Abstract
AIMS/INTRODUCTION: This study was designed and carried out to investigate the association of dipeptidyl peptidase-4 inhibitor (DPP-4i) use with pancreatic cancer (PC) in individuals with diabetes in Japan. MATERIALS AND METHODS: The JMDC Claims Database, which contains the medical and prescription information of Japanese employment-based health insurance programs, was used. The primary outcome was duration to the first occurrence of PC (International Classification of Diseases 10th Revision code C25), both all and hospitalized, from prescription of DPP-4is or other oral glucose-lowering agents (GLAs). RESULTS: Individuals with diabetes who received DPP-4is (n = 61,430) or other oral GLAs (n = 83,304) were analyzed. Follow-up periods (median [interquartile range]) were 17 months (8-33) for DPP-4is and 14 months (7-28) for other oral GLAs. Kaplan-Meier curve analysis to determine the duration of first use of DPP4i or other oral GLA to diagnosis of PC disclosed no differences between the two groups in duration to all or hospitalized PC (log-rank test: all, P = 0.7140; hospitalized, P = 0.3446). Cox proportional hazards models showed that use of DPP-4is did not affect the PC risk adjusted for medications, age, sex and risk comorbidities (all, hazard ratio 1.1, 95% confidence interval 0.8-1.3, P = 0.6518; hospitalized, hazard ratio 1.1, 95% confidence interval 0.8-1.4, P = 0.6662). Similar results were obtained when individuals with ≥2 years oral GLA treatment and those with medical checkup data (e.g., smoking or drinking habit) available were analyzed. CONCLUSION: This database study shows that there is not a significant PC risk due to DPP-4i treatment in individuals with diabetes in Japan, but larger studies with longer follow up are required to confirm these findings.