Tesamorelin

Identity and regulatory status

Tesamorelin is human growth hormone-releasing hormone, all forty-four residues, carrying two modifications. A (3E)-hex-3-enoyl group sits on the tyrosine at position 1, and the C-terminal leucine is amidated. PubChem CID 16137828 records the formula C221H366N72O67S and an average mass near 5,136 g/mol. FDA labeling gives the same formula, identifies the drug substance as an acetate salt with approximately seven acetate equivalents, and states a free-base mass of 5,135.9 Da. The FDA Global Substance Registration System lists the free peptide under UNII MQG94M5EEO with CAS 218949-48-5, INN 8786 and USAN SS-04, and registers the acetate separately under UNII LGW5H38VE3 with CAS 901758-09-6.

Three CAS numbers circulate and each has a different referent. 218949-48-5 is the free peptide. 804475-66-9 resolves in PubChem to the same compound record, CID 16137828. 901758-09-6 returns a PUGREST.NotFound from a PubChem name query, which has led to its being treated as spurious, but the GSRS acetate record carries it, and the acetate is what the approved product contains. A fourth disagreement sits inside GSRS itself: the record displays a calculated formula of C221H365N71O69S at 5,040 Da, which does not reconcile with the amidated C-terminus described in that record's own structural notes. PubChem and the labeling agree with each other; the GSRS calculation agrees with neither.

Approval status is the unusual part of this record. Tesamorelin was approved by the FDA in 2010 for the reduction of excess abdominal fat in adults with HIV-associated lipodystrophy, and that remains the only approved indication. DailyMed currently carries two prescribing-information documents for the compound, both stamped with an initial US approval year of 2010, the newer describing a more concentrated 11.6 mg vial presentation. Approval covers one product, one indication and one population. It does not extend to material supplied outside that channel, which has not passed the identity, purity or sterility controls that apply to a licensed article. Growth hormone-releasing hormone analogues, tesamorelin among them named, are also prohibited in sport under the WADA code.

What the registration trials measured

Two phase 3 trials carry the approval. Falutz and colleagues randomised 412 adults with HIV and abdominal fat accumulation on antiretroviral therapy, 86% male, to 2 mg subcutaneous tesamorelin daily or placebo for 26 weeks. Visceral adipose tissue measured by CT fell 15.2% in the treated arm and rose 5.0% on placebo (P<0.001). IGF-1 rose 81.0% against a 5.0% fall on placebo. Triglycerides fell 50 mg/dL against a 9 mg/dL rise, and the total cholesterol to HDL ratio fell 0.31 against a 0.21 rise. Glycaemic measures did not differ significantly between the groups at 26 weeks.

Pooling both phase 3 trials across 806 randomised participants, the same group reported a visceral adipose tissue treatment effect of −15.4% at week 26, with an absolute change of −24 ± 41 cm² on tesamorelin against +2 ± 35 cm² on placebo. Mean IGF-1 rose 108 ± 112 ng/mL against −7 ± 64. Subcutaneous adipose tissue showed no significant change in either arm, which is the finding that distinguishes this compound's measured effect from general fat loss: the change recorded was in the visceral compartment specifically, and total body weight was not the endpoint.

Both trials enrolled a population selected for a particular physiology — adults on long-term antiretroviral therapy with central fat accumulation, a group in which growth hormone secretion is characteristically reduced. That selection is the single most important qualifier on every figure above. The trials establish what happened in that population over 26 weeks. They do not establish what happens in anyone else, and the compound has never been tested against a general obesity population.

The 52-week extension and what withdrawal did

A second phase 3 trial, 404 participants, used a re-randomisation design that answers a question the 26-week readouts cannot. Participants received tesamorelin or placebo 2:1 for six months, at which point those on tesamorelin were re-randomised either to continue or to switch to placebo. At 26 weeks, visceral adipose tissue had fallen 10.9%, a reduction of 21 cm², against 0.6% in the placebo arm (P<0.0001). Among those who continued through twelve months, the reduction reached approximately 18% (P<0.001).

Among those switched to placebo at six months, the investigators recorded that the initial reduction was rapidly lost. That sentence is the durability finding for this compound, and it is the one most often left out of secondary summaries, which tend to quote the 18% figure without the arm that produced its counterpart. No published trial has measured what happens after any longer period of continuous administration; 52 weeks is the ceiling of the randomised record.

Outside HIV: two trials and a registry posting

Makimura and colleagues randomised 60 abdominally obese adults without HIV, selected for reduced growth hormone secretion, to 2 mg subcutaneous tesamorelin daily or placebo for twelve months. Visceral adipose tissue fell by 35 cm² relative to placebo (95% CI −58 to −12, P=0.003), carotid intima-media thickness by 0.04 mm (P=0.02), C-reactive protein by 0.15 mg/L (P=0.04) and triglycerides by 37 mg/dL (P=0.02). IGF-1 rose 92 µg/L (P<0.0001). No change was seen in fasting glucose, two-hour glucose or glycated haemoglobin. The entry criterion again required documented reduced growth hormone secretion.

Liver fat has been measured twice. In HIV-associated fatty liver disease, Stanley and colleagues randomised 61 participants with a hepatic fat fraction of at least 5% for twelve months and reported an absolute effect on hepatic fat fraction of −4.1 percentage points (95% CI −7.6 to −0.7, P=0.018), a relative reduction of 37%, with 35% of the treated arm and 4% of the placebo arm falling below a 5% fat fraction. Fasting glucose and glycated haemoglobin did not differ between groups at twelve months.

The equivalent trial without HIV exists but has not been published. NCT03375788, run at Massachusetts General Hospital, randomised adults aged 18 to 65 with a BMI of at least 30 and hepatic steatosis to 2 mg subcutaneous tesamorelin daily or placebo for twelve months. Results were posted to the registry in November 2025: median liver fat change of −5.3% (IQR −13.0 to −1.4) in 18 analysed participants against +3.6% (IQR −6.2 to +7.9) in 16 on placebo. A registry posting is not a peer-reviewed publication, and no such publication was located in PubMed at the time of writing.

The cognition record does not converge

Baker and colleagues randomised 152 adults aged 55 to 87, of whom 61 completers had mild cognitive impairment and 76 were cognitively healthy, to 1 mg subcutaneous tesamorelin daily or placebo for 20 weeks followed by a ten-week washout. A composite cognitive outcome favoured the treated arm at P=.03 in the intention-to-treat analysis and P=.002 among completers, with executive function at P=.005 and verbal memory at P=.08. IGF-1 rose 117% and body fat fell 7.4%. Fasting insulin rose 35% in the mild cognitive impairment group and did not change in the healthy adults.

Two subsequent trials did not reproduce that result. Ellis and colleagues randomised 73 adults with HIV and abdominal obesity 3:2 to 2 mg subcutaneous tesamorelin daily or standard of care for six months and reported no significant between-group difference in neurocognitive change (P=0.673), with a within-arm trend on tesamorelin that did not reach significance (mean change 0.146, 95% CI −0.002 to 0.294, P=0.060). Waist circumference fell by a median 2.7 cm (P=0.015). A 22-participant pilot at 1 mg daily for ten weeks, published in 2026, reported no significant change on its study measures.

Three trials, three different populations, three different doses and durations, and one positive composite outcome out of three. Any summary presenting the cognition literature as a settled finding is quoting the 2012 trial and omitting the two that followed it. Neither of the later trials was powered as a confirmatory replication, so the honest reading is that the question was raised in 2012 and has not since been answered in either direction.

Safety findings the trials recorded

Glucose is the signal the labeling treats most seriously. Elevated glycated haemoglobin at or above 6.5% was recorded from baseline to week 26 in 5% of treated participants against 1% on placebo, and the labeling reports a hazard ratio for developing diabetes of 3.3 (95% CI 1.4 to 9.6). This sits alongside the individual trials, several of which reported no significant between-group difference in glucose parameters at 26 or 52 weeks. Both statements come from the same programme. The trial-level comparisons were not powered for incident diabetes; the pooled hazard ratio is the figure that addresses it.

IGF-1 rises substantially and the labeling quantifies how far. At 26 weeks, 47% of treated patients had IGF-1 above 2 standard deviation scores and 36% above 3; at 52 weeks the corresponding figures were 34% and 23%. Anti-tesamorelin IgG antibodies were detected in 50% of patients treated for 26 weeks and 47% of those treated for 52 weeks, and in 85% of the subset who had hypersensitivity reactions. Visceral fat reduction and IGF-1 response were similar in patients with and without those antibodies.

Recorded adverse events in the 26-week phase 3 pool of 543 treated and 263 placebo participants were injection site reactions (17% against 6%), arthralgia (13% against 11%), myalgia and peripheral oedema (6% against 2% each) and pain in extremity (6% against 5%). Hypersensitivity reactions occurred in 4%. The labeling carries warnings on new malignancy, elevated IGF-1, fluid retention presenting as oedema, arthralgia and carpal tunnel syndrome, glucose intolerance, hypersensitivity and injection site reactions. The ten-year prospective cohort safety study that would have addressed the malignancy question, NCT01579695 with 391 participants, was terminated in 2018 with no results posted.

Pharmacokinetics, and the half-life that circulates

The approved labeling reports a mean elimination half-life of 8 minutes in healthy subjects after a single subcutaneous dose in one document, and 11 minutes in the other. Absolute subcutaneous bioavailability was determined to be less than 4%. Median time to peak concentration was 0.15 hours, mean peak concentration 2,956 pg/mL, and mean volume of distribution 4.8 ± 1.9 L/kg. A population pharmacokinetic analysis across 38 HIV-infected patients and healthy subjects fitted a one-compartment model with plasma clearance of 1,060 L/h and a volume of distribution of 200 L, and reported that the fraction absorbed by a first-order process was 13.1% higher on day 14 than on day 1.

Figures of 26 to 38 minutes appear widely and do not come from either labeling document. The nearest measured values in that range are canine: Ferdinandi and colleagues reported an elimination half-life of 21 to 45 minutes in dogs given the compound intravenously and subcutaneously at doses up to 600 µg/kg per day. That same non-clinical paper is the origin of the structural rationale repeated everywhere — that the hexenoyl group on Tyr1 confers resistance to dipeptidyl aminopeptidase-IV deactivation and slowed in vitro degradation in rat, dog and human plasma. The rationale is sourced. The human half-life figure attached to it is not.

Endogenous secretion has been characterised once directly. Stanley and colleagues gave 2 mg subcutaneously daily for two weeks to 13 healthy men of mean age 45 and reported an increase in mean overnight growth hormone of 0.5 ± 0.1 µg/L (P=0.004), in average log growth hormone peak area of 0.4 ± 0.1 (P=0.001) and in basal secretion of 0.008 ± 0.003 µg/L per minute (P=0.008), with IGF-1 up 181 ± 22 µg/L. Neither fasting glucose (P=0.93) nor insulin-stimulated glucose uptake (P=0.61) changed significantly. Thirteen men over two weeks is the whole of the direct human secretion-dynamics evidence.

What is not known

The randomised record stops at twelve months, and one trial showed the visceral fat reduction was rapidly lost when treatment was withdrawn at six months, so nothing is established about what happens after a year in either direction. The ten-year prospective cohort study designed to address the malignancy question, NCT01579695 with 391 participants, was terminated in 2018 and posted no results; the labeling's warning on new malignancy therefore rests on background risk reasoning rather than on completed long-term surveillance. Glucose sits unresolved: individual trials reported no significant between-group differences at 26 and 52 weeks, while the pooled labeling reports a hazard ratio of 3.3 for developing diabetes, and no trial was powered to settle that. The compound has never been tested in an unselected obesity population, in healthy young adults, in adolescents, or during pregnancy or lactation, and every trial outside HIV either required documented reduced growth hormone secretion at entry or remains unpublished. The cognition question raised in 2012 has not been answered by the two later trials, neither of which was designed as a replication. Half of treated patients developed anti-tesamorelin IgG antibodies, and while efficacy appeared unaffected over 52 weeks, the consequence of that response over longer exposure is unstudied. Physical behaviour in solution is separately undocumented and is covered elsewhere on this site.

Questions