Skip to content
PeptideAI
Approved drugfat lossblood sugarcardiovascularinflammation

Tirzepatide

Dual GIP and GLP-1 receptor agonist delivering around 21% mean weight loss at 15 mg weekly - currently the most effective approved obesity drug on the market.

Also known as twincretin, GIP/GLP-1 dual agonist, tirz, Mounjaro, Zepbound, LY3298176

Approved drugLicensed by a major regulator for human use, with phase-3 trial data behind it.

Phase 3 across obesity (SURMOUNT), diabetes (SURPASS), sleep apnoea and heart failure with preserved ejection fraction, plus a head-to-head win over semaglutide in SURMOUNT-5 (20.2% vs 13.7%). The evidence base is genuinely excellent.

How it works

Tirzepatide is built on the GIP backbone with an Aib substitution and a C20 fatty-diacid chain, and it is deliberately imbalanced - roughly five-fold more potent at GIPR than GLPR relative to native hormones. GLP-1 agonism supplies the familiar satiety, delayed gastric emptying and glucose-dependent insulin release. The GIP arm adds improved adipocyte lipid buffering and insulin sensitivity, and central GIPR signalling in the area postrema appears to blunt the nausea that limits pure GLP-1 dosing, which is why higher effective exposures are tolerable. Whether the GIP contribution is agonism or functional receptor desensitisation is still debated in the literature.

Targets: GIP receptor, GLP-1 receptor

Dosing

ProtocolDoseFrequencyRoute
Standard obesity titrationSame day each week, any time of day.2.5 mg – 15 mgonce weeklysubcutaneous
Type 2 diabetes maintenanceSame day each week.5 mg – 15 mgonce weeklysubcutaneous
  • · 2500 mcg weekly for 4 weeks, then 5000 mcg for 4 weeks, then increase by 2500 mcg every 4 weeks as needed to a maximum of 15000 mcg. Many people get most of the benefit at 5000-10000 mcg and never need the top dose.
  • · Maintenance doses of 5, 10 or 15 mg; HbA1c reductions of 2.0-2.4% are typical.

Titration

Four weeks minimum per step. Tirzepatide is generally better tolerated than semaglutide at equivalent efficacy, but the 12.5 and 15 mg steps are where most people hit their GI ceiling.

Cycling

Not cycled. Like semaglutide it is a maintenance therapy - SURMOUNT-4 showed rapid regain after withdrawal. Some people hold a lower maintenance dose (2.5-5 mg weekly) once at goal weight rather than stopping outright.

Work out your exact syringe units →

Pharmacology

Half-life
About five days, supporting once-weekly dosing; steady state at roughly four weeks.
Onset
Appetite drops within the first week; weight loss in SURMOUNT-1 was still continuing at 72 weeks.
Routes
subcutaneous
Molecule
Acylated 39-amino-acid GIP/GLP-1 dual agonist peptide
Sequence length
39 amino acids
Molecular weight
4813.5 Da

Handling

Diluent
Bacteriostatic water
Typical mix
1 or 1.5 or 2 mL
Vial sizes
10, 15, 20, 30, 60 mg
Lyophilised
Refrigerate at 2-8 C long term; tolerates room-temperature shipping.
Reconstituted
Refrigerated, use within about 28-30 days.
Light sensitive
Yes — keep it out of the light

Oral — not a viable route

No oral formulation exists. Unlike semaglutide there is no SNAC-paired tablet, so there is no route by which a swallowed dose survives.

Mixing

A 30 mg vial in 1.5 mL gives 20000 mcg/mL, so 5 units on a U-100 syringe is 1000 mcg. Always recompute units from your own vial size and diluent volume - the same 5 units is 1500 mcg if you mix that vial with 1 mL and 750 mcg if you mix it with 2 mL. Swirl gently; the peptide is surfactant-free and foams easily.

Side effects

  • very commonNauseaReported by roughly a quarter to a third of users, concentrated around dose increases.
  • very commonDiarrhoeaSlightly more common than with semaglutide.
  • very commonLoss of lean massSame problem as all incretins - protect it with resistance training and high protein, not hope.
  • commonConstipation
  • commonInjection-site reactionMore frequent with grey-market reconstitutions than with pens.
  • uncommonGallbladder diseaseRate rises with the speed of weight loss.
  • uncommonHypoglycaemia when combined with insulin or sulfonylureasTirzepatide alone almost never causes it.
  • uncommonDelayed gastric emptying and aspiration risk under anaesthesiaSame class risk as semaglutide. Tell any anaesthetist or endoscopist you are on it; most centres want weekly GLP-1 agonists held for about a week before an elective procedure.
  • rarePancreatitisRare but real; severe epigastric pain means stop.

Do not use if

  • Personal or family history of medullary thyroid carcinoma or MEN2 - boxed warning.
  • History of pancreatitis.
  • Pregnancy; tirzepatide also reduces the effectiveness of oral contraceptives, so switch to a non-oral method for four weeks after starting and after each dose increase.
  • Severe gastroparesis.

Combining it

  • redundantsemaglutideOverlapping GLP-1 agonism; no reason to run both.
  • cautioninsulin-analoguesCut mealtime insulin substantially at initiation to avoid hypoglycaemia.
  • synergycagrilintideAmylin plus incretin is the current frontier combination; formally studied for semaglutide, used anecdotally with tirzepatide.
  • conflictipamorelinGhrelin-receptor agonists drive appetite in the opposite direction and will partly undo the point of the drug.

What to monitor

  • · Weight and waist weekly.
  • · HbA1c every 3-6 months where relevant.
  • · Body composition every 3-4 months.
  • · Resting heart rate and blood pressure.
  • · Contraception method if you are on the pill and could become pregnant.

Legal status

FDA- and EMA-approved as Mounjaro (diabetes) and Zepbound (obesity). Grey-market 'research' tirzepatide is sold widely and is not legal for human use.

References

  • Jastreboff et al. 2022, SURMOUNT-1, NEJM - 20.9% weight loss at 72 weeks on 15 mg (trial)
  • Aronne et al. 2024, SURMOUNT-4, JAMA - regain after withdrawal (trial)
  • Aronne et al. 2025, SURMOUNT-5, NEJM - head-to-head vs semaglutide (trial)
  • Zepbound and Mounjaro US prescribing information (label)

Mechanism in depth

Tirzepatide is not a GLP-1 drug with GIP bolted on; it is a GIP peptide engineered to also hit GLP1R, and the asymmetry matters. Relative to the native hormones it is roughly balanced at GIPR but substantially weaker at GLP1R, and it is strongly cAMP-biased at GLP1R with markedly reduced beta-arrestin recruitment. Reduced beta-arrestin means less receptor internalisation and desensitisation, so a molecule that binds GLP1R less tightly can still deliver more sustained downstream signal than a stronger, more balanced agonist - this is the pharmacological trick, and it is why potency measured at the binding site badly predicts clinical effect here. The GIP arm does three things that the GLP-1 arm does not. In white adipose tissue GIPR activation increases blood flow and lipoprotein lipase activity, improving the tissue's ability to buffer a postprandial lipid load rather than spilling it into liver and muscle - this is the most credible explanation for why tirzepatide improves insulin sensitivity more than weight loss alone predicts. In the central nervous system GIPR is expressed in the area postrema and hypothalamus, and central GIPR agonism reduces emesis in animal models, which is the leading explanation for why tirzepatide is better tolerated than a GLP-1 agonist of equivalent efficacy. And GIPR signalling in bone and in the vasculature produces effects nobody has fully characterised yet. There is a genuine open controversy here worth stating plainly: GIPR antagonism also produces weight loss in humans (that is exactly what MariTide does), so the field does not agree on whether tirzepatide's GIP arm works through sustained agonism or through functional receptor desensitisation that ends up looking like antagonism. Both camps have data. Nobody has resolved it.

What usually goes wrong

The characteristic tirzepatide failure is the 12.5 and 15 mg steps. Tolerability is genuinely good up to 10 mg and then falls off a cliff for a lot of people, and because the drug is so effective they keep climbing anyway. Second, insulin-treated users who do not cut their prandial insulin at initiation - a 2.0-2.4% HbA1c reduction arriving over eight weeks will cause hypoglycaemia if the background regimen is untouched. Third, sulfur burps, reflux and constipation, which are mundane and are nonetheless the reason a lot of people quit. Fourth, the lean-mass problem is larger than with semaglutide purely because the total weight loss is larger. Fifth, the oral contraceptive interaction produces unplanned pregnancies and it is written on the label precisely because it happened. And sixth, grey-market reconstitution error: a 30 mg vial in 1.5 mL versus 1 mL versus 2 mL gives 1000, 1500 or 750 mcg per 5 units, and people who copy someone else's unit count from a forum without recomputing for their own vial routinely take double or half what they intend.

Titration ladder

  1. 2.5 mgWeeks 1-4 — Starting dose, not a therapeutic one. Its job is to let the gut adapt.
  2. 5 mgWeeks 5-8 — The first properly effective dose. A large fraction of people would get most of their result by simply staying here for a year.
  3. 7.5 mgWeeks 9-12 — Optional step. Only climb if weight loss has genuinely stalled for four weeks and side effects are mild.
  4. 10 mgWeeks 13-16 — The common practical ceiling. SURMOUNT-1 gave 19.5% at this dose versus 20.9% at 15 mg - a small return for the extra nausea.
  5. 12.5 mgWeeks 17-20 — Where most people hit their gastrointestinal limit.
  6. 15 mgWeek 21 onward — Maximum approved dose. Diminishing returns are real above 10 mg.

Bloodwork worth running

MarkerWhenWhy it matters
HbA1cBaseline, 3 months, then 3-6 monthly.Tirzepatide is the strongest glycaemic agent in the class - SURPASS trials routinely produced 2.0-2.4% reductions, which is enough to make a co-prescribed insulin or sulfonylurea dangerous if left unchanged.Act if: Below 5.5% on background insulin or sulfonylurea means cut that agent.
Fasting glucose and, ideally, CGM during escalationContinuously or weekly through escalation.The 2.0%+ HbA1c reductions arrive fast. If you are on insulin, a continuous glucose monitor for the first eight weeks is worth more than any bloodwork panel.Act if: Any reading under 3.9 mmol/L (70 mg/dL) means the background agent is too high.
ALT, AST and where available liver fat by MRI-PDFFBaseline and 6 months.Tirzepatide produces large reductions in liver fat and, in the SYNERGY-NASH trial, histological improvement in MASH. If MASLD is part of why you are using it, this is the marker that shows it working.Act if: Rising ALT on treatment is not expected and needs another explanation.
Body composition (DEXA, or bioimpedance as a poor substitute)Baseline and every 3-4 months.Not bloodwork, but the single most important measurement on this drug. Tirzepatide produces the largest weight loss in the approved class and therefore the largest absolute lean-mass loss.Act if: If more than about a third of the weight lost between scans is lean tissue, the protein and training plan has failed and the dose should stop climbing until it is fixed.
Lipid panelBaseline and 6 months.Triglycerides fall sharply and the GIP arm's effect on adipose lipid handling shows up here more than with pure GLP-1 agonists.Act if: None; confirmation marker.
Creatinine and eGFRBaseline and after any prolonged vomiting or diarrhoea.Same dehydration-driven pre-renal risk as the rest of the class.Act if: A 30% rise means stop and rehydrate.
Ferritin, B12, vitamin D and albuminBaseline and 6-monthly.Twenty percent body-weight loss over 18 months means a very long period of low food volume.Act if: Ferritin under 30 ng/mL or albumin under 35 g/L means intake is genuinely inadequate.
LipaseSymptom-driven only.Only as a diagnostic test in someone with abdominal pain. Asymptomatic elevations are common and meaningless.Act if: Three times the upper limit of normal plus pain means stop and image.

Pharmacokinetics

Tmax
24 h
Bioavailability
80%
Volume of distribution
10.3 L
Protein binding
99%
Time to steady state
28 days
Crosses blood-brain barrier
partial
Accumulates
Yes — doses stack before steady state
Metabolism
Proteolytic cleavage of the peptide backbone, beta-oxidation of the C20 fatty di-acid and amide hydrolysis. No CYP450 route, so no classical drug-drug interactions - but delayed gastric emptying does alter the absorption of oral drugs, and that is the reason oral contraceptives are specifically called out.
Elimination
Metabolites in urine and faeces. Intact tirzepatide is not detectable in either, which tells you the molecule is fully catabolised before excretion.

Receptor targets

  • GIP receptor (GIPR)Comparable to native GIP; the frequently quoted Kd is around 0.1-0.2 nM from Lilly's discovery work, not re-verified here

    Enhanced glucose-dependent insulin secretion, improved adipose lipid buffering and insulin sensitivity, and central signalling in the area postrema that appears to blunt nausea.

  • GLP-1 receptor (GLP1R)Roughly five-fold weaker than native GLP-1 at the binding site, but cAMP-biased with reduced beta-arrestin recruitment

    Satiety, delayed gastric emptying, glucagon suppression and glucose-dependent insulin release - the same clinical picture as semaglutide, produced by a weaker binder with better signalling persistence.

  • Albumin99% bound

    C20 di-acid depot binding, giving the 5-day half-life.

Trials

  • SURMOUNT-1 Phase 3 · n=2539 · 72 weeks · 2022

    Mean weight reduction of 15.0%, 19.5% and 20.9% at 5, 10 and 15 mg weekly versus 3.1% for placebo.

  • SURMOUNT-4 Phase 3 withdrawal · n=670 · 88 weeks · 2024

    Continued tirzepatide gave a further 5.5% loss while switching to placebo produced 14.0% regain - the definitive demonstration that this is maintenance therapy, not a course.

  • SURMOUNT-5 Phase 3 head-to-head · n=751 · 72 weeks · 2025

    20.2% weight loss with tirzepatide versus 13.7% with semaglutide 2.4 mg.

  • SYNERGY-NASH Phase 2 · n=190 · 52 weeks · 2024

    Resolution of metabolic dysfunction-associated steatohepatitis without worsening of fibrosis in a substantially higher proportion than placebo.

What to expect, and when

Days 1-4: appetite drops, often more abruptly than with semaglutide. Weeks 1-4: gastrointestinal adaptation; nausea peaks 24-72 hours after each dose step. Week 4: steady state. Weeks 4-12: the fastest phase of weight loss, and where early rapid responders separate from slow responders - a SURMOUNT-5 post hoc analysis found early rapid loss predicted the final result. Weeks 20-52: steady accrual, roughly 1-2% of body weight per month. Weeks 72-88: still falling in SURMOUNT-1 and SURMOUNT-4, which is unusual - most obesity drugs have plateaued by then. Stopping: SURMOUNT-4 showed 14% regain over 88 weeks off drug.

Stacking and comparisons

Cagrilintide is the only addition with a mechanistic argument that does not overlap - amylin acts through calcitonin and RAMP-based receptors, not incretin receptors, and appears to spare lean mass. It has been formally studied with semaglutide, not tirzepatide, so combining it with tirzepatide is an extrapolation people make routinely and no trial supports. Anything that is already a GLP-1 or GIP agonist is redundant by definition, including retatrutide, semaglutide and VK2735. Ghrelin-receptor agonists (ipamorelin, MK-677, hexarelin) directly oppose the drug's purpose and MK-677 additionally worsens insulin sensitivity. The useful non-peptide additions are the same as for semaglutide: heavy resistance training, 1.6-2.2 g/kg protein, creatine, magnesium, adequate fibre, and enough sodium and water to survive the gastrointestinal losses. One tirzepatide-specific interaction that people miss: it reduces the effectiveness of oral contraceptives through delayed gastric emptying, so a non-oral method is needed for four weeks after starting and after each dose increase.

Against semaglutide: tirzepatide won SURMOUNT-5 decisively (20.2% versus 13.7%) and is generally better tolerated per unit of weight lost. Semaglutide's remaining advantages are the SELECT cardiovascular outcome data in non-diabetics, a longer record, and an oral form. Against retatrutide: retatrutide's phase 3 numbers are higher, but it adds glucagon-driven tachycardia and transient hyperglycaemia and is not approved anywhere, so you are trading a licensed product with a supply chain for a research chemical. Against CagriSema: REDEFINE-4 failed to show CagriSema was non-inferior to tirzepatide, which is a quiet but important result - the amylin combination did not beat the dual agonist. Against MariTide: MariTide dosing is monthly and regain after stopping appears much slower, but it is phase 3 and unavailable.

Rough cost

$350–$1100/month. Manufacturer single-dose vial programmes sit around 350-500 per month for lower doses; US list price for Zepbound or Mounjaro without coverage is roughly 1,000-1,100. Grey-market research vials are far cheaper but tirzepatide is among the most counterfeited peptides on the market. Market observation from mid-2026, not verified pricing.

Genuinely uncertain

  • Whether the GIP component acts as a sustained agonist or produces functional receptor desensitisation is genuinely unresolved in the literature, and the fact that GIPR antagonism also causes weight loss is the reason.
  • Receptor affinity figures are from Lilly's published discovery work and were not independently resolved in this session.
  • The amino-acid sequence is not reproduced here because it was not verified against a primary source; the listed modifications are from the published molecular description.
  • Long-term GIPR agonism effects on bone metabolism are not characterised in humans.
  • No cardiovascular outcomes trial has read out for tirzepatide in the way SELECT has for semaglutide, so cardiovascular benefit is currently inferred from risk factors rather than demonstrated on events.
  • Cost figures are market observations, not verified pricing.

Papers