Tesofensine's reported phase 3 is missing from ClinicalTrials.gov.

Why this note exists

Tesofensine occupies an unusual position. It is not a compound with no evidence — the sort where the honest note is three paragraphs long and ends early. Nearly a thousand people took it in randomized, placebo-controlled trials before 2010, and the weight-loss numbers from those trials are large enough that they get quoted alongside GLP-1 results.[1][5][18] It is a compound where the published evidence and the company-reported later trial are not equally inspectable, and where the two most frequently cited facts about it — a headline weight-loss percentage and a completed phase 3 — are each carrying a qualification that rarely travels with them.

In the index generated for this note, 29 of the 104 vendors pepmg tracks carried a tesofensine listing, 44 listings in total.[27] Twenty-five of the 44 are sold as oral capsules, tablets or strips rather than as material for reconstitution — unusual on a shelf otherwise built around injectable powders.[27]

A note on what kind of evidence this is: every efficacy and safety figure below is human data and carries its design, phase and participant count. Two paragraphs report animal work and are labeled as such. Company statements that have no corresponding publication are attributed to the company and identified as unpublished. pepmg reports doses only as their sources published them, and does not convert a trial dose into a protocol.

The chemistry

It is on peptide shelves, and it is not a peptide

This is the least interesting fact in the note and the one most worth getting out of the way. PubChem's record for tesofensine gives the IUPAC name as (1R,2R,3S,5S)-3-(3,4-dichlorophenyl)-2-(ethoxymethyl)-8-methyl-8-azabicyclo[3.2.1]octane — an azabicyclooctane, the tropane skeleton — with molecular formula C₁₇H₂₃Cl₂NO and a molecular weight of 328.3.[25] The published literature describes it consistently as a triple reuptake inhibitor of noradrenaline, dopamine and serotonin.[1][5]

That matters for one practical reason. A buyer reasoning by analogy from other items on the same shelf — subcutaneous, peptide, short half-life, reconstituted from powder — will be reasoning about the wrong molecule. Tesofensine is an orally active central stimulant with a long elimination profile, and its consistent signal across the human trials is cardiovascular, not local.[1][5][23]

The origin

The weight loss was a side effect in trials for two other diseases

Tesofensine entered clinical development as NS 2330, for Parkinson's disease and Alzheimer's disease. The registry still holds those studies: a 261-patient and a 254-patient phase 2 in Parkinson's disease, and a 430-patient phase 2 in mild-to-moderate dementia of the Alzheimer's type, all sponsored by Boehringer Ingelheim, all completed.[9][10][11] The 430-patient dementia study remains the largest tesofensine trial on ClinicalTrials.gov, and it was not a study about weight.[11][12]

The obesity program grew out of a pooled analysis of those neurology trials, published in Obesity in 2008. Four randomized, double-blind, multicentre trials, two in each disease, compared tesofensine in 740 patients against placebo in 228, dosed once daily for 14 weeks with no weight-loss program of any kind.[5] Weight change after 14 weeks was +0.5 percent on placebo against −0.5, −0.9, −1.8 and −2.8 percent at 0.125, 0.25, 0.5 and 1.0 mg, with a dose effect at P = 0.015.[5] In the obese subgroup, the proportions reaching at least 5 percent weight loss were 2.1 percent on placebo against 8.2, 14.1, 20.9 and 32.1 percent.[5]

Two things belong in the same breath as those numbers. The first is the heart rate: −0.4, +2.1, +4.2, +6.0 and +6.8 beats per minute across the same dose ladder, significant from 0.25 mg upward, with no effect on blood pressure observed in that analysis.[5] The second is what kind of result this is. It is a post-hoc weight analysis of trials designed to test a different question in two neurodegenerative diseases, in patients who were mostly not obese — 14 percent of the placebo group and 21 percent of the tesofensine group were.[5] It is a strong signal generator and it is not an obesity efficacy trial.

The pivotal evidence

203 patients, 24 weeks, and a sentence the authors wrote themselves

The trial everyone is citing is TIPO-1, published in The Lancet in November 2008. It was a phase 2, randomized, double-blind, placebo-controlled trial at five Danish obesity management centres. After a two-week run-in, 203 patients with a body-mass index of 30 to 40 were prescribed an energy-restricted diet and randomized to tesofensine 0.25 mg (n=52), 0.5 mg (n=50) or 1.0 mg (n=49), or placebo (n=52), once daily for 24 weeks. The primary outcome was percentage change in bodyweight, analysed by modified intention to treat. 161 participants, 79 percent, completed.[1][6]

The placebo row reports total weight loss; the three tesofensine rows report additional percentage-point reductions beyond diet and placebo at 24 weeks, as reported in the trial publication. A phase 2 trial, 203 randomized, 161 completing, at five centres in one country.[1]

The 9.2 figure is a placebo-adjusted percentage-point difference, not total weight loss, and the reason it gets quoted is that it sits in GLP-1 territory while coming from a daily oral tablet. It is a real result from a real randomized trial, and the qualifications are all in the same paper. Every arm, including placebo, was on an energy-restricted diet, so the placebo arm's own 2.0 percent is part of the comparison.[1] One patient in five did not complete.[1] The most common adverse events attributed to tesofensine were dry mouth, nausea, constipation, hard stools, diarrhoea and insomnia.[1] At 24 weeks the 0.25 mg and 0.5 mg groups showed no significant increase in systolic or diastolic blood pressure against placebo, but heart rate in the 0.5 mg group was up 7.4 beats per minute.[1]

And the authors' own conclusion is the cleanest summary anyone has written of where this compound stands. Their interpretation reads that the results "suggest that tesofensine 0.5 mg might have the potential to produce a weight loss twice that of currently approved drugs" — and then, in the same breath, that "these findings of efficacy and safety need confirmation in phase III trials."[1] That was written in 2008. Eighteen years later, no phase 3 trial of tesofensine appears on ClinicalTrials.gov.[12]

The complication

The expression of concern, and what it does and does not cover

On April 6, 2013, The Lancet published an expression of concern attached to the 2008 tesofensine trial.[2] It followed an inspection by the Danish Health and Medicines Authority, conducted in November 2011. As reported at the time, the inspection raised three matters: at one site, informed consent was taken by non-medical personnel, which Danish rules require a medical practitioner to do; the inspection questioned whether the trial's blinding had held; and adverse events were incompletely recorded, with investigators instructed not to register headache, migraine, stress and depression as adverse events where patients had reported those conditions before randomisation.[4] The report's conclusion, as quoted, was that "the overall side-effect profile that is…published in The Lancet is not in accordance with the actual course of the trial."[4]

The same report found that the efficacy data held. It concluded that "data on primary and secondary end-points, and serious adverse events, were reported in accordance with the source data."[4] That distinction is the whole point of reading the document rather than the headline: the weight-loss numbers in the table above are not what was called into question. The adverse-event profile printed alongside them is.

The trial's authors replied in The Lancet in July 2013 under the title "Under-reporting of adverse effects of tesofensine."[3] The paper has not been retracted. Checked on September 7, 2026, PubMed's record for the 2008 trial still carries the expression of concern, and pepmg found no notice withdrawing it.[1][2]

The honest reading: this is a thirteen-year-old, unresolved editorial flag on the single most-cited efficacy trial for a compound sold today, and it concerns exactly the half of the trial — tolerability — that a buyer has the least other data on.

The gap

The phase 3 that is described in press releases and absent from the registry

The sponsor's account is specific. Saniona's pipeline page states that its partner Medix completed a phase 3 study of tesofensine at 0.25 and 0.50 mg daily, and that the study "confirmed the compelling efficacy and favorable safety profile of tesofensine in obesity previously observed in Phase 2."[18] The company's November 6, 2024 statement puts the phase 3 program at 372 patients, describes average weight loss of about 10 percent over 24 weeks with more than half of patients exceeding that, and cites a broader clinical safety database of roughly 1,600 patients.[19]

Set that against what the obesity program actually left on the registry, which is three studies and nothing else: TIPO-1 itself at 200 enrolled, a 140-patient phase 2 long-term safety evaluation in patients with obesity, and a 32-patient phase 1/2 study of energy balance.[6][7][12]

pepmg cannot check any of it, and says so rather than repeating it as evidence. That study does not appear among the 13 tesofensine studies returned by ClinicalTrials.gov on September 7, 2026, and a PubMed search on the same date returned no peer-reviewed report of a phase 3 obesity trial of tesofensine — the most recent primary human obesity result in the indexed literature remains a 21-patient randomized trial of the metoprolol combination from 2022.[12][13] pepmg did not locate a public protocol, dropout table or full adverse-event report for that company-described trial in the sources searched. It may well have found what the company says it found. It cannot be examined.

Counts from the ClinicalTrials.gov registry search for tesofensine on September 7, 2026. The registry is not the world — a trial can run outside it — but it is the record a reader can audit.[12]

The regulatory sequence in Mexico is worth stating precisely, because it is routinely compressed into "approved in Mexico." In February 2023 Cofepris' technical committee expressed a favorable opinion on tesofensine for the treatment of obesity.[20] The sponsor's own characterisation is that the opinion "did not represent neither a market authorization nor a rejection" but a non-binding technical opinion, one step in reviewing new molecules.[19] On November 6, 2024, under the headline "Mexican Application for Tesofensine Not Yet Approved," the company said Medix was "entering a dialogue with the agency regarding the path forward as it appears that the decision by Cofepris has not been based on the full data package as submitted by Medix."[19] pepmg found no later announcement of an approval.

Tesomet

The combination exists because of the heart rate, and its trials are very small

Tesomet is tesofensine co-formulated with metoprolol, a beta-1 blocker, and the reason for it is the cardiovascular signal that runs through every dataset above. That rationale was worked out in animals: a 2013 study in conscious telemetrized rats reported that tesofensine caused dose-dependent increases in heart rate and blood pressure alongside its effect on food intake, and that combined treatment with metoprolol "fully prevented the cardiovascular sympathetic effects of tesofensine while leaving the robust inhibitory efficacy on food intake unaffected."[26] That is rat data, and it is included here because it explains the product, not because it establishes anything in people.

The human trials of the combination are small, and the largest published one had safety as its primary endpoint. In a randomized controlled trial published in the European Journal of Endocrinology in 2022, 21 adults with hypothalamic obesity, 16 of them female, were randomized to Tesomet (0.5 mg tesofensine / 50 mg metoprolol) or placebo for 24 weeks alongside diet and lifestyle counselling; 18 completed.[13] Against placebo, Tesomet produced an additional mean weight change of −6.3 percent (95% CI −11.3 to −1.3; P = 0.017) and more patients reaching at least 5 percent weight loss, 8 of 13 against 1 of 8 (P = 0.046); the reduction in waist circumference, 5.7 cm, did not reach significance (P = 0.054).[13]

The adverse events from that trial are the most useful tolerability data in the human record, precisely because it was designed to collect them: sleep disturbances in 50 percent on Tesomet against 13 percent on placebo, dry mouth 43 against 0 percent, headache 36 against 0 percent, and one Tesomet-related serious adverse event, an exacerbation of pre-existing anxiety that led to discontinuation.[13] No significant differences in heart rate or blood pressure were observed between groups — which is what the metoprolol is there to do.[13] Twenty-one people is twenty-one people, and the authors present it as a safety trial.

An independent systematic review published in Obesity in September 2026 places that result in context. Reviewing seven randomized controlled trials of pharmacological and lifestyle interventions in acquired hypothalamic obesity, it reports that "Tesomet was associated with reductions in both body weight and lean mass," and grades the certainty of evidence, under GRADE, as low for anthropometric and body-composition outcomes, low for metabolic and behavioural outcomes, and moderate for adverse events.[24] The lean-mass finding is not a detail to drop.

The registry records where the program went. A 48-week Tesomet study in hypothalamic injury-induced obesity enrolled 21 and completed; a Tesomet study in Prader-Willi syndrome enrolled 18 and completed in July 2019, with results posted to the registry and, as of September 7, 2026, no publication indexed in PubMed that pepmg could find.[14][15] Two larger follow-on studies, one in hypothalamic obesity and one in Prader-Willi syndrome, were registered in late 2021 and then withdrawn with zero participants randomized.[16][17] Both registry records give the same reason, and it is worth quoting because the alternative explanation is the one readers assume: "Due to financial considerations Sponsor is unable to complete the trial and assess the planned objectives/endpoints. No subjects have been randomized to treatment in the clinical trial and the decision therefore has no safety concern for patients."[16][17]

The market

What the index shows, and a 2026 paper that describes the same market from the outside

In the index generated for this note, 29 of the 104 vendors pepmg tracks carried tesofensine, across 44 listings, 40 of them recorded in stock.[27] Twenty-five of the 44 are oral forms — capsules, tablets or dissolving strips.[27] Twenty-six of the 44 record a size of 0.5 mg, the TIPO-1 dose, though the index stores some oral products by pack total rather than per-unit strength, so that figure describes the recorded field rather than a uniform product.[27]

The listings that are hardest to square with the evidence are the combinations. The index contains tesofensine sold pre-combined with other actives — a dissolving strip pairing it with 5-amino-1MQ, and capsules pairing it with taurine and magnesium L-threonate.[27] No trial in the registry or the indexed literature has tested either combination. The only tesofensine combination with published human data is the one with metoprolol, which exists to blunt a cardiovascular effect.[12][13]

An independent description of the same phenomenon arrived this month. A study in Drug Testing and Analysis, published September 2026 by an anti-doping laboratory, opens by noting that tesofensine, "although still under regulatory review, … has been marketed online as a dietary supplement promoted for weight management and metabolic enhancement."[23] The study itself is a small human pharmacology experiment: six volunteers ingested 483 µg of tesofensine bought as a dietary supplement, with urine collected for up to 600 hours.[23] It identified four principal metabolites, found marked interindividual variability with peak concentrations of 1–4 ng/mL reached anywhere between 4 and 46 hours, and reported detection windows of up to 500 hours — roughly three weeks.[23] Six volunteers, no control group, no efficacy endpoint; it is a metabolism study and pepmg reports it as one.

That paper exists because of a regulatory change that did land. WADA added tesofensine as a named example of a prohibited stimulant under section S6 of the 2025 Prohibited List, effective January 1, 2025; S6 stimulants are prohibited in-competition only.[21][22] USADA's advisory on that list pairs the addition with a warning that tesofensine has been appearing on supplement labels.[21] For a tested athlete, a three-week detection window and an in-competition ban are the operative facts, and they have nothing to do with whether the compound works.[21][23]

Four things this note is not saying

First, it is not saying the weight loss is fake. TIPO-1 was a randomized, double-blind, placebo-controlled trial, the Danish inspection specifically found that its primary and secondary endpoint data matched the source records, and an independent pooled analysis of four other randomized trials found a dose-dependent effect in patients who were not even trying to lose weight.[1][4][5] The signal is consistent across datasets.

Second, it is not saying the compound was abandoned for safety. The two withdrawn trials give financial considerations as the reason, in the registry, in writing, and add that no participant had been randomized.[16][17] A separate published human abuse-liability study — a single-dose, randomized, double-blind crossover in 52 recreational stimulant users against placebo, D-amphetamine, bupropion and atomoxetine — concluded that tesofensine's abuse potential "is no greater than that of bupropion or atomoxetine."[8]

Third, it is not saying the Mexican phase 3 does not exist or did not succeed. It is saying that pepmg could not find it in the registry or the peer-reviewed literature, that everything known about it comes from the sponsor, and that a company statement is a weaker class of evidence than a published trial no matter how favourable the number in it.[18][19][12]

Fourth, it is not saying anything about what is in a vendor's capsule. pepmg indexes prices and reports published evidence; it does not test products and makes no claim about the contents of any listing.[27]

What it is saying is narrower, and it is the whole file in one sentence: tesofensine is a non-peptide stimulant whose best evidence is an eighteen-year-old phase 2 trial in 203 people that carries an unresolved expression of concern, whose authors asked for phase 3 confirmation that has never appeared on the registry, and whose more recent published obesity evidence reviewed here includes a 21-person trial of a different, metoprolol-containing formulation.[1][2][12][13]

Questions people are asking