The biggest unanswered question in obesity pharmacology just got a real answer: in head-to-head data, tirzepatide and semaglutide reduce cardiovascular events at similar rates, even though tirzepatide produces meaningfully more weight loss. We break down what the new Nature Medicine real-world analysis and the SURMOUNT-5 post-hoc actually show, the dosing framework for choosing between the two, why adding GIP receptor agonism may not produce additive cardiovascular benefit, and the public comment deadline for the PCAC peptide review that is six weeks away.
The standard framing of tirzepatide versus semaglutide has been settled for two years: tirzepatide produces more weight loss. SURMOUNT-5, the first Phase 3 head-to-head, made this clear. What has not been settled, until now, is whether that weight-loss advantage translates into a cardiovascular outcomes advantage. Two new analyses, both published in late 2025, suggest it does not.
The first is a real-world cohort study published in Nature Medicine in November 2025 by Krueger and colleagues at Brigham and Women's Hospital and Harvard Medical School. The team used US insurance claims data from 2018 to 2025 to emulate two cardiovascular outcome trials: SUSTAIN-6 (semaglutide versus sitagliptin) and SURPASS-CVOT (tirzepatide versus dulaglutide). They then conducted a direct head-to-head comparison of semaglutide and tirzepatide in patients with elevated cardiovascular risk.
The result: in the direct head-to-head comparison, the hazard ratio for tirzepatide versus semaglutide was 1.06 (95% confidence interval 0.95 to 1.18). The confidence interval crosses 1.0, meaning the difference is not statistically significant. But the point estimate trends slightly in semaglutide's favor, not tirzepatide's. The cleanest reading: tirzepatide and semaglutide produce similar cardiovascular event reductions in real-world use. Tirzepatide is not superior. If anything, the data hints at the opposite.
The second is a post-hoc analysis of SURMOUNT-5, the Phase 3 trial that directly compared the two compounds at maximum tolerated doses in adults with obesity. Mamas and colleagues, publishing in European Heart Journal Open in September 2025, modeled 10-year cardiovascular disease risk based on the changes in cardiometabolic risk factors observed in SURMOUNT-5. The model projected approximately 70 fewer new cases of type 2 diabetes and 10 fewer cardiovascular events per 1,000 patients on tirzepatide compared with semaglutide at max tolerated doses, over a 10-year horizon. Read that carefully: these are modeled projections from a risk equation, not observed cardiovascular outcomes from a trial. They depend entirely on the assumption that improvements in surrogate markers translate proportionally to hard endpoints, which they often do not in real life.
Two things to flag prominently. First, the Mamas et al. analysis has six Eli Lilly co-authors. Lilly makes tirzepatide. This is industry-funded modeling work that favors the sponsor's compound. It is not invalid, but it requires the disclosure most coverage leaves out. The Krueger et al. real-world study, by contrast, is academic, conducted by Brigham and Women's pharmacoepidemiology group with no listed pharmaceutical funding. When those two evidence sources point in different directions, the academic real-world data should carry more weight than the industry-funded model.
Second, semaglutide has the longer cardiovascular outcomes track record. SUSTAIN-6 (Marso et al., 2016, NEJM) showed a 26 percent relative reduction in major adverse cardiovascular events in type 2 diabetes, with primary endpoint events in 6.6 percent of semaglutide patients versus 8.9 percent on placebo, hazard ratio 0.74 (95% CI 0.58 to 0.95). SELECT (Lincoff et al., 2023, NEJM) showed a 20 percent reduction in nondiabetic adults with established cardiovascular disease and obesity, with a primary endpoint event in 6.5 percent of the semaglutide group versus 8.0 percent of placebo, hazard ratio 0.80 (95% CI 0.72 to 0.90), n=17,604. Tirzepatide's SURPASS-CVOT (2025) showed noninferiority to dulaglutide, which itself has established cardiovascular benefit. So tirzepatide's cardiovascular benefit is real but inherited through a noninferiority margin, while semaglutide's is established through direct placebo comparison.
The bottom line: tirzepatide produces more weight loss; in cardiovascular outcomes the two are roughly tied, with semaglutide having the longer evidence base. Anyone telling you tirzepatide is unambiguously better for cardiovascular protection is either selling something or has not read the head-to-head data.
If you are working with patients, athletes, or yourself on a GLP-1 protocol decision, the choice between tirzepatide and semaglutide is now less about superiority and more about matching the compound to the goal. Here is the framework we use.
Start with the primary endpoint.
If the goal is maximum weight loss in someone without established cardiovascular disease, tirzepatide has the edge. SURMOUNT-1 produced approximately 22.5 percent weight loss at 72 weeks at the 15 mg dose. SURMOUNT-5 confirmed the advantage over semaglutide at max tolerated doses. If the goal is cardiovascular event prevention in someone with established disease, semaglutide has the longer evidence base (SELECT, SUSTAIN-6, PIONEER-6). If the goal is glycemic control in type 2 diabetes, tirzepatide produced larger A1C reductions in SURPASS-2 against semaglutide 1 mg, though semaglutide has stronger data at the 2 mg dose.
Then look at dosing cadence.
Semaglutide is weekly subcutaneous, escalating from 0.25 mg to 2.4 mg for obesity (Wegovy) or 0.25 mg to 1 mg for diabetes (Ozempic, off-label to 2 mg). Tirzepatide is weekly subcutaneous, escalating from 2.5 mg to 15 mg for obesity (Zepbound) or 5 mg to 15 mg for diabetes (Mounjaro). Both require a slow titration to manage gastrointestinal side effects. Tirzepatide has more dose levels (2.5, 5, 7.5, 10, 12.5, 15 mg), which gives you more room to find the highest tolerated dose without dropping back to a sub-therapeutic level.
Dose tolerability is real, especially at max.
SURMOUNT-5 reported similar discontinuation rates between the two compounds, but the absolute rate at max tirzepatide (15 mg) was numerically higher. In TRIUMPH-1 for retatrutide, discontinuation rates were 4.1 percent at 4 mg, 6.9 percent at 9 mg, and 11.3 percent at 12 mg versus 4.9 percent on placebo. This is a class pattern: gastrointestinal adverse events scale with dose. For real-world protocols, the optimal endpoint may not be the maximum approved dose, but the highest dose the individual tolerates with sustained adherence over twelve months. A patient on 10 mg tirzepatide for two years will likely do better than one on 15 mg for six months who stops.
A note on supply and compounding.
The legal landscape for compounded GLP-1s has shifted decisively against the cash-pay compounded market. The FDA resolved the tirzepatide shortage in December 2024, with enforcement deadlines of February 19, 2025 for state-licensed (503A) compounders and March 19, 2025 for outsourcing facilities (503B). The semaglutide shortage was resolved in February 2025, with parallel deadlines of April 22 and May 22, 2025. On April 30, 2026, the FDA went further: it proposed excluding semaglutide, tirzepatide, and liraglutide from the 503B Bulks List entirely, removing the last bulk-compounding pathway for these drugs. The public comment period on that proposal is open through June 29, 2026. The practical effect for patients: large-scale compounded copies at the 150 to 300 dollar monthly price point are no longer legally available. Anyone still being prescribed compounded semaglutide or tirzepatide in 2026 should understand the legal exposure their prescriber and pharmacy are accepting.
Practical decision rule.
For most adults with obesity and no established cardiovascular disease: start tirzepatide and titrate to highest tolerated dose. For adults with established cardiovascular disease or strong family history: semaglutide has the better evidence base. For type 2 diabetes with hemoglobin A1C above 9 percent: tirzepatide for the larger glycemic reduction. None of these are absolute, and any of these patients could be reasonably treated with the other compound. The choice is closer than the marketing suggests.
The interesting question raised by the head-to-head data is mechanistic: why does adding GIP receptor agonism to GLP-1 produce so much more weight loss, but not a corresponding boost in cardiovascular benefit? Understanding this matters, because retatrutide adds glucagon to GIP and GLP-1, and we now have to predict whether its weight-loss advantage will or will not extend to cardiovascular endpoints.
GIP, glucose-dependent insulinotropic polypeptide, is one of the two main incretin hormones (the other is GLP-1). It is released from K cells in the proximal small intestine in response to nutrient ingestion. Its primary recognized function is amplifying glucose-stimulated insulin secretion from pancreatic beta cells. For decades, GIP was considered the less interesting incretin: it has weaker effects on satiety than GLP-1 and its action appeared to be impaired in type 2 diabetes.
What tirzepatide demonstrated, and what was genuinely surprising to the field, is that pharmacologic GIP receptor agonism in the obesity context produces effects that endogenous GIP does not. The GIP receptor is expressed not just on pancreatic beta cells but in adipose tissue, the central nervous system (including the hypothalamus), and the cardiovascular system. Pharmacologic activation of adipose GIP receptors appears to enhance lipid storage capacity in subcutaneous depots and reduce ectopic fat deposition, a beneficial pattern. Central GIP receptor activation appears to suppress appetite synergistically with GLP-1. These are the mechanisms that drive tirzepatide's weight loss advantage.
The cardiovascular question is different. GLP-1 receptor activation produces several effects that plausibly drive its cardiovascular benefit: improved endothelial function, reduced inflammation in atherosclerotic plaques, modest blood pressure reduction, and direct effects on myocardium under ischemic stress. The mechanisms driving SELECT and SUSTAIN-6's outcome reductions are well-characterized in mechanistic studies. GIP receptor activation, in contrast, has a much less established cardiovascular profile. Some preclinical work suggests GIP signaling may even be cardioprotective in its own right. Some suggests it may oppose certain GLP-1 cardiovascular effects. The honest answer is that the cardiovascular pharmacology of GIP receptor agonism is not yet settled, and the human data from SURPASS-CVOT and the real-world comparison suggests that whatever GIP is contributing on the cardiovascular side, it is not producing a clinically detectable boost above what GLP-1 already provides.
For retatrutide, this means the prediction is uncertain. Adding glucagon receptor agonism (the third agonism in retatrutide) introduces effects on hepatic glucose output, lean mass preservation, and energy expenditure that may have their own cardiovascular profile, positive or negative. TRIUMPH-3, the ongoing cardiovascular outcomes trial in adults with established cardiovascular disease, will provide the answer. Until then, treating retatrutide as obviously cardioprotective just because its weight loss is bigger is the same mistake that early tirzepatide enthusiasts made.
If you have not been tracking the FDA peptide regulatory timeline, here is what you need to know in five paragraphs.
On April 15, 2026, the FDA announced removal of twelve peptides from Category 2 of the 503A bulks list following nomination withdrawals. The peptides removed include BPC-157, KPV, TB-500, MOTS-c, DSIP (Emideltide), Semax, Epithalon, GHK-Cu, Melanotan II, LL-37, Dihexa acetate, and PEG-MGF. Removal from Category 2 is not the same as addition to Category 1. The substances are now in regulatory limbo, neither prohibited from compounding nor explicitly authorized.
On April 16, 2026, the FDA published a Federal Register notice scheduling a Pharmacy Compounding Advisory Committee meeting for July 23 and 24, 2026, at FDA's White Oak Campus in Silver Spring, Maryland. The PCAC will consider seven peptides for potential inclusion on the 503A Bulk Drug Substances List: BPC-157, KPV, TB-500, MOTS-c, DSIP (Emideltide), Semax, and Epithalon. A separate PCAC meeting is planned for before the end of February 2027 to consider five additional peptides: GHK-Cu, Melanotan II, LL-37, Dihexa acetate, and one other.
What PCAC actually does: the committee reviews scientific evidence on each substance (physical and chemical characterization, safety in compounding, evidence of effectiveness, and historical compounding use) and votes on a non-binding recommendation. The FDA then decides whether to initiate notice-and-comment rulemaking to add the substance to Category 1. Under typical FDA timelines, rulemaking adds another twelve to twenty-four months after a positive PCAC recommendation. A positive vote at the July meeting is the start of the process, not the end.
The actionable item: the public comment period is open under Docket FDA-2025-N-6895 on regulations.gov. Comments submitted by July 9, 2026 will be provided to the committee in advance of the meeting. Comments submitted by July 22, 2026 will still be considered by FDA but may not reach the committee. The docket closes after July 22. If you have research experience with any of the seven peptides under review, comments submitted with citations to published evidence carry meaningfully more weight than general statements of support. Anyone planning to make an oral presentation at the meeting must notify the FDA by June 30, 2026.
The framing matters here. Most peptide media is treating the April 15 Category 2 removal as if it were FDA approval. It is not. It is a procedural step that opens the door to a multi-year regulatory pathway. The July 23 to 24 PCAC meeting is the next checkpoint in that pathway, and the public comment window is open right now.
That is next week. We will see you Tuesday.
The Peptide Signal Team
For research and educational purposes only · Not medical advice · Consult a qualified physician before any human use