Tirzepatide matches dulaglutide on heart outcomes but misses superiority in SURPASS-CVOT

In the SURPASS-CVOT trial, tirzepatide (Mounjaro) prevented cardiovascular death, heart attack and stroke at rates similar to dulaglutide (Trulicity) in adults with type 2 diabetes and established cardiovascular disease, but it did not reach statistical superiority over the older drug, investigators reported in full on Sept. 18 at the European Association for the Study of Diabetes annual meeting in Vienna ^[1][2].

The trial enrolled 13,165 adults at 640 sites in 30 countries between May 29, 2020, and June 27, 2022 ^[1][2]. All were at least 40 years old with type 2 diabetes, atherosclerotic cardiovascular disease and an HbA1c between 7% and 10.5% ^[2]. They were randomly assigned to once-weekly tirzepatide at the maximum tolerated dose (5 mg, 10 mg or 15 mg) or once-weekly dulaglutide 1.5 mg, with sham dose escalation in the dulaglutide group ^[1][2]. Everyone received standard care for their heart disease, and glucose treatment could be adjusted during the trial with non-incretin drugs ^[1]. About 86% were taking a statin, roughly 30% an SGLT2 inhibitor and about 49% insulin at the start ^[2]. Median follow-up was four years, and median study drug exposure was about 47.6 months ^[1][2].

The primary endpoint was the first occurrence of cardiovascular death, myocardial infarction or stroke. Tirzepatide met the bar for noninferiority to dulaglutide (hazard ratio 0.92; 95.3% CI, 0.83-1.01; P for noninferiority = .003) but did not meet superiority (P = .086) ^[1]. In the presentation, myocardial infarction occurred in 4.7% of the tirzepatide group versus 5.4% of the dulaglutide group, stroke in 3.5% versus 3.8%, cardiovascular death in 5.6% versus 6.2%, and death from any cause in 8.6% versus 10.2% ^[1].

On secondary outcomes, all-cause mortality was lower with tirzepatide (HR 0.84; 95% CI, 0.75-0.94), as was the expanded composite of cardiovascular death, heart attack, stroke or coronary revascularization (HR range 0.81-0.96) ^[1]. For individual components, the trial trended toward lower risk of heart attack (HR 0.86; 95% CI, 0.74-1.00), stroke (HR 0.91; 0.76-1.09), cardiovascular death (HR 0.89; 0.77-1.02) and cardiovascular death plus heart failure hospitalization or urgent visit (HR 0.91; 0.81-1.03) ^[1]. Nicholls also described favorable effects on HbA1c, body weight and estimated kidney filtration rate compared with dulaglutide ^[1].

An indirect, model-based analysis that matched SURPASS-CVOT participants to patients from the REWIND trial of dulaglutide versus placebo estimated a 28% lower risk of the primary outcome for tirzepatide compared with a hypothetical placebo (HR 0.72; 95% CI, 0.55-0.94; P = .02) ^[1]. In REWIND, dulaglutide's effect on major adverse cardiovascular events was modest (HR 0.88; 95% CI, 0.79-0.99) ^[1].

Serious adverse events occurred in about 32% of both groups. Nausea was reported by 25.1% of the tirzepatide group and 22.4% of the dulaglutide group, and diarrhea by 24.8% and 19.1% ^[1]. Discontinuation because of an adverse event occurred in 13.3% versus 10.2% ^[2]. "Safety outcomes were similar in the two groups, although an increase in gastrointestinal adverse events was observed in the tirzepatide group," Nicholls said ^[1].

Eldad Einav, MD, a cardiologist who reviewed the trial, wrote that the design leans on indirect comparisons and statistical assumptions, and that dosing favored tirzepatide, which was titrated to as much as 15 mg, while dulaglutide stayed at a fixed 1.5 mg ^[1]. He wrote that the 16% lower risk of all-cause death was a secondary outcome not adjusted for multiple comparisons, came without a clear reduction in cardiovascular death or major adverse cardiovascular events, and could reflect non-cardiovascular factors or chance ^[1]. He added that the FDA typically grants a "reduces risk of cardiovascular death, MI or stroke" claim when a randomized trial shows superiority over placebo, and that no diabetes drug has received such an indication based on noninferiority to an active comparator ^[1].

Why it matters for patients

For people with type 2 diabetes and existing heart disease, the trial offers direct evidence about tirzepatide's cardiovascular safety in that group: it did not produce more heart attacks, strokes or cardiovascular deaths than dulaglutide, an older GLP-1 drug with a known cardiovascular benefit ^[1]. It did not show that tirzepatide prevents those events better than dulaglutide ^[1].

The lower all-cause death rate is the most striking number, but it was a secondary finding, was not adjusted for multiple comparisons, and one outside reviewer said a direct cardiovascular explanation is uncertain ^[1]. Gastrointestinal side effects were more common with tirzepatide, and roughly one in three participants in each group had a serious adverse event ^[1].

The trial compared two active drugs rather than testing either against placebo, so the size of any benefit relative to no GLP-1 treatment was not measured directly here; the 28% figure comes from a model that matched patients across two different trials ^[1]. Whether this data changes US prescribing labels has not been disclosed.

What happens next

Einav wrote that stronger evidence would probably require superiority on a prespecified cardiovascular endpoint in additional outcomes research, such as SURMOUNT-MMO, plus consistent real-world data and analyses addressing differences in patient populations and background therapy ^[1]. No results or completion dates for that study were reported. No company statement about seeking a new cardiovascular indication based on SURPASS-CVOT, and no FDA decision date, has been reported ^[1].