It enrolled 14,752 adults with type 2 diabetes, with or without previous cardiovascular disease, and followed them for a median of 3.2 years on exenatide 2 mg a week or placebo. A first heart attack, stroke or cardiovascular death occurred in 11.4% on exenatide and 12.2% on placebo. The drug was shown to be safe, but the benefit just missed statistical significance. It was published in the New England Journal of Medicine in 2017.1
How the trial was run
EXSCEL was a large trial, and it enrolled a broad group: 10,782 of the participants, 73.1%, had previous cardiovascular disease, while the rest had diabetes without it. Extended-release exenatide or a matching placebo was injected once a week.1
The trial tested two hypotheses together: that once-weekly exenatide was not worse than placebo for safety, and that it was better for efficacy. The main outcome was the first of three events: cardiovascular death, non-fatal heart attack or non-fatal stroke.1
What it found
- The main outcome occurred in 839 of 7,356 people on exenatide, 11.4%, and 905 of 7,396 on placebo, 12.2%.
- As rates, that is 3.7 against 4.0 events per 100 person-years, a hazard ratio of 0.91.
- Safety was shown. Superiority was not: the test for benefit gave P = 0.06, just above the usual threshold of 0.05.
- Cardiovascular death, heart attack, stroke and hospital stays for heart failure or acute coronary syndrome did not differ clearly between the groups.
Holman et al., 2017.1
How to read a near miss
A hazard ratio of 0.91 with a confidence interval running from 0.83 to 1.00 sits right at the edge. The data are compatible with a small benefit, but they do not rule out none. That is different from ELIXA, whose hazard ratio of 1.02 pointed to no effect at all.1,2 It is also weaker than the trials of liraglutide, semaglutide and dulaglutide, which found hazard ratios of 0.87, 0.74 and 0.88 and met their goals.3,4,5
The trial does not explain the difference from the others, and the drugs, doses and people all differ. Half of the participants were followed for between 2.2 and 4.4 years, the middle range around the median of 3.2. Another weekly GLP-1 medicine, efpeglenatide, met its goal in the AMPLITUDE-O trial with a hazard ratio of 0.73, in people with heart or kidney disease followed for a median of 1.81 years.1,6 A near miss like EXSCEL's is best read as: safe, and possibly modestly helpful, but not proven to protect the heart.
About exenatide
Exenatide was first sold as Byetta, a twice-daily injection with a half-life of 2.4 hours, approved in 2005.7 The twice-daily form is injected within the hour before the two main meals. The extended-release form tested in EXSCEL was given once a week instead, at a fixed 2 mg dose. A 2021 review groups once-weekly exenatide with the other long-acting GLP-1 medicines, which lower fasting blood sugar and HbA1c more than the short-acting ones.8
What it means in practice
For someone choosing a GLP-1 medicine with heart disease in mind, the trials point to specific drugs. The labels of Victoza, Ozempic and Trulicity carry approved uses for reducing major cardiovascular events, based on LEADER, SUSTAIN 6 and REWIND.3,4,5 Exenatide's trial did not prove a benefit, so that evidence does not carry over to it.
Limits
The trial was funded by Amylin Pharmaceuticals and is registered as NCT01144338.1 Our page on Ozempic and Mounjaro covers the weekly diabetes injections most used today.
Sources
- Holman RR, Bethel MA, Mentz RJ, et al. Effects of once-weekly exenatide on cardiovascular outcomes in type 2 diabetes (EXSCEL). N Engl J Med 2017;377:1228-1239. Funded by Amylin Pharmaceuticals; NCT01144338. doi:10.1056/NEJMoa1612917
- Pfeffer MA, Claggett B, Diaz R, et al. Lixisenatide in patients with type 2 diabetes and acute coronary syndrome (ELIXA). N Engl J Med 2015;373:2247-2257. doi:10.1056/NEJMoa1509225
- Marso SP, Daniels GH, Brown-Frandsen K, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes (LEADER). N Engl J Med 2016;375:311-322. doi:10.1056/NEJMoa1603827
- Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes (SUSTAIN-6). N Engl J Med 2016;375:1834-1844. doi:10.1056/NEJMoa1607141
- Gerstein HC, Colhoun HM, Dagenais GR, et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND). Lancet 2019;394:121-130. doi:10.1016/S0140-6736(19)31149-3
- Gerstein HC, Sattar N, Rosenstock J, et al. Cardiovascular and renal outcomes with efpeglenatide in type 2 diabetes (AMPLITUDE-O). N Engl J Med 2021;385:896-907. doi:10.1056/NEJMoa2108269
- BYETTA (exenatide) injection prescribing information. AstraZeneca; DailyMed, version published September 8, 2025; Initial U.S. Approval and section 12.3. dailymed.nlm.nih.gov
- Nauck MA, Quast DR, Wefers J, Meier JJ. GLP-1 receptor agonists in the treatment of type 2 diabetes: state-of-the-art. Mol Metab 2021;46:101102. doi:10.1016/j.molmet.2020.101102
Related analyses
Ozempic vs Mounjaro: what the head-to-head trial and the two labels show
When to take your GLP-1 shot: time of day, changing the day and time zones
Ozempic side effects: how common each one is by dose, and when they ease
Nausea on a GLP-1: how common, how long, when it peaks and what helps
Log your doses, blood sugar and blood pressure in one place.
Open Dose Insider