Higher carbohydrate quality, but not quantity, is associated with lower lung cancer risk in a population-based prospective cohort study

This is the authors' abstract. We don't add key points for this paper.

Background The role of diet in lung cancer is increasingly recognized, but it remains unclear whether carbohydrate quality or carbohydrate quantity is more relevant to risk. We therefore examined the associations of the carbohydrate quality index (CQI) and low-carbohydrate diet score (LCD) with incident lung cancer in a population-based prospective cohort. Methods We included 101,755 participants from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. CQI and LCD were used to assess carbohydrate quality and quantity, respectively. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for lung cancer across quartiles of CQI and LCD. Restricted cubic spline, subgroup, and sensitivity analyses were conducted to assess dose–response relationships and the robustness of the findings. Results During 897,809.4 person-years of follow-up, 1,706 incident lung cancer cases were identified. Higher CQI was significantly associated with a lower risk of lung cancer after multivariable adjustment. In the fully adjusted model, compared with the lowest quartile, the HRs (95% CIs) for lung cancer were 0.82 (0.72, 0.93), 0.78 (0.68, 0.89), and 0.73 (0.64, 0.84) for the second, third, and fourth quartiles of CQI, respectively ( P for trend <0.001). By contrast, LCD was not significantly associated with lung cancer risk after multivariable adjustment, with corresponding HRs (95% CIs) of 0.93 (0.81, 1.07), 1.02 (0.89, 1.17), and 1.09 (0.95, 1.25) across increasing quartiles ( P for trend = 0.117). Restricted cubic spline analyses showed a significant non-linear inverse association for CQI ( P for nonlinearity = 0.006), but not for LCD ( P for nonlinearity = 0.150). The inverse association for CQI remained robust in multiple sensitivity analyses, including further adjustment for LCD and alternative smoking exposure models. Conclusion Higher carbohydrate quality, but not carbohydrate quantity, was associated with a lower risk of lung cancer, suggesting a greater relevance of carbohydrate quality to lung cancer risk.