(067) Risk of Incident Inflammatory Bowel Disease in New Users of Dipeptidyl Peptidase 4 Inhibitors and Sodium Glucose Cotransporter 2 Inhibitors in the FDA Sentinel System

Background: Previous studies reported inconsistent findings of the risk of inflammatory bowel disease (IBD: Crohn’s disease [CD], ulcerative colitis [UC], and indeterminate colitis) following dipeptidyl peptidase-4 inhibitors (DPP-4i) exposure.

Objectives: To describe characteristics of new users of DPP-4i and sodium glucose cotransporter 2 inhibitors (SGLT-2i) and to estimate the overall risk of incident IBD in order to inform future inferential analyses.

Methods: We identified eligible DPP-4i or SGLT-2i initiators aged ≥18 years in 6 data partners of U.S. FDA Sentinel System from March 2013 to December 2022, who met the following criteria in the year prior to the first dispensing of a study drug (index date): presence of type 2 diabetes; use of oral antidiabetic drugs; and absence of IBD, IBD treatment, diverticulitis, colitis, intestine or colon surgery, endoscopy, or use of study drugs. IBD was identified by an ICD9 or ICD10 diagnosis code, preceded by endoscopy and biopsy procedures and followed by IBD treatment. Follow-up began on the index date and ended at the first occurrence of an IBD event, exposure to the other study drug, treatment cessation, death, insurance disenrollment, or study end date. We excluded IBD events that occurred in the first 180 days of follow-up to rule out latent IBD cases that may have presented before the index date. We estimated IBD risk, overall and by age and sex, in a combined cohort of DPP-4i and SGLT-2i to maintain the blinding for inferential analyses.

Results: In the DPP-4i (n=695,410) and SGLT-2i (n=577,383) cohorts, mean (±standard deviation) age was 66.2 (±10.9) and 62.6 (±9.9) years and 42% and 30% were aged ≥70 years, respectively. In the DPP-4i cohort, 48% were male and 61% had index exposure in 2014 to 2018. In the SGLT-2i cohort, 57% were male and 62% had index exposure in 2019 to 2022. Relative to SGLT-2i users, DPP-4i users had a higher score on the Diabetes Complications Severity Index (1.9 ±2.0 vs 1.6 ±1.8, SMD= 0.158) and less concurrent use of insulin (21% vs 35%). Median at-risk period was 302 (interquartile range [IQR] 163 to 631) days in DPP-4i users and 270 (IQR 150 to 523) days in SGLT-2i users. In the combined cohort, 242 IBD events were identified, corresponding to 19 cases/100,000 persons. Risk of incident IBD was 20 cases/100,000 persons for men and 18 cases/100,000 persons for women. For adults aged 18-49, 50-59, 60-69, and ≥70 years, IBD risk ranged from 16 to 23 cases/100,000 persons. We observed similar trends in risks of UC and CD by sex or age.

Conclusions: Descriptive analysis did not suggest the risk of incident IBD in DDP-4i or SGLT-2i users varied by sex or age. Baseline characteristics differed between these two user cohorts. These data will inform the design of future inferential analyses to assess DPP-4i use in relation to IBD.