Summary: In 9650 adults with type 2 diabetes and atherosclerotic cardiovascular disease and/or chronic kidney disease, oral semaglutide lowered major adverse cardiovascular events compared with placebo overall (hazard ratio 0.86; 95% CI 0.77 to 0.96). This prespecified analysis found the benefit was consistent whether or not participants were taking an SGLT2 inhibitor at baseline (p for interaction 0.66), and the combination appeared safe.
PICO Summary
| Element | Detail |
|---|---|
| Population | 9650 adults with type 2 diabetes and established atherosclerotic cardiovascular disease and/or chronic kidney disease; international double-blind randomized controlled trial (SOUL; NCT03914326); mean follow-up 47.5 months. |
| Intervention | Oral semaglutide. Prespecified subgroups by baseline SGLT2 inhibitor use: on SGLT2i n=2596 (1296 in the semaglutide arm); not on SGLT2i n=7054 (3529 in the semaglutide arm). |
| Comparison | Matching placebo, with or without concomitant SGLT2 inhibitor (placebo arm: 1300 on SGLT2i, 3525 not on SGLT2i at baseline). |
| Outcome | First major adverse cardiovascular event (cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke). Overall: HR 0.86 (95% CI 0.77 to 0.96). On SGLT2i at baseline: 143/1296 vs 158/1300; HR 0.89 (95% CI 0.71 to 1.11). Not on SGLT2i at baseline: 436/3529 vs 510/3525; HR 0.84 (95% CI 0.74 to 0.95). p for interaction 0.66. Serious adverse event profiles were similar with or without concomitant SGLT2 inhibitor. |
Oral semaglutide CV benefit by baseline SGLT2i use (SOUL)
RCT prespecified analysis · type 2 diabetes · 47.5 mo
Oral semaglutide's reduction in major adverse cardiovascular events was consistent regardless of baseline SGLT2 inhibitor use, and the combination appeared safe.
Expert Commentary
This prespecified analysis of the SOUL trial was designed to test effect modification, not to establish efficacy within each subgroup, and on that question its message is reassuring: the overall 14 percent relative reduction in major adverse cardiovascular events was not meaningfully altered by background SGLT2 inhibitor use, with a non-significant interaction term (p 0.66). The result should be read as consistency rather than proof of benefit in every stratum. Among participants already taking an SGLT2 inhibitor the point estimate remained favourable (HR 0.89) but the confidence interval crossed 1.0 (0.71 to 1.11), so an isolated benefit in that group is not demonstrated; this subgroup was also the smaller one and the analysis was not powered for stratum-specific significance. The principal limitation is that baseline SGLT2 inhibitor use was non-randomized, so the subgroups differ in ways beyond the assigned drug and residual confounding cannot be excluded. The trial was sponsored by the manufacturer, with several authors employed by the sponsor, which warrants the usual caution even within a rigorously conducted double-blind design. Can I use this with my patients? Yes, for a patient with type 2 diabetes and high cardiovascular risk already on an SGLT2 inhibitor, these data support adding oral semaglutide without expecting attenuated cardiovascular benefit or a new safety signal. Confirmatory dedicated combination trials would strengthen the case further.
References
Marx N, Deanfield JE, Mann JFE, et al. Oral Semaglutide and Cardiovascular Outcomes in People With Type 2 Diabetes, According to SGLT2i Use: Prespecified Analyses of the SOUL Randomized Trial. Circulation. 2025;151(23):1639-1650. doi:10.1161/CIRCULATIONAHA.125.074545
