The reverse of the more common switch has even less behind it. Moving from semaglutide to tirzepatide at least has Lilly's switching trials and a published series; moving from tirzepatide to semaglutide has no trial in the registry, one case report and a model that converts average weight loss, not doses for a person. This page sets out what each source states (read 2026-10-03), the washout arithmetic the tirzepatide label allows, and the semaglutide steps in syringe units. It is not a conversion and gives no instruction.
The opposite direction is on switching from semaglutide to tirzepatide, which carries the SURPASS-SWITCH trials and the semaglutide washout; this page adds what is specific to leaving tirzepatide. The doses themselves are on the tirzepatide and semaglutide dosage pages.
What the labels say: a starting dose, and nothing about switching
| Question | ZEPBOUND (tirzepatide) | WEGOVY (semaglutide) |
|---|---|---|
| Starting dosage | 2.5 mg weekly | 0.25 mg weekly |
| Switching section | None | Injection ↔ tablet only |
| Exception for prior GLP-1 use | None | None |
| Half-life | ≈5–6 days | ≈1 week |
| Steady state | After 4 weeks | — |
The WEGOVY label's starting dose is written for everyone; it does not shorten the escalation for someone arriving from tirzepatide, and it does not mention tirzepatide at all. That silence is the record — not an instruction to restart at 0.25 mg, and not permission to skip steps.
The registry: no trial of this direction
A ClinicalTrials.gov search on 2026-10-03 for switch, tirzepatide and semaglutide returned two records, neither of which moves anyone onto semaglutide: Pfizer's NCT07845500 (switching to a monthly investigational drug, berobenatide, not yet recruiting) and Boehringer Ingelheim's NCT07855900 (survodutide, not yet recruiting). A search for prior or previous tirzepatide treatment returned none. Lilly's switching trials run the other way, from dulaglutide or another GLP-1 receptor agonist onto tirzepatide, and are described on the reverse page.
The equivalence model, read the other way round
The most-cited source for "equivalent doses" is a 2026 model by Builes-Montaño and colleagues (Diabetes Therapy, PMID 42252377), built from 48 phase 3 arms in 16,524 people with type 2 diabetes. Read from the tirzepatide side:
| Tirzepatide | Semaglutide pairing in the model | Probability of equivalence (±2 points) |
|---|---|---|
| 10 mg | 2.4 mg | 99.4% |
| 15 mg | 7.2 mg | 94.8% |
| 10–15 mg | Lower semaglutide doses | Not equivalent |
Three limits travel with it. It is a model of average percentage weight loss, not of tolerability or of any one person's response. It was built in type 2 diabetes, not in obesity without diabetes. And two of its three authors are Novo Nordisk employees, which the paper declares. The pairing also implies that the 7.2 mg strength — approved on the WEGOVY label in 2026 — is the only semaglutide dose modelled as matching tirzepatide's top dose.
The washout arithmetic
The ZEPBOUND label gives an elimination half-life of about 5 to 6 days. Fraction of the steady-state level remaining after the last weekly injection, by that half-life:
| Time since last tirzepatide dose | Remaining at t½ 5 days | Remaining at t½ 6 days |
|---|---|---|
| 1 week | 38% | 45% |
| 2 weeks | 14% | 20% |
| 3 weeks | 5% | 9% |
| 4 weeks | 2% | 4% |
A semaglutide injection given one week after the last tirzepatide dose therefore lands on top of roughly 40% of the previous drug's level. No source read for this page studies that overlap, in either direction. The semaglutide labels use a one-week gap for their own injection-to-tablet switch, which concerns one molecule, not two. The half-life background is on the tirzepatide half-life page.
The single published case
La Vignera and Condorelli (JCEM Case Reports 2026, PMID 42670558) describe a 40-year-old man with class 2 obesity (BMI 39.8) and marked insulin resistance (HOMA-IR 5.0) who lost under 5% of body weight after six months of tirzepatide escalated to 15 mg weekly, then 20% over six months on semaglutide 2.4 mg weekly. The authors propose that reduced GIP receptor signalling in adipose tissue may explain it, and state that the idea needs prospective validation. It is one person; it shows that the reverse switch has been reported to work, not how often.
Population data run the other way on average. In SURMOUNT-5, people starting tirzepatide 10 or 15 mg lost more weight than those starting semaglutide 1.7 or 2.4 mg (−20.2% against −13.7% at 72 weeks, NEJM 2025, PMID 40353578) — a trial of starting one drug or the other, not of switching.
Why the question comes up
No study counts tirzepatide-to-semaglutide moves on their own. The closest records are about stopping and changing agents in general. Among 288 randomly sampled Cleveland Clinic patients without diabetes who stopped injectable semaglutide or tirzepatide within a year (2022–2024), 47.6% stopped for cost or insurance, 14.6% for side effects, 11.8% for shortages and 1.7% for unsatisfactory weight loss (Gasoyan et al., Obesity 2025, PMID 41039650). In the LUX MED network in Poland, 29.7% of 42,423 patients with repeat GLP-1 prescriptions changed agent at least once between 2018 and 2025, with tirzepatide uptake accelerating after its launch (Łupina et al., Acta Diabetologica 2026, PMID 42142150). Coverage and supply, more than response, are what the records show.
The semaglutide steps in units, research vials
The labels dial semaglutide in milligrams on pens. On a vial, units on a U-100 syringe = mg ÷ mg/mL × 100:
| Semaglutide dose | 2.5 mg/mL | 5 mg/mL | 10 mg/mL |
|---|---|---|---|
| 0.25 mg (label start) | 10 units | 5 units | 2.5 units |
| 0.5 mg | 20 units | 10 units | 5 units |
| 1 mg | 40 units | 20 units | 10 units |
| 1.7 mg | 68 units | 34 units | 17 units |
| 2.4 mg (model pair for tirzepatide 10 mg) | 96 units | 48 units | 24 units |
For orientation only: tirzepatide 10 mg at 10 mg/mL is 100 units, and semaglutide 2.4 mg at 5 mg/mL is 48 — the model's "equivalent" pair occupies different volumes because the molecules are dosed in different milligrams. Water volumes are on semaglutide reconstitution, the formula on mg to units, and any vial in the calculator's semaglutide setting.
What is not established
A dose conversion between the two molecules, a washout interval, whether escalation can be shortened after tirzepatide, and how often a non-responder to one responds to the other. Each would need a trial that does not exist in the registry as of 2026-10-03. The compounds are covered in Medibact's Semaglutide guide and Tirzepatide guide. Nothing on this page is an instruction; switching and dosing belong with the prescriber.
Sources and dates
Opened 2026-10-03: the WEGOVY label via DailyMed (set id ee06186f-2aa3-4990-a760-757579d8f77b, effective 2026-06-18) — sections 2.2 and 2.5; the ZEPBOUND label via openFDA (effective 2026-08-28) — sections 2 and 12.3; ClinicalTrials.gov API searches as described; PubMed abstracts 42252377, 42670558, 41039650 and 42142150; PubMed abstract 40353578 (SURMOUNT-5). Washout percentages and units are arithmetic by this site. Corrections go to the contact page.