EXSCEL

EXSCEL is the trial that keeps the GLP-1 class honest. It enrolled 14,752 adults with type 2 diabetes — 73.1% with existing cardiovascular disease — and randomized them to once-weekly exenatide or placebo, following them for a median of 3.2 years. Holman and colleagues published it in the New England Journal of Medicine in 2017. It is one of the largest trials in the class.

The primary composite of cardiovascular death, nonfatal heart attack and nonfatal stroke occurred in 11.4% on exenatide versus 12.2% on placebo, a hazard ratio of 0.91. That met the noninferiority bar comfortably but missed superiority, with a p-value of 0.06 — close enough to be tantalizing, not close enough to claim a benefit.

The safety picture was clean. Rates of cardiovascular death, heart attack, stroke, heart-failure hospitalization and acute coronary syndrome did not differ significantly, and neither did acute pancreatitis, pancreatic cancer or medullary thyroid carcinoma. For a class carrying a rodent-based boxed warning, that last sentence is worth reading twice.

EXSCEL matters because it disproves the idea that cardiovascular protection is automatic across GLP-1 drugs. LEADER, SUSTAIN-6 and REWIND showed benefit; EXSCEL and ELIXA were neutral. Molecular structure, half-life and trial population all differ, and a class effect that some members do not share is not a class effect. When a clinician says a particular GLP-1 has cardiovascular evidence, that is a statement about a specific drug.

Worth asking your clinician: whether the specific product you are prescribed carries a cardiovascular indication on its label, and whether that should influence the choice given your own risk profile. Our EXSCEL deep dive and cardiovascular study roundup lay the trials out side by side.