Is there a link between liraglutide and pancreatitis? A post hoc review of pooled and patient-level data from completed liraglutide type 2 diabetes clinical trials.

OBJECTIVE To report the incidence of pancreatitis in type 2 diabetes trials of liraglutide and details of all pancreatitis cases. RESEARCH DESIGN AND METHODS Data from Novo Nordisk-sponsored trials with liraglutide (phase 2 and 3; NN2211 identifiers) completed by 19 April 2013 were pooled. All pancreatitis cases were reviewed. RESULTS Total exposure to liraglutide and active comparators was 5,021 and 1,354 patient-years, respectively (n = 6,345 and 1,846, respectively). Eight cases of acute pancreatitis (AP) with liraglutide and one with any comparator (glimepiride) were found. The incidence of AP was 1.6 cases/1,000 patient-years exposure (PYE) for liraglutide vs. 0.7 cases/1,000 PYE for total active comparators. One of the eight AP cases reported with liraglutide did not meet diagnostic criteria for AP. In six of these eight cases, recognized risk factors for AP were present and/or the onset of AP occurred >6 months after liraglutide initiation. All patients were receiving multiple medications. Four cases of chronic pancreatitis (CP) with liraglutide and none with comparators were found. One of these four cases fulfilled diagnostic criteria for CP; these criteria were not met or information was missing in the remaining three. CONCLUSIONS Based on the small number of cases observed, the incidences of reported AP and CP were numerically greater with liraglutide than with comparators. Not all cases fulfilled diagnostic criteria, and confounding variables were present in 75% of the AP cases with liraglutide therapy, precluding firm conclusions.