Tirzepatide vs. Semaglutide: What the Head-to-Head Trials Actually Show

Ask which is "better" — tirzepatide or semaglutide — and most of what ranks for that question is either a vendor's own comparison chart or a blog stitching together two unrelated trials as if they were run side by side. That second approach is a real methodological problem: comparing one trial's tirzepatide arm against a completely different trial's semaglutide arm, run in different years on different populations, is not the same evidence as randomizing the same group of people to one drug or the other. The good news for anyone actually trying to answer this is that nobody has to settle for the indirect version — a real, randomized, head-to-head trial exists, and a second one exists for a different population. This article traces every number back to those two trials directly, then separately covers what the indirect comparison shows and where it's weaker evidence, before getting to the question this site's readers actually have: does any of this tell you something useful about a compounded provider's tirzepatide or semaglutide, specifically?

The real head-to-head trial: SURMOUNT-5

SURMOUNT-5 is a phase 3b, open-label, randomized, controlled trial that did exactly what the indirect comparisons can't: it put 751 adults with obesity — but without type 2 diabetes — on either tirzepatide or semaglutide, in the same trial, at the same time, and followed them for 72 weeks. Neither drug was capped at a token dose to make the comparison closer; each participant was titrated to their own maximum tolerated dose — 10 mg or 15 mg for tirzepatide, 1.7 mg or 2.4 mg for semaglutide, matching each drug's own highest approved obesity dose. The primary endpoint was straightforward: percent change in body weight from baseline to week 72¹.

The result was not close. Least-squares mean weight change was -20.2% with tirzepatide against -13.7% with semaglutide (P<0.001) — roughly a 6.5-percentage-point gap on top of a shared baseline. Waist circumference told the same story: -18.4 cm with tirzepatide versus -13.0 cm with semaglutide, also P<0.001. Participants on tirzepatide were also more likely than those on semaglutide to hit every weight-loss threshold the trial measured — 10%, 15%, 20%, and 25% of body weight lost. The trial's own conclusion states it plainly: "treatment with tirzepatide was superior to treatment with semaglutide with respect to reduction in body weight and waist circumference at week 72¹." Gastrointestinal side effects — nausea, diarrhea, vomiting, constipation — were the most common adverse events in both arms, mostly mild to moderate, and concentrated during dose escalation, consistent with both drugs' known safety profile.

A second head-to-head trial exists too — for a different population and a different question

SURMOUNT-5 answers the weight-loss question in people without diabetes. A separate, earlier head-to-head trial — SURPASS-2 — answers a related but distinct question in people who do have type 2 diabetes: glycemic control. SURPASS-2 randomized 1,879 adults with type 2 diabetes to tirzepatide (5 mg, 10 mg, or 15 mg) or semaglutide, and measured the change in HbA1c over 40 weeks. Every tirzepatide dose beat semaglutide on the primary endpoint: HbA1c fell by 2.24 and 2.30 percentage points on the 10 mg and 15 mg tirzepatide doses, against 1.86 points on semaglutide (P<0.001 for both). Weight loss was also greater with tirzepatide — treatment differences of -3.6 kg and -5.5 kg at those same two doses².

One detail matters enough to flag rather than gloss over: SURPASS-2 capped its semaglutide arm at 1 mg — the dose semaglutide was approved at for diabetes when this trial was designed and run. Semaglutide's diabetes label has since added a 2 mg dose, and its obesity-approved dose (in Wegovy) goes to 2.4 mg — both higher than what SURPASS-2 tested. That doesn't erase the trial's finding, but it does mean SURPASS-2 is head-to-head evidence about tirzepatide versus a specific, now-superseded semaglutide ceiling, not necessarily about tirzepatide versus semaglutide at every dose semaglutide is prescribed at today. SURMOUNT-5, run more recently and specifically titrating each drug to its own current maximum, doesn't have that same limitation for the weight-loss question.

What the indirect comparison shows — and why it's weaker evidence, clearly labeled as such

Before SURMOUNT-5 read out, the closest thing available was comparing each drug's own placebo-controlled monotherapy trial. Those numbers are worth including, but only with the caveat stated up front: this is a cross-trial comparison, not a randomized one. SURMOUNT-1 tested tirzepatide against placebo in 2,539 adults with obesity, reporting mean weight loss of -19.5% (10 mg) and -20.9% (15 mg) at 72 weeks, against -3.1% for placebo³. STEP 1 tested semaglutide 2.4 mg against placebo in a separate trial of adults with overweight or obesity, reporting -14.9% at 68 weeks against -2.4% for placebo⁴. Lined up next to each other, those numbers point the same direction as SURMOUNT-5's actual head-to-head result — a roughly 5-to-7-point gap favoring tirzepatide — but different trials enrolled different participants in different years with slightly different follow-up windows (72 weeks vs. 68), and neither trial was designed to be compared against the other. Where SURMOUNT-5 exists, it's the better evidence; the placebo trials are included here because they're what the field relied on before SURMOUNT-5 was run, and because their broad agreement with SURMOUNT-5's actual result is itself informative.

A 2026 systematic review and meta-analysis pooling 12 studies — a mix of head-to-head-adjacent randomized and observational data — reached a similar pooled estimate: a mean difference of -4.61 percentage points in body weight favoring tirzepatide (95% CI, -6.03 to -3.20), and higher odds of hitting every weight-loss threshold checked. The same paper is explicit that this pooled estimate carries substantial heterogeneity (I²=87-97% across outcomes) because it mixes trial designs and populations — its own authors describe the result as "numerically greater weight loss," not proof that any individual patient will see that exact gap⁵.

What none of this tells you about a compounded provider's vial

Here's the part that actually matters for a reader comparing telehealth providers rather than reading NEJM for its own sake: every number above was generated on the FDA-approved, brand-name product, at a dose and titration schedule the manufacturer specified, administered under trial-protocol supervision. Tirzepatide's approved brands are Mounjaro and Zepbound; semaglutide's are Ozempic and Wegovy. A direct check against FDA's own Drugs@FDA database confirms both are approved, prescription-status drug products, not investigational compounds⁶⁷.

That's precisely what a compounding pharmacy's product is not being compared against in these trials. Our explainer on what's actually still legal to compound covers the narrow exceptions that let some pharmacies prepare tirzepatide today — but a compounded preparation is, by definition, not the identical branded product SURMOUNT-5 and SURPASS-2 tested. It may use a different concentration, a different vial-and-syringe protocol instead of a pre-filled pen, and it isn't reviewed by FDA for identity, purity, or potency the way an approved drug product is. None of that means a given compounded product doesn't work — the active molecule, when it's genuinely present at the labeled concentration, is the same molecule the trials tested. It means the specific efficacy numbers above are evidence about tirzepatide and semaglutide as molecules and about the branded products specifically, not a verified guarantee about what's actually in any one provider's vial. A trial can't audit a pharmacy it was never testing.

The bottom line

If the question is simply "does the clinical evidence show a real difference between these two molecules" — yes. Two independent, randomized, head-to-head trials, run in two different populations, both found tirzepatide produced significantly greater weight loss (SURMOUNT-5) and better glycemic control (SURPASS-2) than semaglutide, at doses reflecting each drug's own approved maximum. That is genuine evidence, not a coin flip dressed up as a debate. If the question is "will switching from a semaglutide provider to a tirzepatide provider get me the same 6-point gap SURMOUNT-5 reported" — the honest answer is that depends on factors the trial didn't test: whether the compounded product is accurately dosed, whether you tolerate the higher-dose titration tirzepatide requires, and whether you're being compared against the trial's specific population at all. For what to actually look for in a provider once you've settled on a compound, our tirzepatide provider rankings grade who states their dosing and sourcing clearly and who doesn't; and once you have an actual prescribed dose in hand, our reconstitution calculator turns that into the exact syringe reading — pure arithmetic on what you type in, not a substitute for verifying what's actually in the vial.