Tirzepatide Side Effects: What Its FDA Label and Pivotal Trials Actually Report

Tirzepatide is the single highest-traffic compound on this site — our tirzepatide provider rankings is the board everything else here points toward — and until now, its side effects have only ever come up inside a legality article and a head-to-head comparison against semaglutide, never as their own focused review. That's a real gap, and an unusual one to have left open, because tirzepatide isn't like this site's other deep-dive subject, retatrutide: it isn't investigational. It's FDA-approved, under two brand names — Zepbound for weight reduction and Mounjaro for type 2 diabetes — which means there's an actual official answer to "what are tirzepatide's side effects," sitting in a public, current, FDA-approved label. Most of what ranks for that question still doesn't quote the label directly. This article does: every figure below was read straight off Zepbound's and Mounjaro's own current prescribing information, fetched live from DailyMed, and the peer-reviewed pivotal trials behind them — not a compounding pharmacy's marketing page, not a forum thread, and not a paraphrase of either.

The boxed warning FDA requires on both labels

Zepbound and Mounjaro carry the identical boxed warning, because they're the same active ingredient reviewed under the same nonclinical finding. Zepbound's own label states it directly: "In rats, tirzepatide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors... at clinically relevant plasma exposures... It is unknown whether ZEPBOUND causes thyroid C-cell tumors, including [medullary thyroid carcinoma, MTC], in humans as human relevance of tirzepatide-induced rodent thyroid C-cell tumors has not been determined¹." Mounjaro's own label reports the identical finding — in both male and female rats this time — and the identical unresolved question about human relevance². Both labels go further than a caution: both drugs are flatly contraindicated in anyone with a personal or family history of MTC or Multiple Endocrine Neoplasia syndrome type 2, stated in FDA's own boxed language, not a soft warning buried further down the page.

Zepbound's own Table 1: gastrointestinal events, by exact percentage

Zepbound's label reports adverse-reaction data from a pool of two placebo-controlled trials — Study 1 (SURMOUNT-1, NCT04184622, 2,539 adults with obesity or overweight and no diabetes) and Study 2 (SURMOUNT-2, NCT04657003, 938 adults with overweight or obesity and type 2 diabetes, which tested only the 10 mg and 15 mg doses, not 5 mg — the reason the pooled table's 5 mg column has a noticeably smaller patient count than its 10 mg and 15 mg columns). Table 1 of the label reports these adverse reactions occurring in at least 2% of Zepbound-treated patients, more often than on placebo¹:

One detail in the label's own text is worth flagging specifically rather than skipped: measured as an overall category rather than by individual symptom, gastrointestinal adverse reactions occurred in 56% of patients at every one of the three doses tested — 5 mg, 10 mg, and 15 mg alike — against 30% on placebo¹. That's a real, reported finding: the type and severity of GI events shift with dose (more vomiting, less constipation, as the table above shows), but the label's own pooled number for "any GI adverse reaction" doesn't rise with dose the way nausea or vomiting individually do. Discontinuation due to gastrointestinal adverse reactions specifically did rise with dose — 1.9% at 5 mg, 3.3% at 10 mg, 4.3% at 15 mg, against 0.5% on placebo — and permanent discontinuation for any adverse reaction ran 4.8%, 6.3%, and 6.7% at those same three doses, against 3.4% on placebo¹.

Beyond GI: what the label documents specifically

Severe gastrointestinal adverse reactions — not just any GI event, but ones the trials classified as severe — were reported in 1.7% (5 mg), 2.5% (10 mg), and 3.1% (15 mg) of Zepbound-treated patients, against 1% on placebo; the label states plainly that Zepbound "is not recommended in patients with severe gastroparesis¹." Acute gallbladder disease is a separate, named warning: cholelithiasis (gallstones) was reported in 1.1% of Zepbound-treated patients against 1% on placebo, cholecystitis in 0.7% against 0.2%, and cholecystectomy (gallbladder removal) in 0.2% against none on placebo¹. Acute pancreatitis is where the label's own numbers are worth reading closely rather than assumed: in this specific pool of weight-reduction trials, confirmed acute pancreatitis occurred in 0.2% of Zepbound-treated patients — statistically indistinguishable from the 0.2% reported on placebo (0.14 vs. 0.15 patients per 100 years of exposure)¹. That's the label's own reported finding for these particular trials specifically; it isn't the same as saying tirzepatide carries no pancreatitis risk at all — postmarketing reports separately include hemorrhagic and necrotizing pancreatitis, "sometimes resulting in death," alongside ileus, intestinal obstruction, and severe constipation including fecal impaction, none of which were common enough in the controlled trials themselves to generate a rate, but all of which FDA requires listed because they were reported after approval¹. Immediate hypersensitivity reactions occurred in 2.1% of Zepbound-treated patients against 0.4% on placebo, and acute kidney injury tied to volume depletion — dehydration driven by GI-related fluid loss — is called out as its own numbered warning, with the label specifically instructing renal-function monitoring during dose initiation and escalation¹.

Mounjaro's own numbers: a different population, a different pattern

Mounjaro's label reports its own separate Table 1, built from a different pool — SURPASS-1 and SURPASS-5, two placebo-controlled trials in 718 adults with type 2 diabetes, at the same 5 mg/10 mg/15 mg doses². The individual-symptom rates run lower across the board than Zepbound's obesity-population numbers: nausea at 12%, 15%, and 18% (5/10/15 mg) against 4% on placebo; diarrhea at 12%, 13%, and 17% against 9%; vomiting at 5%, 5%, and 9% against 2%; constipation at 6%, 6%, and 7% against 1%². Unlike Zepbound's flat 56%-at-every-dose overall GI rate, Mounjaro's pooled GI category does climb with dose — 37.1%, 39.6%, and 43.6% against 20.4% on placebo — and discontinuation due to GI adverse reactions follows the same upward pattern: 3.0%, 5.4%, and 6.6% against 0.4%². It's the same molecule and the same class of side effect in both populations, but the label itself reports genuinely different numbers for each — a reminder that "tirzepatide's side effects" isn't a single fixed table so much as two related ones, each specific to the population and trials that generated it.

What a real head-to-head trial against semaglutide adds

Our companion article on tirzepatide versus semaglutide covers the efficacy side of SURPASS-2 in full; that same trial's own results text also reports adverse-event data neither label above captures, because SURPASS-2 compared tirzepatide directly against an active comparator rather than placebo. Across the trial's three tirzepatide dose arms and its one semaglutide arm, nausea ran 17%-22% (tirzepatide) versus 18% (semaglutide), diarrhea 13%-16% versus 12%, and vomiting 6%-10% versus 8% — genuinely similar GI rates between the two drugs in this specific trial, not the wide gap the individual placebo-controlled tables above might suggest on their own⁴. One number is worth reading carefully rather than skipped: serious adverse events were reported in 5%-7% of patients on tirzepatide against 3% on semaglutide in this trial⁴ — SURPASS-2's own published results don't break down what those serious events were, so this article reports the rate exactly as published rather than speculating about its cause.

What SURMOUNT-1's own peer-reviewed publication adds

Zepbound's label figures above are pooled across two trials; SURMOUNT-1's own NEJM publication reports numbers from that single trial alone, and they aren't identical. The paper's own results state that adverse events caused treatment discontinuation in 4.3%, 7.1%, and 6.2% of participants on 5 mg, 10 mg, and 15 mg tirzepatide, respectively, against 2.6% on placebo³ — a different shape than the label's pooled 4.8%/6.3%/6.7%/3.4%, because the label's Study 2 (SURMOUNT-2) contributes its own type 2 diabetes population's data into that pooled figure and SURMOUNT-1 alone does not. Neither number is wrong; they're answering slightly different questions — one trial's own reported rate, versus FDA's pooled two-trial rate — and this article treats them as exactly that rather than picking whichever sounds better. SURMOUNT-1's own plain-language summary of its safety data: "The most common adverse events with tirzepatide were gastrointestinal, and most were mild to moderate in severity, occurring primarily during dose escalation³" — the same pattern the label's own numbers above quantify in more detail.

This data describes the FDA-approved brand — not any specific compounded vial

Every number in this article was generated on Zepbound or Mounjaro specifically: the FDA-approved, brand-name finished product, at FDA's own approved dose-escalation schedule, tested and labeled as that exact formulation. Our explainer on what's still legal to compound covers the narrow conditions under which a compounding pharmacy can still legally prepare tirzepatide today — but a compounded preparation is a different finished product from Zepbound or Mounjaro even when it contains the same active molecule, and it isn't independently reviewed by FDA for identity, purity, or potency the way the branded product is. That matters specifically for the adverse-event numbers above, which describe the labeled products' own escalation schedule — not necessarily how quickly, or how consistently, an individual compounded seller titrates a patient's dose. Some providers publish their own schedule directly: RxPepsDirect's own review documents a real, dose-scaling titration schedule for its compounded tirzepatide, climbing from a 12 mg starting dose to 66 mg by month six — numbers that don't map onto Zepbound's or Mounjaro's own 2.5 mg-to-15 mg FDA-approved ladder at all, since compounded formulations aren't bound to the same milligram scale the approved product uses. Our review of Precision Telemed, by contrast, prices its lowest, 2 mg tirzepatide tier separately from every higher maintenance dose — a starting-dose-versus-maintenance-dose price structure that at least implies a titration step, without publishing the actual schedule the way RxPepsDirect does. Neither example is a safety claim about either company; it's a reminder that the label's own escalation-driven GI numbers above assume the specific schedule Zepbound and Mounjaro were tested on, and a compounded provider's own schedule is a separate thing to check for yourself.

One more label detail worth knowing if you're of reproductive age

Both labels also flag something unrelated to the GI/gallbladder/pancreatitis findings above but genuinely worth knowing before starting: tirzepatide delays gastric emptying, which can reduce how completely an oral medication — including oral contraceptives — gets absorbed, especially during dose initiation and each escalation step. Our full look at what the label says about birth control, plus the separate, real fertility-restoring signal in PCOS trial data, covers both in detail; it isn't a side effect in the same sense as nausea or gallstones, but it's a real, labeled interaction a reader deciding between providers should know applies regardless of which one they pick.

The bottom line

Tirzepatide's own FDA-approved labels — for Zepbound and for Mounjaro separately — report gastrointestinal adverse events as the dominant, dose-related signal in both populations, a boxed warning about thyroid C-cell tumors carried over from rodent carcinogenicity data whose human relevance remains undetermined, a real but modest gallbladder-disease signal, and a pancreatitis finding that, specifically in the weight-reduction trial pool, didn't actually exceed placebo. None of that is a paraphrase — it's what the label itself states, verified against a live fetch of both current prescribing information documents. What it describes, precisely, is the FDA-approved brand product on its own tested escalation schedule; what it doesn't automatically describe is any individual compounded provider's own vial. Once you have an actual prescribed dose in hand, our reconstitution calculator turns that into the exact syringe reading — pure arithmetic on the number you enter, not a substitute for reading the label, or your prescriber's own guidance, first. And for how the providers on our tirzepatide rankings each disclose their own pharmacy sourcing and dosing structure, that comparison — not this article — is the place to look.