Emestedastat is a once-a-day pill designed to lower the stress hormone cortisol inside the brain. Its biggest Alzheimer’s trial has just reported, and the pill missed its main goal. This guide explains the result in simple words, one step at a time.
Emestedastat, also known by the brand name Xanamem, is made by Actinogen Medical, a small Australian company. On 19 November 2026 Actinogen shared the results of a trial called XanaMIA. The trial tested the pill in 247 people with Alzheimer’s disease for 36 weeks.
The main result was clear. People on the pill did a little better than people on a dummy pill, but the gap was tiny. It was 0.10 points on the main score, and a gap that small can easily happen by chance.
The company then pointed to a second finding. In people older than 75, more patients stayed stable on the pill than on the dummy pill. That sounds hopeful, and it may be real. But a result in one part of a trial needs careful reading. This guide shows why, using only the numbers the company has published.
01What is emestedastat (Xanamem)?
Emestedastat is a small tablet taken once a day. Scientists at the University of Edinburgh first designed it, and Actinogen has developed it since then. Older papers call it UE2343. Xanamem is the brand name the company uses.
The pill is built around one idea. Cortisol is the body’s main stress hormone. A short burst of cortisol is useful, because it helps us wake up and react. Too much cortisol for too long is thought to harm brain cells, mainly in the areas that handle memory. Emestedastat tries to lower cortisol inside brain cells while leaving the rest of the body alone.
It is important to say what the pill is not. No country has approved it. It cannot be bought in a pharmacy. Everything known about it comes from trials.
02How emestedastat works: the mechanism of action
The body keeps cortisol in two forms. One is active cortisol, which switches on stress signals. The other is cortisone, a resting form that does nothing until it is changed back.
Inside many cells sits an enzyme with a long name, 11β-HSD1. An enzyme is a tiny machine that changes one substance into another. This one turns resting cortisone into active cortisol. So a brain cell can raise its own cortisol level, even when the level in the blood is normal.
Emestedastat blocks that enzyme. With the enzyme blocked, less cortisol is made inside the cell. The adrenal glands, which make cortisol for the whole body, keep working as usual. That is the plan in one line: quieter cortisol in the brain, normal cortisol elsewhere.
Does the pill reach its target? A brain scan study says yes (Villemagne et al., Journal of Alzheimer’s Disease 2024, via PubMed). At 10 mg a day, the pill occupied about 80% of the enzyme in the outer brain. In the memory region the figure was about 69%. Higher doses added little.
That is a good start, but it answers only half the question. Reaching the target is one thing. Helping patients is another. A 2025 review of this enzyme found the same gap (Di Vincenzo et al., International Journal of Molecular Sciences, via PubMed). It covered emestedastat and an older drug called ABT-384. Both blocked the enzyme well, and neither had clearly improved thinking in patients.
03XanaMIA trial results: what the emestedastat study found
XanaMIA was a fair test. Half the people got the real pill and half got a dummy pill, called a placebo. A computer decided who got which. Neither patients nor doctors knew. This design is called randomised and double-blind, and it is the standard way to test a medicine.
The clinic list comes from the trial’s registry record. One detail in the entry rules is worth knowing. To join, a person needed a raised level of pTau181 in the blood. This protein rises in Alzheimer’s disease, so the test helps make sure people in the trial truly have it.
What is CDR-SB?
Every trial needs one main measure, fixed in advance. Researchers call it the primary endpoint. In XanaMIA it was a score named CDR-SB, short for Clinical Dementia Rating, Sum of Boxes.
The idea is simple. A trained clinician talks to the patient and to a family member. Six areas of life are rated, such as memory, managing at home and personal care. The six ratings are added up. The total runs from 0 to 18, and a higher score means more difficulty. People in this trial started at about 4.3.
In Alzheimer’s the score climbs over time. A medicine works on this measure if the score climbs more slowly than it does on placebo.
The main result
After 36 weeks, the pill group was 0.10 points better than the placebo group. The P value was 0.65.
A P value tells us how easily chance alone could produce a gap of this size. The usual pass mark is below 0.05. A value of 0.65 is far from that mark. In plain words, the trial could not tell the pill apart from the dummy pill. So the pill missed its primary endpoint.
Two other measures were checked, one for thinking skills and one for daily tasks. Both leaned slightly towards the pill. Neither was close to the pass mark. The thinking test had a P value of 0.59.
04Why the early signal faded
Many readers will remember good news from this trial earlier in the year. In January 2026 an independent safety board took an early look at the data. One job of such a board is to stop a trial that is clearly failing. It told Actinogen to carry on.
Now the company has shown what that early look contained. In the first 88 people, the pill was ahead by 0.58 points at week 24. The P value was 0.03, which is a pass. In Actinogen’s words, this benefit was “not fully sustained” in the final data.
Three numbers tell the story. In the early group at week 24, the gap was 0.58 points. The slides also show 0.38 points at week 36 for the early data, with a P value of 0.39. In all 247 people at week 36 it was 0.10 points.
People counted after the early look 247 − 88 = 159
Rough gap in those 159 people (247 × 0.10 − 88 × 0.38) ÷ 159 = about −0.05 points
The last line is a rough estimate. It assumes the 0.38 figure covers all 88 early participants and that the groups can be averaged simply. Even with that caution, the direction is plain. For the full trial to land at 0.10, the later participants must have shown close to nothing.
Why does this happen? Cricket gives an easy picture. A batsman who scores 90, 70 and 85 in his first three innings has an average above 80. Nobody expects that to last for 30 innings. Small samples swing widely, and early numbers are often lucky. As more data arrive, the average settles near the true value.
Trials work the same way. A result in 88 people can look strong and still shrink when 247 are counted. This is why the final, full result is the one that counts.
05Xanamem in people over 75: what the result shows
The hopeful part of the announcement is about older patients. Actinogen looked only at people above the trial’s middle age of 75. The middle value splits a group in two, so this was about half of the trial.
The company says it planned this check before seeing the data. That matters, and it is to the company’s credit. A check chosen in advance is more trustworthy than one found by searching afterwards.
In this older half, two things were measured.
First, the share of people who did not get worse. On the pill, 40% had a score that stayed the same or improved. On placebo it was 22%. The P value was 0.02, which is a pass.
Second, the average score. This is the same measure used for the whole trial. Here the pill group was ahead by 0.39 points. The P value was 0.22, which is not a pass.
So the two measures disagree. One says yes and the other says not proven. The headline figure of 30% slowing comes from the second measure, the one that did not pass. If 0.39 points is 30%, the placebo group worsened by about 1.3 points (computed).
Why a part of a trial needs extra care
Looking at one part of a trial is called a subgroup analysis. It is useful for finding ideas. It is weak as proof, for a simple reason.
Think of tossing a coin ten times in each of ten rooms. Across all rooms, heads and tails come out even. But in one room you may well see eight heads. Nothing is special about that room. Chance looks like this when you check many small groups.
A trial that misses its main goal can be split in many ways: by age, by sex, by disease stage, by blood results. Each split is another room. The slides show at least two age cuts, over 75 and over 70. They also show two ways of scoring. The more ways one looks, the more likely one of them passes by luck.
Two more facts belong here. The pill group and the placebo group were not perfect twins at the start. The pill group was two years older on average. It also had more carriers of ApoE4, a gene type that raises Alzheimer’s risk: 75.6% against 62.6%. Small gaps like these can tilt a result in either direction.
The over-75 finding may still be true. For now it is a question that needs its own test. The normal way to settle it is a new trial in people over 75, with that group as the main goal from day one.
06Earlier emestedastat trials: has this happened before?
XanaMIA is the third patient trial of this pill. Reading the results in order, with one re-analysis in between, shows a pattern.
The second row explains why XanaMIA was built the way it was. The 2024 paper found a hint in 34 people with high pTau181 (Taylor et al., Journal of Alzheimer’s Disease, via PubMed). So the new trial enrolled only people with high pTau181. In that group, the hint did not hold up.
Now the same step is being taken again. The main goal was missed, and a smaller group gives hope. This time the group is people over 75. It is fair to test that idea. It is also fair to remember that the last such idea was tested and did not hold.
A 2026 paper describes the depression trial (Taylor et al., British Journal of Psychiatry, via PubMed). The early work on the molecule is in a 2017 paper (Webster et al., British Journal of Pharmacology, via PubMed). For the reasons behind the 10 mg dose, see a 2025 review (Rolan et al., Clinical Pharmacology in Drug Development, via PubMed).
07Is emestedastat safe? Side effects so far
On safety, the news from XanaMIA is reassuring but thin. Actinogen says no serious side effect was blamed on the pill over 36 weeks. It says more than 500 people have now taken emestedastat across eight trials.
The company has not yet given counts. There is no table of side effects by group, no number of people who stopped treatment, and no count of deaths.
One earlier finding deserves a follow-up. In the depression trial, blood tests showed raised liver enzymes in 10% of people on the pill and 1% on placebo. Most rises were mild, and one person stopped treatment. That trial lasted six weeks. XanaMIA lasted 36 weeks, so its liver results will be worth reading when they come.
Actinogen also says the pill does not interfere with other medicines. That would matter for older patients, who often take several. About 60% of people in the trial were already taking a standard Alzheimer’s medicine.
08Actinogen’s money and share price
Actinogen is a small company with one medicine in testing. So this result hits it hard.
Trading in its shares was paused on 13 November while the results were prepared. The last price before the pause was 4.5 Australian cents. At that price the whole company was valued at about A$174 million. When trading resumed on 19 November, the shares fell by as much as 88% during the day, according to Stockhead. BioSpace reported a fall of more than 80%.
Company value after an 80% fall A$174M × 0.20 = about A$35M
Company value after an 88% fall A$174M × 0.12 = about A$21M
Cash held A$18.9M
When a company is priced near its cash, investors are placing almost no value on its medicine. That is the market’s blunt verdict on the trial.
Actinogen says it has A$18.9 million, no debt, and enough to last into 2028. It also expects a tax refund for research spending. To stretch the money, it has paused spending on a further large trial. Its chief executive, Dr Steven Gourlay, says the company will take the data to possible partners.
This is the practical meaning of the result. A new, larger trial would be hard to fund from A$18.9 million, which is why that spending is paused. The pill’s future now depends on whether a larger company believes the over-75 signal.
09What happens next for emestedastat
Three things are on the calendar.
The first is the full data. Actinogen says more analysis will be presented at a science meeting. The missing pieces listed above should appear then.
The second is the extension study. After the 36 weeks, people could move to a phase where everyone gets the real pill. It has 168 people. The next update is due in the first three months of 2027, when about 100 of them will have had 24 weeks of treatment.
A caution applies to extension data. Everyone knows they are on the real pill, and there is no placebo group to compare with. The company says the first 29 people showed no average worsening at 24 weeks. That is pleasant to hear, but it cannot prove the pill works.
The third is talks with partners and with the US medicines regulator, the FDA. Actinogen says these will take place over the coming months.
How does it compare with Leqembi and Kisunla?
Actinogen says the 30% slowing in older patients is similar to the figures for two approved Alzheimer’s medicines. Those are lecanemab, sold as Leqembi, and donanemab, sold as Kisunla. Its slides quote 27% for lecanemab.
These numbers look alike, but the comparison is uneven in three ways. The approved medicines showed their effect in the whole trial, which was their main goal. By contrast, the emestedastat figure comes from half of a trial that missed its main goal. Their trials ran for about 18 months, and this one ran for 36 weeks. Their results passed the statistical test, and the 0.39-point gap did not.
The pill does have real points in its favour. It is a tablet, while the two approved medicines are given by drip or injection. If a future trial proved it works, that would matter a great deal to patients.
Witfire has covered other trials where a sound idea did not carry through. Two examples are the ravulizumab ARTEMIS trial and the eplontersen heart trial.
10Emestedastat and India: what this means for families here
India has a large and growing number of people with dementia. A national study put the figure at 7.4% of people aged 60 and above, or about 8.8 million people (Lee et al., Alzheimer’s & Dementia 2023, via PubMed). Alzheimer’s disease is the most common cause of dementia.
For Indian families, three points follow from this trial.
The pill is not available here, or anywhere. None of the 34 trial clinics was in India. No regulator has approved emestedastat. Any website offering Xanamem for sale is not offering the real, tested product.
A simple tablet would suit India well, if one is ever proven. Eisai launched Leqembi in India this year after approval by the national regulator, CDSCO. Leqembi is given by drip or injection and is meant for early disease. A once-a-day tablet would be far easier to supply across the country. That is why trials like this one matter to India, even when they disappoint.
The blood test is the quiet lesson. XanaMIA used a blood test, pTau181, to choose its patients. Blood tests for Alzheimer’s are cheaper and simpler than brain scans. As they spread, they could help Indian doctors confirm the disease earlier and help Indian hospitals join future trials.
There is a lesson for Indian drug developers too. A hint from a small group is a reason to run a new trial. It is a weak base for a costly final trial. Emestedastat shows how expensive that step can be when the hint does not hold.
11Witfire Risk Score: 7.0 / 10
This score rates the risk around emestedastat as a medicine in development, after the XanaMIA result. A higher number means more risk.
12What to watch
- The full results. Scores for each group at each visit, and the result in people aged 75 and under.
- The size of the over-75 group. How many people were in it, and how many in each arm.
- Liver blood tests. Whether the rise seen in the depression trial appeared again over 36 weeks.
- First quarter of 2027. The next update from the extension study, with about 100 people at 24 weeks.
- A partner. Whether a larger company agrees to fund a trial in older patients (ASX: ACW).
- The FDA. What the US regulator says a new trial would need to show.
The plain reading is that emestedastat did not work in this trial. A gap of 0.10 points with a P value of 0.65 is a miss, and the company has said so openly.
The over-75 finding is worth testing and too weak to lean on. It passed on one measure and failed on the main one. The pill’s own history adds a warning, because the last promising small group did not hold up in a bigger trial.
For readers, the useful habit is simple. When a trial misses its main goal and a smaller group looks good, ask two things. Was that group chosen in advance? Has it been tested again? Here the answers are yes, and not yet.
FAQEmestedastat: questions and answers
What is emestedastat?
Emestedastat is a once-a-day tablet from Actinogen Medical, also known by the brand name Xanamem. It blocks an enzyme called 11β-HSD1, which makes the stress hormone cortisol inside cells. It has been tested in Alzheimer’s disease and depression. No country has approved it.
Did emestedastat work in the XanaMIA trial?
It missed its main goal. After 36 weeks, people on the pill were 0.10 points better than people on placebo on the CDR-SB score. The P value was 0.65, so the trial could not tell the pill apart from placebo.
What did the trial show in people over 75?
In people above the age of 75, 40% on the pill did not get worse, against 22% on placebo. That difference passed the statistical test, with a P value of 0.02. The average score gap in the same group was 0.39 points and did not pass, with a P value of 0.22. The finding needs a new trial to confirm it.
What is CDR-SB?
CDR-SB stands for Clinical Dementia Rating, Sum of Boxes. A clinician rates six areas of daily life, such as memory and personal care, after talking to the patient and a family member. The total runs from 0 to 18. A higher score means more difficulty.
Why was the early result better than the final result?
An early look at 88 people showed a gap of 0.58 points at week 24. With all 247 people counted at week 36, the gap was 0.10 points. Small groups swing widely by chance, so early results often shrink when more people are added.
How does emestedastat work?
Cells can turn resting cortisone into active cortisol using an enzyme called 11β-HSD1. Emestedastat blocks that enzyme, so less cortisol is made inside brain cells. A brain scan study found that 10 mg a day occupied about 80% of the enzyme in the outer brain.
Is emestedastat safe?
Actinogen says no serious side effect was blamed on the pill over 36 weeks, and that more than 500 people have taken it. Full safety tables are not yet published. In an earlier depression trial, raised liver enzymes were seen in 10% of people on the pill and 1% on placebo.
Is Xanamem approved or available in India?
No. Xanamem is not approved in India or in any other country, and none of the trial clinics was in India. It cannot be bought legally as a medicine.
How is Xanamem different from Leqembi and Kisunla?
Leqembi and Kisunla are antibody medicines that clear amyloid from the brain, and both are approved. Xanamem is a tablet that targets cortisol and is still being tested. The approved medicines met the main goal of their trials. Xanamem did not.
What happens next for Xanamem?
Actinogen has paused spending on a further large trial. It plans to discuss the data with possible partners and with the US regulator. An extension study in 168 people continues, with an update due in the first quarter of 2027.
Author. Dr. Akhilesh Vats is Editor-in-Chief of Witfire Elite and founder of ACME Research Solutions, a pharmaceutical contract research organisation.
Disclosure: Editorial analysis, not investment or medical advice. The Witfire Risk Score is our own framework, weighting Regulatory Exposure (30%), Competitive Displacement (25%), Capital Position (20%), Evidence Integrity (15%) and Execution & Credibility (10%) on a 1–10 scale. Scores are judgments, not forecasts.
Put this framework on your own file.
The same method used in these briefs, applied privately to a decision you are actually facing — a target, a partner, a facility, a filing.
- CMC & manufacturing risk assessment
- Regulatory / CRL readiness review
- Due diligence on a target or partner
- Investor brief or market analysis
- Commissioned editorial
- Something else
The signal before the market catches up.
A low-noise briefing on the pharma events that change strategy, risk and value. No more than one a week.
