Decaffeinated coffee consumption and risk of total and site-specific cancer

Abstract

Background: Coffee is generally considered safe regarding cancer risk. However, concerns have emerged over methylene chloride, a chemical used in decaffeination, due to its carcinogenic properties. The potential cancer risk from methylene chloride residue in decaffeinated coffee remains unclear.

Patients and methods: This prospective cohort study included 75 988 women (Nurses' Health Study, 1984-2020) and 45 349 men (Health Professionals Follow-up Study, 1986-2020). Decaffeinated coffee consumption was assessed at baseline and every 4 years using a validated food frequency questionnaire. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of total and 14 site-specific cancers associated with decaffeinated coffee consumption, adjusted for regular coffee intake and other potential confounding variables.

Results: During up to 36 years of follow-up, we documented 34 120 incident cancer cases (22 688 in women and 11 432 in men). Overall, decaffeinated coffee intake was not associated with higher total cancer risk [per 1 cup/day higher intake, HR 1.00 (95% CI 0.98-1.01)]. For specific cancer type, an inverse association was observed for colorectal [HR 0.96 (95% CI 0.92-0.99)] and aggressive prostate [HR 0.93 (95% CI 0.87-0.99)] cancer. An observed higher lung cancer risk for decaffeinated coffee attenuated to null when restricted to never smokers. However, we observed elevated bladder cancer risk among male never smokers: compared with nondrinkers, 0.1-0.9, 1-1.9, 2-2.9 and ≥3 cups/day of decaffeinated coffee intake were associated with HRs of 1.30 (95% CI 1.01-1.68), 1.42 (95% CI 1.00-2.03), 1.43 (95% CI 0.91-2.23), and 1.79 (95% CI 0.92-3.50), respectively, with borderline significant linear trends (P_trend = 0.06). This positive association remained robust in various sensitivity analyses and became even stronger with an 8-year lag. No association between decaffeinated coffee and bladder cancer was observed among women (P_{sex-heterogeneity} = 0.03).

Conclusions: Higher decaffeinated coffee intake was not associated with total cancer risk. However, the observed significant higher risk of bladder cancer in men warrants further studies.

Keywords: bladder cancer; cancer risk; cohort studies; decaffeinated coffee; epidemiology.