234 hours, about ten days, modelled by Lehr and colleagues from 320 Alzheimer patients on repeated swallowed doses. That is unusually long for a daily pill, and the practical consequence is that taking it every day keeps building the level in the body for roughly six to eight weeks, so judging a dose at two weeks is reading it before the drug has finished arriving.
What it actually is
Tesofensine is a pill you swallow, not an injection, which already makes it odd company here. It is a small manufactured chemical rather than a peptide, and its core structure belongs to the same scaffold family as cocaine. It was built as a brain drug, first for Parkinson disease and then for Alzheimer disease, and it failed at both. It got a second life only because the people in those trials kept losing weight. In the one proper obesity trial, the group on the highest dose lost about a tenth of their body weight in half a year, against two percent for the group on a dummy pill. Everyone in that trial was also on an energy restricted diet, so I could not find a published result for the drug on its own.
What it is supposed to do
It blocks reuptake of three of the brain's own signalling chemicals at once: noradrenaline, serotonin and dopamine. Reuptake is the vacuuming up a nerve cell normally does after it releases one of these, so blocking it leaves more of each sitting in the gap between one nerve cell and the next. What that has to do with body weight is not something the entry can answer. The weight effect turned up afterwards, when somebody pooled the weight data off trials run for something else. At what it was actually given for, it did not beat a dummy pill: at week 14 none of the three doses separated from placebo on the standard score clinicians use for Parkinson symptoms.
What people take it for
People take it to lose weight. That is what it is sold and talked about for, and the 10.6 percent figure that every vendor page inherits came from the 1.0 mg arm of the one published obesity trial, a dose that was dropped from the development programme. It is not approved for that or for anything else in any country I could find.
The dose question
What circulates
0.25 mg to 0.5 mg by mouth, once daily
What was studied
0.25, 0.5 and 1.0 mg by mouth, once daily
Astrup 2008, TIPO-1, 203 obese adults over 24 weeks
Lands on a studied dose
The convention is the trial dose, which is rare here. What does not travel with it is the rest of the trial: every obesity trial paired the drug with an energy restricted diet, so I could not find a published result for the drug alone, and they all ran 24 weeks and stopped. The 1.0 mg arm behind the 10.6 percent was dropped at the FDA meeting before the large trial meant to settle the heart rate question, and that trial was never run. The paper the numbers come from carries a Lancet Expression of Concern from 2013 about unrecorded adverse events, and I found it still standing.
Reported because it is what people use. Nothing here recommends any amount.
How long it lasts
234 hours, about ten days, modelled by Lehr and colleagues from 320 Alzheimer patients on repeated swallowed doses. That is unusually long for a daily pill, and the practical consequence is that taking it every day keeps building the level in the body for roughly six to eight weeks, so judging a dose at two weeks is reading it before the drug has finished arriving.
Route studied
Swallowed once a day. Fourteen weeks in the Parkinson trial, 24 weeks in the obesity trials.
Route used
Swallowed once a day, from vendor capsules and tablets of 250 and 500 micrograms.
How to check you have the right molecule
Two checks. On the chemistry, hold a certificate against the entry's numbers: PubChem identifier 11370864, formula C17H23Cl2NO, molecular weight 328.3. On the name, watch for two products that are not tesofensine on its own. Tesomet is a fixed combination of 0.5 mg tesofensine with 50 mg metoprolol, a beta blocker put there specifically to cancel the heart effect, so the clean blood pressure result people cite from it belongs to two drugs rather than to the pill by itself. Nupenta, which gets listed alongside it in the sentence claiming Mexican approval, is a pantoprazole brand from Macleods Pharmaceuticals in India, a stomach acid drug and an entirely different molecule. I found no independent analysis of a grey market tesofensine product, so I could not find a published test of a vendor capsule to compare yours against.
Two mistakes, and they stack. The first is the approval. A sentence mirrored across vendor pages, uncited, says it has been approved in Mexico as Tesomet and Nupenta for obesity since 2023. What actually happened in February 2023 is that the Mexican regulator's new molecules committee issued a favourable opinion with qualifications, which the sponsor's own description calls a non binding technical opinion that does not represent a market authorization nor a rejection. Mexican outlets ran it under headlines using the verb aprueba, which is where the confusion starts. In November 2024 the sponsor announced that the filing had not been approved, and in February 2025 announced a resubmission. As of 5 August 2026 I could not find a marketing authorisation in any country. The second is treating the clean cardiovascular numbers as belonging to tesofensine. The modern trial with no significant difference in heart rate or blood pressure, 21 adults with hypothalamic obesity over 24 weeks, tested tesofensine given together with a beta blocker whose job is to cancel exactly that effect. That flat result belongs to the combination, not to the pill on its own.
What is known about harm
The heart is the open question, and it rises with the dose. Pooled off the neurology trials, heart rate ran up 2.1, 4.2, 6.0 and 6.8 beats per minute across ascending doses through 1.0 mg, against minus 0.4 on the dummy pill, with blood pressure unmoved. The obesity trial conceded a rise of 7.4 beats per minute at its middle 0.5 mg dose, cleared 0.25 and 0.5 mg on blood pressure, and said nothing at all about the 1.0 mg arm, the one that produced the largest weight loss. Bello and Zahner's 2009 review of the programme fills that hole and reports dose dependent heart rate rises with significant blood pressure increases at the highest dose tested. Secondary sources put the 1.0 mg arm at 6.8 over 5.8 mm Hg and 8.5 beats per minute, but the full text of the trial paper returned an access error to me, so treat those two figures as unverified against the paper. Two other things belong here. The trial built to settle the heart question, NCT00481104, 140 obese patients with vital signs as its main measure, has no results posted. And the closest marketed comparator, sibutramine, same class of cardiovascular signal, was withdrawn in 2010 after its outcomes trial. Side effect reporting from the pivotal trial is itself in question: after a Danish Health and Medicines Authority inspection, the journal attached an Expression of Concern in 2013 covering consent delegated to non medical staff at one site, blinding integrity questioned in an earlier inspection, and monitors instructing investigators not to record headache, migraine, stress and depression as adverse events where a patient had reported those conditions before the trial started. The investigators' reply was titled Under-reporting of adverse effects of tesofensine.