The short version

Plain English first. Everything below this section is the same material with the sources, the numbers and the caveats attached.

What tirzepatide is

A 39-residue synthetic peptide built on the GIP backbone that activates two receptors at once. It is an approved prescription medicine in Canada, the United States and the European Union. We list a 30 mg research vial. This entry exists so the numbers are checkable before anyone buys one — including the numbers further down that argue against it.

The structure, in the regulator's own words

From the Canadian Product Monograph: “Tirzepatide is based on the GIP sequence and contains aminoisobutyric acid (Aib) in positions 2 and 13, a C-terminal amide, and Lys residue at position 20 is attached to 1,20-eicosanedioic acid via a linker.” FDA's pharmacology review spells the linker out: a γ-Glu followed by two 8-amino-3,6-dioxaoctanoic acids, ending in a C20 diacid.

What “dual agonist” does and does not license you to say

The mechanism sentence FDA has actually approved is one line: “Tirzepatide is a GIP receptor and GLP-1 receptor agonist.” That is the entire claim. Unpacking it without inventing anything:

• GIP and GLP-1 are the two incretin hormones released from the gut after a meal. Tirzepatide is a single molecule that binds both receptors — not a co-formulation of two drugs. Semaglutide is built on the GLP-1 backbone and hits GLP-1 only.
• What the evidence supports: more HbA1c reduction and more weight reduction than a GLP-1-only comparator, at the doses tested. That is measured, twice, head to head.
• What it does not support: the popular line that “the GIP arm is what adds the extra weight loss.” Receptor activity is not balanced 1:1 between the two, and no approved label attributes any fraction of the clinical effect to the GIP component. That is mechanism-talk standing in for a result nobody has produced.

What the salt costs, and it is not the usual peptide story

None of the approved products is a salt, and none is a powder. Mounjaro and Zepbound are ready-to-use aqueous solutions at pH 6.5–7.5. What circulates as “research tirzepatide” is a lyophilised powder, and lyophilised synthetic peptides are almost always isolated as acetate or trifluoroacetate salts — the counterions left behind by the purification step. House arithmetic on the label's 4813.53 Da free base:

A big molecule carries its counterions lightly — the worst case here is 6.6%, against 28% on a hexapeptide like GHRP-6. Then residual water gets added on top, and that is where it stops being negligible.

Which is why “purity ≥98% by HPLC” does not answer this question. Chromatographic purity tells you how much of the peptide present is the right peptide. It says nothing about counterion mass and nothing about water. Three different quantities are all legitimately called “10 mg” — gross weight, peptide content, and net peptide content — and only the third corresponds to a labelled dose of the approved medicine.

It is an approved medicine, and that changes the question

Most compounds in this library are unapproved everywhere. Tirzepatide is the opposite problem: it is approved, which means there is a real label, a real manufacturer of record, and a real answer to nearly every question — and none of that transfers to a vial bought online.

The Canadian indication, verbatim: “MOUNJARO (tirzepatide injection) is indicated for once-weekly administration as an adjunct to diet and exercise to improve glycemic control for the treatment of: Adult patients with type 2 diabetes mellitus.” Zepbound's Canadian monograph adds chronic weight management at BMI ≥ 30, or 27–30 with a weight-related comorbidity, and moderate-to-severe OSA in adults with obesity.

One honest flag on our own sourcing: an automated read of both Canadian monographs returned “Date of Initial Authorization” fields in 2026, which conflict with the Summary Basis of Decision. Those are almost certainly revision dates picked up from the header, and the table above uses the SBD and Drug Product Database dates instead. We are printing that rather than quietly resolving it, because a reader who opens the monograph will see the 2026 field and deserves to know why.

What the trials actually measured

Two large programmes, both real, both randomised, and both measuring things that stand in for outcomes rather than outcomes themselves. The numbers below come from FDA-approved labelling wherever the label carries them, because those have been through statistical review.

SURPASS — type 2 diabetes. Primary endpoint: HbA1c

Two qualifications on the semaglutide head-to-head that vendor pages drop. The comparator was semaglutide 1 mg — the highest approved Ozempic dose at the time, but not the 2.0 mg later approved. And 40 weeks is a short window on which to judge a therapy meant to run for years.

SURMOUNT — obesity. Primary endpoint: percent body weight

SURMOUNT-5 — the head-to-head everyone cites

751 adults without diabetes, 72 weeks, both arms titrated to maximum tolerated dose — tirzepatide 10 or 15 mg against semaglutide 2.4 mg, which is the fair comparison the SURPASS-2 critique asks for.

A genuine head-to-head at both drugs' maximum approved doses, which is rare and worth crediting. What it measures is still weight.

Surrogates and outcomes — read this before quoting any number above

Every primary endpoint in the section above is a surrogate. SURPASS measured HbA1c. SURMOUNT measured percent body weight. SURMOUNT-OSA measured an apnoea index. None of those is an event and none is a death. There are exactly two tirzepatide trials with hard endpoints, and they say different things.

What that licenses, and what it does not. In people with type 2 diabetes and established cardiovascular disease, over four years, tirzepatide was not worse than dulaglutide for cardiovascular death, heart attack or stroke. It was not shown to be better. And because the comparator was an active drug rather than placebo, “no difference” cannot be read as no benefit or as benefit — it is a comparison between two treatments, not against nothing.

All-cause mortality favoured tirzepatide, but the reported driver was a reduction in non-cardiovascular death — a secondary finding, on a secondary endpoint, in a trial whose primary superiority test failed. That is hypothesis-generating and nothing more.

SUMMIT — the one hard-endpoint win, and how narrow it is

731 patients with heart failure with preserved ejection fraction and obesity, placebo controlled, median follow-up about two years. Composite of cardiovascular death or worsening heart failure: 36 events versus 56. Hazard ratio 0.62, 95% CI 0.41–0.95, p = 0.026. Symptom scores, six-minute walk distance and hsCRP all improved.

What has not reported yet

SURMOUNT-MMO is the trial that would answer the question properly: about 15,374 adults aged 40 and over with BMI ≥ 27, placebo-controlled — which SURPASS-CVOT was not — with morbidity and mortality as the endpoint. Estimated completion October 2027.

Until it reports, there is no placebo-controlled outcome evidence for tirzepatide in obesity without diabetes. Anyone saying it prevents heart attacks in that population is describing a trial that has not finished.

What happens when you stop

There is a randomised answer to this, and it is the single most relevant trial to anyone thinking about a “cycle”.

SURMOUNT-4 ran a 36-week open-label lead-in on tirzepatide, then randomised 670 people to continue or switch to placebo for another 52 weeks.

What that establishes is that the effect is maintained by continuous exposure. It is not induced and then kept. Nothing in the tirzepatide record supports the idea of running a course and holding the result, and this is the one trial designed to test exactly that.

We could not open the primary paper, so the per-arm percent-weight-change figures from randomisation to week 88 are not printed here — only the regain figure, which is the one the summary carried.

Harms, from the label rather than from forums

This is the part an approved medicine gives you that a grey-market vial cannot: rates, denominators, and a regulator's own words about what is not known.

Canada's monograph adds a sentence the US label does not carry, and it is the important one: “It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors.” In other words the regulator is saying it does not know whether screening helps. Any reassurance built on “just get your calcitonin checked” is not supported by the label.

Rates, from the pooled placebo-controlled obesity trials

Two things worth reading off that table. Pancreatitis did not separate from placebo — 0.2% either way — which is why it is a warning rather than a labelled adverse reaction with an excess rate; acute pancreatitis was also an exclusion criterion, so the trials are not where you would find it. And gallbladder disease is confounded by the weight loss itself, which is an independent gallstone risk factor.

How many people stopped

The longer trial gives roughly double the figure. Worth remembering when someone quotes 6.7% from a 72-week study as though it were the lifetime number.

The regulator has already finished deciding

Tirzepatide is the fully worked example of a grey market being closed. Every other compound in this library is somewhere earlier in that arc. This one has run the whole way, on the record, with dates.

FDA's own reasoning, verbatim: “an indefinite or overly long period of enforcement discretion for continued compounding of drugs that may be essentially copies of an approved drug that is no longer in shortage would not be appropriate.” And on the bulks list, having evaluated the nominations, FDA “did not identify a clinical need for outsourcing facilities to compound semaglutide, tirzepatide, and liraglutide from bulk drug substances.”

We checked the other list directly: tirzepatide does not appear on FDA's 503A bulk substances list in any of the three categories. Not Category 1, not Category 2, not Category 3. That is a negative finding from a direct read of the FDA page, and it means something different from the Category 2 listings that GHRP-2 and GHRP-6 carry — those were assessed and judged risky; this one has an approved product, so the question of compounding it from bulk does not arise the same way.

What FDA says about “research use only”

Directly on the labelling convention this entire market runs on: “These products have been sold directly to consumers for human use with dosing instructions.” And on salt forms: “The agency does not have information on whether these salts have the same chemical and pharmacologic properties as the active ingredient in the approved drug.”

The warning letters put it more bluntly still, on marketing: “Your claims imply that your products are the same as an FDA-approved product when they are not.”

The two products, side by side

Canada, and sport

On 9 April 2026 two Canadian bodies published peptide warnings on the same day with different lists. Tirzepatide is on one of them. If you are in Alberta, it is the one that names it.

All three checked by opening the document and looking for the word, not assumed. The 9 April Health Canada advisory names fourteen peptides and includes retatrutide — a triple agonist approved nowhere — but not tirzepatide, and not semaglutide either. The plausible reading is that it targets never-approved research-market compounds, while tirzepatide is handled through the ordinary machinery of having a DIN. That is our reading, stated separately from the observation, which is the part we checked.

The general rule, in Health Canada's words: research-use-only labelling “does not make these products legal or exempt from regulatory requirements.”

Sport — and this one is different from everything else here

Tirzepatide is not on the WADA Prohibited List. It is on the 2026 Monitoring Program, in force from 1 January 2026, monitored both in and out of competition. WADA's own description: the programme “includes substances that which are not on the Prohibited List, but that WADA wishes to monitor in order to detect patterns of misuse in sport.”

Two things follow. Not prohibited is not the same as permitted — individual sports bodies can be stricter. And monitoring is a leading indicator: it is the position semaglutide occupied before tirzepatide joined it. Our sourcing here is weaker than we would like — WADA's own 2026 documents were blocked to us across three routes, so this rests on WADA's definition plus two secondary sources that each name tirzepatide. An athlete should check the current list directly rather than trust this page.

Reconstitution, and why the tidy-looking numbers are the bad ones

Start from the fact that frames everything else: there is no approved lyophilised tirzepatide anywhere in the world. Every approved presentation is a ready-to-use solution. So there is no approved reconstitution instruction, no approved diluent and no approved post-reconstitution stability data. Every protocol in circulation is somebody's invention.

The approved ladder is six rungs: 2.5, 5, 7.5, 10, 12.5 and 15 mg, once weekly, stepping no faster than every four weeks. House arithmetic below, on a U-100 syringe where 1 unit = 0.01 mL.

And at 10 mg/mL, the top two rungs exceed a 1 mL barrel: 12.5 and 15 mg need 125 and 150 units. Someone maintaining at 15 mg is either giving two injections a week or has quietly switched concentration partway up the ladder — and switching concentration without recalculating is precisely the dosing-error pattern FDA describes in its adverse-event reports.

Errors this arithmetic does not capture, listed rather than glossed: syringe dead space, which carries twice the drug at 20 mg/mL that it does at 10; meniscus reading, a fixed absolute error and therefore proportionally worse at small volumes; and deliberate overfill, which biases in the opposite direction from counterion mass without reliably cancelling it. We also did not verify syringe graduation intervals against a manufacturer specification — the unit arithmetic is exact, but the claim about what a barrel can resolve rests on the common case.

Storage — and the two clocks nobody separates

Tirzepatide is the clearest illustration we have found of something that matters across this whole library: sterility and stability are different questions with different answers, and the industry routinely reports one number for both.

Read the approved formulations side by side and the reason for the dating falls out:

House arithmetic, and it lands somewhere interesting. The multi-dose products carry 5.4 mg of benzyl alcohol per 0.6 mL, which is 9.0 mg/mL — 0.90% w/v. That is essentially the concentration in bacteriostatic water. But they also carry 0.18% phenol as a second preservative, which a vial reconstituted with plain bacteriostatic water does not, in a container-closure system nobody has qualified.

What the stability literature actually contains

We found no study of reconstituted lyophilised tirzepatide held at any temperature over weeks with periodic assay. What exists is forced-degradation work, which answers a different question — can an assay detect degradation products — not how long a vial lasts.

Read honestly, that says tirzepatide is not exotically fragile — losing about 12% under a full day of direct sunlight is fairly robust for a peptide. It says nothing about a punctured vial at therapeutic concentration over four weeks: different concentration, different matrix, different container, no sterility dimension at all, and analytical standard solutions rather than a formulated product. The same paper reports no apparent degradation over 24 hours at room temperature and one week refrigerated, and reports nothing beyond one week.

Questions we actually get

Is it better than semaglutide?

For weight, at maximum tolerated doses, over 72 weeks, in people without diabetes: yes, and by a clear margin — −20.2% against −13.7% in a proper head-to-head. For cardiovascular outcomes: nobody knows, because the two have never been compared on events.

Does it reduce heart attacks?

Not shown. Its cardiovascular outcomes trial was run against dulaglutide rather than placebo, met non-inferiority, and failed superiority at p = 0.09. The one hard-endpoint win is a 731-patient trial in obesity-phenotype heart failure. The placebo-controlled outcomes trial does not report until October 2027.

Can I run a course and keep the results?

The only randomised test of that question says no. In SURMOUNT-4, 82% of people switched to placebo regained more than a quarter of what they had lost within a year. The effect is maintained by continuous exposure.

Why is the powder version cheaper if it is the same molecule?

Because you cannot tell whether it is the same molecule, at the same amount, in a sterile container. Three different quantities are all called “10 mg”, and a purity figure answers none of them. FDA's own position on the salts is that it “does not have information on whether these salts have the same chemical and pharmacologic properties as the active ingredient in the approved drug.”

What is the safest reconstitution?

We are not going to publish a protocol for a drug we do not sell. What the arithmetic shows is that 10 mg/mL is the only practical concentration where every rung of the approved ladder lands on a whole syringe unit, and that the tidy-sounding 30 mg vial plus 2 mL is the worst of the common choices.

Does “research use only” make it legal?

No. Health Canada: that labelling “does not make these products legal or exempt from regulatory requirements.” Tirzepatide has a Canadian DIN, and selling it without one is illegal regardless of which advisory list names it.

Will it show on a drug test?

It is not prohibited by WADA. It is on the 2026 Monitoring Program, which means WADA is collecting data on how often it turns up. That is the position semaglutide held before tirzepatide joined it. Check the current list yourself rather than trusting this page.

Anything to tell a doctor before surgery?

Yes, and this is the one people miss. The label warns about pulmonary aspiration under general anaesthesia because the drug delays gastric emptying, and instructs patients to disclose it before any procedure. Somebody using an unlabelled vial has nothing to show and may not think to mention it.

References

1. Mounjaro (tirzepatide) US Prescribing Information, revision 01/2026 — mechanism, pharmacokinetics, SURPASS trial results in §14, adverse reaction rates, boxed warning. accessdata.fda.gov
2. Zepbound (tirzepatide) US Prescribing Information, revision 08/2026 — obesity and OSA indications, SURMOUNT results, the full Warnings and Precautions list, and the storage sections underpinning our two-clock reading.
3. Mounjaro and Zepbound Canadian Product Monographs, Eli Lilly Canada — the verbatim Canadian indications, and the calcitonin-monitoring sentence the US label does not carry.
4. SURMOUNT-5. Aronne LJ, et al. New England Journal of Medicine 2025;393(1):26–36. DOI 10.1056/NEJMoa2416394 — the head-to-head against semaglutide 2.4 mg. Read through a cardiology trial summary; the primary paper was not opened.
5. SURPASS-CVOT. Nicholls SJ, Pavo I, Bhatt DL, et al. New England Journal of Medicine 2025;393:2409–2420 — non-inferiority met, superiority not met. Read through two trial summaries, which disagree on enrolment (13,165 vs 13,299); we could not open the paper to adjudicate, so this page says “about 13,200”.
6. SUMMIT, heart failure with preserved ejection fraction. Published in NEJM (NEJMoa2410027) and JACC — the single hard-endpoint result. Read through the ACC summary.
7. SURMOUNT-4, the randomised withdrawal trial — the 82% regain figure. Read through the ACC summary; the primary paper was not opened, which is why no per-arm weight figures appear.
8. SURMOUNT-MMO, NCT05556512 — approximately 15,374 participants, placebo-controlled, estimated completion October 2027.
9. US Food and Drug Administration. Declaratory Order, 19 December 2024 — the tirzepatide shortage resolved, with the 60- and 90-day compounding wind-downs. fda.gov
10. US Food and Drug Administration. FDA proposes to exclude semaglutide, tirzepatide and liraglutide from the 503B bulks list, 30 April 2026 — “did not identify a clinical need”. fda.gov
11. US Food and Drug Administration. FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight Loss, page current 1 September 2026 — the salt-form statement, the research-use-only statement, and the adverse-event counts (990 compounded semaglutide, more than 730 compounded tirzepatide, as of 31 May 2026). fda.gov
12. US Food and Drug Administration. Bulk drug substances used in compounding under section 503A — read directly, checking all three categories. Tirzepatide appears in none of them.
13. Health Canada. Unauthorized injectable peptide drugs seized and sold by Canada Peptide, 1 August 2025 — tirzepatide named. And: Think twice before injecting peptides bought online, 9 April 2026 — tirzepatide not named, checked by two independent routes. College of Physicians & Surgeons of Alberta, 9 April 2026 — tirzepatide named.
14. Khalil HA, Hassanein NA, El-Yazbi AF, Mahgoub H. A multimodal HPLC stability indicating approach for the estimation of semaglutide and tirzepatide. BMC Chemistry 2026;20(1):31 — the forced-degradation figures, and the one-week refrigerated limit.
15. World Anti-Doping Agency, Monitoring Program — WADA's own definition. The 2026 programme document itself was blocked to us; see the note below.

What we could not check, listed rather than hidden. WADA's own 2026 Monitoring Program document and the 2026 Prohibited List explanatory note were both robots-blocked across three routes, so the sport section rests on WADA's definition plus two secondary sources that each name tirzepatide — an athlete should verify directly. We could not obtain a UNII code from any regulator source and have left it blank rather than guessed; PubChem and DrugBank were unreachable across four attempts between them. No regulator source prints the 39-residue sequence string — the labels describe the modifications without giving the residues — and vendor sequence strings are not a source we will print. The CAS number was found only on database aggregators. We did not open the NEJM or JAMA primary papers for any trial, using FDA-approved labelling where it carried the numbers and trial summaries where it did not, which is why the SURPASS-CVOT enrolment discrepancy is unresolved and the SURMOUNT-4 per-arm figures are absent. Health Canada's recalls search endpoint was blocked, so our list of Canadian advisories is the ones we located by title and cannot claim to be exhaustive. And the absence of a reconstituted-tirzepatide stability study is a negative from open-web searching, not a systematic review — weak evidence of absence, and we would rather say so.