Tesofensine produced the best weight-loss number of its era, and never became a medicine

What it does inside the brain

Your brain passes messages between nerve cells using chemicals called neurotransmitters. A nerve cell releases a chemical into the tiny gap between itself and the next cell, the next cell reads it, and then the first cell pulls the chemical back in to be used again. That pulling-back is called reuptake, and the molecular machinery that does it is called a transporter. Block the transporter and the chemical stays in the gap longer, so the message keeps being read.

Tesofensine blocks three transporters at once: the ones for noradrenaline, dopamine and serotonin. That is what "triple monoamine reuptake inhibitor" means, and it is the whole mechanism. The three chemicals it acts on are the three most closely tied to appetite, alertness and reward. Serotonin is involved in feeling full. Noradrenaline drives arousal and raises the rate at which the body burns energy. Dopamine carries the signal that something is worth wanting, which is why it matters for food specifically and not just for hunger.

Most appetite drugs push on one of these. Tesofensine pushes on all three. The published trial describes it exactly this way — "an inhibitor of the presynaptic uptake of noradrenaline, dopamine, and serotonin" — and that breadth is the reason the effect was large, and also the reason the side effects were what they were. You cannot raise all three of these chemicals across the brain and only change eating.

What actually happened in the trial

The trial that made tesofensine's reputation ran in five Danish obesity clinics and published in The Lancet in November 2008. It was a phase II study: 203 people with a body mass index between 30 and 40, all put on a reduced-calorie diet, then randomly assigned to one of three doses of tesofensine or to a placebo, once daily for 24 weeks. Neither the patients nor the doctors knew who was getting what. 161 people, or 79 percent, finished.

The results, in the paper's own words:

"After 24 weeks, the mean weight loss produced by diet and placebo was 2.0% (SE 0.60). Tesofensine 0.25 mg, 0.5 mg, and 1.0 mg and diet induced a mean weight loss of 4.5% (0.87), 9.2% (0.91), and 10.6% (0.84), respectively, greater than diet and placebo (p<0.0001)."

Read those numbers carefully, because they are routinely misquoted. The 9.2 percent at the 0.5 mg dose is weight loss greater than the placebo group, not total weight loss. Placebo plus diet took off 2.0 percent. So the 0.5 mg group came in around 11 percent below where they started, and the drug's own contribution was the 9.2.

For 2008 that was a very large number. The paper opens by saying why it mattered: "Weight-loss drugs produce an additional mean weight loss of only 3-5 kg above that of diet and placebo over 6 months, and more effective pharmacotherapy of obesity is needed." Against a 3–5 kg benchmark, tesofensine at 0.5 mg roughly doubled it. The authors said so, and hedged it in the same sentence: "Our results suggest that tesofensine 0.5 mg might have the potential to produce a weight loss twice that of currently approved drugs. However, these findings of efficacy and safety need confirmation in phase III trials."

That confirmation never arrived.

The part that is usually left out

In April 2013, The Lancet published an expression of concern about that trial.

An expression of concern is a formal notice from a journal that something about a published paper is under question and readers should treat it with caution. It is not a retraction — the paper stands — but it is not nothing either, and it is attached to the record permanently. The notice is indexed in PubMed under its own identifier, PMID 23561987, and its title is simply "Expression of concern--effect of tesofensine on bodyweight loss, body composition, and quality of life in obese patients: a randomised, double-blind, placebo-controlled trial." The PubMed record for the 2008 paper carries the link both ways.

This matters more for tesofensine than it would for most compounds, because the 2008 trial is not one piece of evidence among many. It is essentially the whole human efficacy case. The 9.2 percent figure that appears on every page selling this compound comes from that single 203-person phase II study, and that study has a journal-issued caution attached to it. A page that quotes the number and omits the notice has told you the good half of the record.

What a second, smaller study showed about the mechanism

A separate trial, published in the International Journal of Obesity in 2010, went looking for how the weight came off. Thirty-two overweight and moderately obese men took 2.0 mg daily for a week, then 1.0 mg daily for a second week, or placebo, while being told to keep their eating and activity the same. They were measured inside a respiration chamber, which is a sealed room that can account for every calorie a body burns.

Two things came out of it. Appetite changed a great deal: the paper reports that tesofensine "induced higher ratings of satiety and fullness and concomitantly lower prospective food intake than placebo." Energy expenditure barely changed: "No significant effect of TE on total 24-h EE could be demonstrated compared with PL, but higher energy expenditure was observed during the night period (4.6%; P<0.05) when adjusted for changes in body composition." Fat burning did shift, by 18 grams over 24 hours.

The honest reading is that tesofensine works mostly by making people want to eat less, not by meaningfully raising the rate at which they burn energy. That is worth knowing, because "boosts metabolism" is the claim most often made for it, and the one study designed to measure exactly that found the effect small and confined to the night.

Side effects, and the one that ended it

The 2008 trial lists them plainly: "The most common adverse events caused by tesofensine were dry mouth, nausea, constipation, hard stools, diarrhoea, and insomnia." Dry mouth and insomnia are the signature of a drug raising noradrenaline. They are not incidental; they are the mechanism showing up where it was not wanted.

The cardiovascular numbers are where the programme ran into trouble. At 0.25 mg and 0.5 mg there was "no significant increases in systolic or diastolic blood pressure compared with placebo," but heart rate was another matter: "heart rate was increased by 7.4 beats per min in the tesofensine 0.5 mg group (p=0.0001)." At the 1.0 mg dose, blood pressure rose as well.

Seven and a half extra beats a minute, every minute, in a population that already carries cardiovascular risk from obesity, is the kind of finding a regulator weighs against the benefit. The history of appetite drugs is largely a history of exactly this trade — sibutramine, which also raised heart rate and blood pressure through noradrenaline and serotonin reuptake inhibition, was withdrawn from the American and European markets in 2010 after a long cardiovascular outcomes trial found more heart attacks and strokes in the drug group. Tesofensine belongs to the same pharmacological family, and it was being developed into a regulatory climate shaped by that withdrawal.

How it moves through the body

Tesofensine has an unusually long half-life for a small molecule of its type — on the order of 200 to 230 hours, which is eight to ten days. Half-life is the time it takes for the amount in your blood to fall by half. Ten days is very long. Fluoxetine, famous for lingering, is one to four days for the parent drug.

Two consequences follow, and both are practical rather than theoretical.

The first is that the drug accumulates for weeks. With a ten-day half-life it takes roughly five half-lives — about six or seven weeks — before the amount in your blood stops climbing and settles at a steady level. The dose you take on day one and the dose you take on day fifty are the same dose, but the amount circulating in your body on day fifty is several times higher. This is why side effects on a compound like this often appear well after starting, and why they appear without anything having changed. Nothing changed. The accumulation caught up.

The second is that stopping does not stop it. If you take the last dose today, meaningful amounts are still present a month from now. That cuts both ways: there is no withdrawal cliff, but there is also no fast exit if something goes wrong. If your heart rate has climbed and you want it back down, you wait weeks, not hours. Anything that pushes on noradrenaline — a stimulant, a decongestant, a strong dose of caffeine — is stacking onto a drug that has not left, and will not for some time.

It is taken by mouth and it is a small molecule, not a peptide, which is worth stating because it is sold alongside peptides and constantly described as one. Peptides are short chains of amino acids and mostly have to be injected because the gut digests them. Tesofensine is neither a peptide nor injected. If a supplier lists it as a peptide, they are describing their catalogue rather than the compound.

Where it stands now

Tesofensine is not an approved medicine in the United States, the United Kingdom or the European Union. It has no FDA approval for any indication. The compound was later licensed to Saniona, which has pursued it for hypothalamic obesity — a specific, rare condition in which damage to the hypothalamus, usually from a brain tumour or its treatment, destroys the body's ability to regulate appetite at all. That is a different and much narrower target than general obesity, and pursuing it is a reasonable read of what the compound can support.

Meanwhile the landscape it was competing in changed completely. The GLP-1 drugs — semaglutide, tirzepatide — now produce 15 to 22 percent total body weight loss in phase III trials with tens of thousands of participants and cardiovascular outcome data. A phase II result of 9.2 percent above placebo in 203 people, with an expression of concern attached, is no longer a competitive position. That is the honest reason tesofensine is a research compound in 2026 and not a prescription.

What that means if you are looking at it

Everything below is what the record does and does not support. None of it is medical advice, and tesofensine is sold for research use only, not for human consumption.

The dosing you will see quoted — 0.25 mg to 1.0 mg once daily — comes from the 2008 trial and nowhere else. There is no long-term human safety data at any dose. The longest controlled human exposure in the published literature is 24 weeks.

Three things are genuinely unknown, and no amount of reading will resolve them, because the studies were not done. Nobody knows what happens after 24 weeks. Nobody knows whether the heart rate increase translates into cardiovascular events over years, which is the exact question that ended sibutramine and required a 10,000-patient trial to answer. And nobody knows whether the 2008 numbers would replicate, which is what phase III exists to establish and what the expression of concern makes a live question rather than a formality.

What is known is narrower and firmer. It reduces appetite, substantially, in people who take it. It raises resting heart rate at doses that produce meaningful weight loss. It causes dry mouth and insomnia often enough that both appear in the trial's own list. It works through three transporters simultaneously, which is why the appetite effect is large and why the side effects are systemic rather than local. And its single supporting trial carries a journal-issued caution that is part of the record whether or not the page you read it on mentions it.

Anyone taking a compound that raises heart rate by seven beats a minute should know their own resting heart rate and blood pressure before and during, and should know that the drug most similar to this one was pulled from two major markets over cardiovascular outcomes. That is not a scare; it is the specific, documented reason this compound is where it is.

The interactions that are not optional to know about

Three of these are serious enough that they are worth stating on their own, separately from the general side effect list, because they are the ones with a named mechanism rather than a general caution.

Anything else that raises serotonin. Tesofensine blocks serotonin reuptake. So do the SSRI antidepressants — sertraline, fluoxetine, escitalopram and the rest — and so do the SNRIs, tramadol, triptans for migraine, St John's wort, and MDMA. Stacking two serotonin-raising drugs risks serotonin syndrome, which is not a subtle condition: agitation, a racing heart, high temperature, muscle rigidity, tremor, and in severe cases it is fatal. The ten-day half-life makes this worse rather than better, because "I stopped taking it" does not mean it is gone.

MAO inhibitors. These are an older class of antidepressant, and also include the Parkinson's drugs selegiline and rasagiline and the antibiotic linezolid. Combining an MAO inhibitor with a triple reuptake inhibitor is the textbook recipe for both serotonin syndrome and a hypertensive crisis, because one drug stops the breakdown of these chemicals while the other stops their removal from the synapse. This combination is contraindicated with every drug in tesofensine's family, and the required washout is measured in weeks in both directions.

Stimulants and anything that raises blood pressure. Amphetamines, methylphenidate, cocaine, high-dose caffeine, pseudoephedrine and phenylephrine in cold medicine, and most pre-workout formulas all push noradrenaline in the same direction tesofensine already is. The 2008 trial found 7.4 extra beats per minute at 0.5 mg from tesofensine alone. Adding a stimulant on top of that is adding to a number that was already the programme's main safety concern.

There is a fourth worth naming because it is easy to miss: anything that acts on dopamine. Tesofensine came out of Parkinson's research precisely because it raises dopamine, and dopamine-raising drugs can trigger or worsen psychosis in people predisposed to it, and can drive compulsive behaviour — gambling, shopping, eating — which is a well-documented effect of dopamine agonists in Parkinson's patients.

What a page selling this will not tell you

Five things, all checkable, all missing from most of what is written about this compound.

The 9.2 percent is above placebo, not total. Placebo plus diet was 2.0 percent. The number quoted as the headline result is the difference, and the difference is the correct thing to quote — but it is quoted as if it were the total, which makes the drug sound better than the trial found.

There is one efficacy trial. Not a body of literature — one 203-person phase II study, published in 2008, plus a 32-person mechanism study in 2010 that was not designed to measure weight loss. Everything else is commentary on those two.

That trial has an expression of concern attached to it. Lancet 2013;381(9873):1167, PMID 23561987, linked from the PubMed record of the original in both directions.

Phase III never happened. The 2008 authors said in the paper that phase III confirmation was needed. There is no published phase III obesity trial. A drug that stops at phase II has not been shown to work in the sense the word usually means.

It failed in the two diseases it was built for. Parkinson's and Alzheimer's, under the name NS2330. The weight loss was originally a side effect, noticed in trials that were looking for something else and did not find it.

None of that makes it an uninteresting compound. It makes it a phase II compound with a real, measured appetite effect, a real, measured cardiovascular signal, a single supporting trial under formal caution, and a mechanism whose closest approved relative was withdrawn on cardiovascular grounds. That is a specific position, and it is a more useful thing to know than a percentage.