group and 79 (1.2%) patients in the control group [adjusted-hazard-ratio (aHR) 1.10 (confidence-interval 95% (CI95%, 0.78,1.55), p=0.6]. Factors independently associated with increased PLC risk were smoking [aHR 4.25 CI95% (3.01,5.99), p<0.001], non-lung-related malignancy [aHR 1.56, CI95% (1.00,2.42), p=0.050], and peripheral vascular disease [aHR 1.77, CI95% (1.16,2.71), p=0.008] Conclusion Chronic amiodarone treatment is not associated with increased risk for PLC. Smoking and prior non-lung-related malignancy and peripheral vascular disease were independently associated with increased PLC risk, suggesting that such patients should undergo closer surveillance for early detection of PLC. Funding Acknowledgement Type of funding sources: None.