Background: Whether metformin may reduce pancreatic cancer risk is controversial and has rarely been studied in Asian populations. This study evaluated pancreatic cancer risk in type 2 diabetes patients with and without metformin treatment. Methods: An original cohort of 16232 never users and 153408 ever users of metformin who were newly diagnosed of type 2 diabetes between 1999 and 2005 were enrolled from Taiwan's National Health Insurance. A propensity score matched cohort of 16232 ever users and 16232 never users were derived from this original cohort. The patients were followed up until December 31, 2011 and analyses were conducted in both the original cohort and the matched cohort. Hazard ratios were estimated by Cox regression incorporated with the inverse probability of treatment weighting using propensity score. Results: In the original cohort, the incidence rate of pancreatic cancer in never users and ever users were 43.59 and 28.85 per 100000 person-years, respectively, with an overall hazard ratio of 0.658 (95% confidence interval: 0.438-0.988, P=0.0436). While compared to never users, the hazard ratio (95% confidence interval) for cumulative duration of metformin use in the first (<22.10 months), second (22.10-46.70 months) and third tertile (>46.70 months) was 1.445 (0.950-2.196), 0.595 (0.384-0.922) and 0.157 (0.094-0.264), respectively. Results in the matched cohort were very similar, with an overall hazard ratio of 0.492 (0.252-0.961) and the hazard ratio in the respective tertile of cumulative duration was 0.900 (0.368-2.198), 0.648 (0.266-1.581) and 0.094 (0.013-0.692). Conclusions: Metformin significantly reduces pancreatic cancer risk in a dose-response pattern.

Citation Format: Chin-Hsiao Tseng. Metformin and pancreatic cancer risk in Taiwanese patients with type 2 diabetes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2236.