Memorandum

Dr. Peter Butler advocates for GLP-1 agonists as potential second-line therapy based on new JAMA article that finds no association between incretins and increased risk of acute pancreatitis – August 4, 2016

Executive Highlights

  • Cohort study involving 1.53 million type 2 diabetes patients finds no additional risk of acute pancreatitis in those taking GLP-1 agonists or DPP-4 inhibitors versus counterparts on other combinations of oral diabetes mediations (HR, 1.03; 95% CI, 0.87-1.22).
  • Dr. Peter Butler wrote an editorial highlighting this “impressively large” study as convincing evidence of the “reassuringly low” risk of pancreatitis associated with incretins. He advocates GLP-1 agonists as “logical choice” of second-line diabetes therapy (after metformin), a notable statement given his past statements on the opposite side of the incretin-pancreatitis debate.

A study published in the August 1 JAMA from McGill University found no association between the use of incretins and increased risk of acute pancreatitis, compared with the use of two or more other oral diabetes medications. The multicenter, retrospective, observational, population-based case-controlled cohort study included over 1.53 million patients with type 2 diabetes (51% male, 49% female; mean age 56.6 years) who began a diabetes pharmacotherapy regimen of two or more medications between January 2007 and June 2013. Data was collected from health records across seven facilities in the US, Canada, and the UK. Over the course of the study, 5,165 patients across the entire cohort were hospitalized for acute pancreatitis, an incidence rate of 1.49 per 1,000 person-years. The incidence rate was no higher in patients taking incretin-based drugs compared to patients taking other combinations of oral diabetes medications, though incretin therapy trended toward very slightly increased risk (HR=1.03; 95% CI: 0.87-1.22). By drug class, neither treatment with DPP-4 inhibitors nor GLP-1 agonists demonstrated a significantly increased risk for pancreatitis (HR for DPP-4 inhibitors=1.09; 95% CI: 0.86-1.22 and HR for GLP-1 agonists=1.04; 95% CI:0.81-1.35).  

In an accompanying editorial, the controversial Dr. Peter Butler (UCLA, Los Angeles, CA) highlighted this “impressively large” study as a clear piece of evidence pointing to a minimal increased risk of pancreatitis associated with incretins – this comes after years of his casting doubt on the safety of incretins. Furthermore, in a major turnaround, Dr. Butler advocates for GLP-1 agonists as “logical choice” of second-line diabetes therapy (after metformin) in patients with type 2 diabetes and obesity, noting that all available evidence points to a “reassuringly low” risk of pancreatitis. Indeed, this view is particularly notable coming from “lone voice” Dr. Butler, given that he was partially responsible for fanning the flames of the incretin-pancreatitis controversy, often defending the hypothesis of a link between incretins and pancreatitis at scientific meetings and in journals. In 2013, Dr. Butler’s laboratory also published a morphological study suggesting a connection between incretin therapies and pancreatitis. For more background, see our timeline of the controversy, which we have been tracking since it began.

Given that even Dr. Butler finds the evidence on pancreatitis and incretins to-date sufficiently reassuring, we assume the controversy can finally largely be put to rest. Clinical evidence increasingly suggests a very positive benefit/risk profile for GLP-1 agonists in particular, with strong A1c efficacy, low risk of hypoglycemia, weight loss, and, now with the LEADER results, cardioprotective potential. At this year’s Keystone 2016 conference, Dr. Ralph DeFronzo went as far as to describe the GLP-1 agonist class as “miraculous” for its ability to protect and restore beta cell function. We expect this class will become increasingly popular as next-generation GLP-1 agonists with innovative dosing arrive on the market (such as Novo Nordisk’s injectable and oral semaglutide formulations and Intarcia’s implantable exenatide mini-pump ITCA 650) and they become easier to prescribe and take. Although we would be disappointed if any lingering controversy over an increased risk of pancreatitis created a significant hurdle toward adoption of this class among patients, providers, and payers, we wouldn’t be surprised if this “baggage” carries on a long time – Dr. Butler, when casting doubt on GLP-1 safety, made headlines in the New York Times, but we doubt his new views that he was incorrect will prompt many headlines.

  • This JAMA study is one of many pieces of clinical evidence to suggest a lack of association between incretin therapy and pancreatitis. Some notable others include:
    • A 2014 retrospective case-control study of a database in Piedmont, Italy (282,489 type 2 diabetes patients receiving treatment, 1,003 cases of acute pancreatitis) conducted by Dr. Carlo Giorda (Metabolism and Diabetes Unit, Regione Piemonte, Italy) and colleagues found no increased risk of acute pancreatitis with exposure to incretin therapy.
    • A 2014 BMJ article describing the results of a systematic review and meta-analysis of the risk of pancreatitis in type 2 diabetes patients treated with incretin therapies found no support for the assertion that incretins increase the risk of pancreatitis. The analysis pooled the results of 60 studies, including 55 randomized control trials, which collectively enrolled over 350,000 patients (~320,000 patients were from the five non-RCTs). The analysis yielded a hazard ratio of 1.11 that was not statistically significant (95% CI: 0.57 - 2.17).
    • The 2016 LEADER trial found no significant difference in incidence of acute or chronic pancreatitis between the 9,340 patients treated with Novo Nordisk’s GLP-1 agonist liraglutide versus the placebo group. Importantly, the LEADER trial was not powered to make conclusive statements about pancreatitis, but these findings nonetheless demonstrate a reassuring trend.

-- by Abigail Dove, Helen Gao, and Kelly Close