Mitoglitazone dosage and evidence
Written by Aaron CuhaReviewed Sep 2026
The short answer
In people, it was tested in 356 people with type 2 diabetes and 29 people with Alzheimer disease. The company ran both studies, and no large final-stage study followed. In male mice started at 7 months, typical lifespan rose 8.7 percent. A retest started at 18 months failed, and female mice died sooner.
Evidence grade for Mitoglitazone: Moderate. The rule: at least one finished study in people, and at least 100 people who took it. The sum on this page: 2 completed trials, 385 people given it. A grade counts how much has been done to people, not whether it works.
- People given it
- 385
- Human studies
- 2
- Found nothing
- 1
Dose
No human dose established
Amounts that were given in studies sit in Human evidence on this page. None of them is a personal dose.
2 completed trials, 385 people given it
1 effects reported, 1 unwanted, by people using it, not a trial
Goal pages: Fat loss and metabolism, Ageing and longevity
In plain English
In plain English, from Aaron
This is how I would explain it to a friend. The evidence, with the numbers, is further down the page.
Mitoglitazone is a test diabetes pill. It is a cousin of pioglitazone, built to cause less water retention. It is not approved anywhere. The company stopped testing it in 2013.
What people take it for
- Better blood sugar.
- A longer life, after a mouse study.
- Parkinson disease, which it has never been tested for in people.
What studies in people showed
In 356 people with type 2 diabetes, it cut blood sugar as much as pioglitazone over 12 weeks. Signs of water retention were about half as big. In 29 people with mild Alzheimer disease, it missed its main brain scan goal. The maker ran both, and no bigger study came after. There is no Parkinson study in people. Now mice. In males started at 7 months, the typical lifespan rose 8.7 percent. In a retest started at 18 months, males gained nothing and females died sooner.
What people report
Reports, not study results
The good
- We have not collected user reports on this one yet.
The bad
- Blood signs of water retention in the diabetes study, about half as big as on pioglitazone.
Where these come from: The water retention note is from the diabetes study on this page. We have no user reports.
My bottom line
The company that owned it walked away. The mouse retest hurt females. I would not touch it.
An opinion, not a finding. I am a coach, not a doctor.
What it has been measured to do
Measured in people
Goals Mitoglitazone has been tested for in people, biggest studies first. We count only what the studies on this page say, so the real numbers may be higher.
- Blood sugar and insulin712 people · 2 studies
Animal and lab work, not yet tested in people
Real results, but from animals or cells, not people. What works in a mouse often does not work the same way in a person. Nobody can patent this, so nobody would ever earn back the cost of a study in people. That is usually why there is none. The record on this one is set out below.
- Living longer1 animal or lab study
34 of the 36 indexed goals have no study of any kind behind Mitoglitazone
No study in people, animals or cells on this page reports a result for these. That does not prove Mitoglitazone does nothing for them. It does mean anything you have heard about Mitoglitazone and one of these did not come from a study cited here.
Show full evidence
What is Mitoglitazone?
In people, it was tested in 356 people with type 2 diabetes and 29 people with Alzheimer disease. The company ran both studies, and no large final-stage study followed. In male mice started at 7 months, typical lifespan rose 8.7 percent. A retest started at 18 months failed, and female mice died sooner.
Mitoglitazone is a diabetes drug from the same family as pioglitazone. It was built to keep pioglitazone's effect on insulin while avoiding its main downsides, water retention and weight gain. Those downsides are blamed on a switch inside cells called PPAR-gamma, and mitoglitazone barely touches it. Its maker, Metabolic Solutions Development Company, traced its main action to a target inside mitochondria, the cell's power plants. That target turned out to be the carrier that moves pyruvate, a sugar breakdown product, into mitochondria.
In people it reached mid-stage trials twice. In a 12 week diabetes study, 356 people with type 2 diabetes were split by chance. They got 50, 100 or 150 mg of mitoglitazone, 45 mg of pioglitazone, or a dummy pill. 258 finished. The two higher doses lowered fasting blood sugar and HbA1c (a three month blood sugar average) as much as pioglitazone did. Blood signs of water retention were about half as big. A fat tissue hormone called adiponectin rose less, which the authors read as less growth of fat tissue. In a 12 week study of 29 people with mild Alzheimer disease and no diabetes, 16 took 150 mg a day and 13 took a dummy pill. The main brain scan comparison of how the brain uses sugar found no difference. A second way of reading the same scans showed brain sugar use holding steady on the drug and falling on the dummy pill. The authors called that a side look. The maker ran both trials. No large final-stage trial followed either one. No Parkinson disease trial of this drug (MSDC-0160) is registered on ClinicalTrials.gov. The Parkinson work is in mice, worms and nerve cells in a dish.
In mice, the NIA's lifespan program fed mitoglitazone from 7 months of age in its 2021 run. Typical (median) male lifespan rose from 794 to 863 days, 8.7 percent. Survival to old age did not clearly change (up 2.9 percent), and treated males were heavier at 12, 18 and 24 months. Females dipped a little, not clearly (median down 2.3 percent). In the 2022 run, the program retested it from 18 months. The target dose in the food was 300 ppm, but the food only reached 126 ppm. Males got no clear benefit (median up 7.8 percent, could be chance). Pooled across all three labs, females on the late start lived shorter lives: median down 5.8 percent and survival to old age down 7.8 percent. At one lab, the untreated females lived unusually long. So the authors re-ran the numbers with only the other two labs. The harm to females still held (median down 0.7 percent, survival to old age down 6.45 percent). The same happened with pioglitazone.
Mitoglitazone is an experimental drug. It is not approved anywhere.
How does Mitoglitazone work?
Mitoglitazone sticks to the pyruvate carrier, a two-part gate (MPC1 and MPC2) that lets pyruvate into mitochondria. A lab method that chemically tags what a drug touches found this gate is where drugs in this family bind inside mitochondria. It barely binds PPAR-gamma. Slowing pyruvate's way into mitochondria changes which fuel the cell burns. In the maker's model, that makes the body more sensitive to insulin without the water retention and fat growth PPAR-gamma causes. In the human diabetes trial, blood sugar results looked like pioglitazone's, with about half the blood signs of water retention. That fits the model. In mice, in worms and in human nerve cells in a dish, all used as models of Parkinson disease, the drug turned down a growth signal called mTOR. It helped cells clear out their own junk again, protected the nerve cells that make dopamine, and calmed support cells in the brain. That is animal and dish data. Nobody knows why it made male mice started at 7 months live longer, but made female mice started at 18 months die sooner. The mouse program's authors note that pioglitazone, its parent drug, did the same harm to females, and that the cause is not clear.
What have studies in people found about Mitoglitazone?
Two finished mid-stage trials, both run by the maker, both with results posted. One was 356 people with type 2 diabetes for 12 weeks. The other was 29 people with mild Alzheimer disease for 12 weeks. No large final-stage trial, no trial longer than 12 weeks, and no trial of any kind in Parkinson disease, even though one is often described.
- Type 2 diabetes, 356 people split by chance, 258 finished 12 weeks: at 100 and 150 mg, fasting blood sugar and HbA1c fell as much as on pioglitazone 45 mg. Blood signs of water retention were about half those on pioglitazone. The maker paid for it.
- An earlier diabetes trial in 86 people with type 2 diabetes (90 and 220 mg, against pioglitazone and a dummy pill) is registered with results posted (NCT00760578). The main thing it measured was blood sugar after a mixed meal.
- Alzheimer disease, 29 people split by chance: the main brain scan measure showed no difference. A side look at the same scans favoured the drug. The maker paid for it. No benefit to memory or thinking was claimed.
- Parkinson disease: no human trial of MSDC-0160 is registered on ClinicalTrials.gov. The Parkinson evidence is in mice, worms and nerve cells in a dish.
- Where it stands: no trial registered since the Alzheimer study finished in 2013. When a mid-stage trial is not followed by a final-stage one for a decade, that says something about what the maker thought of the results, as well as about the drug.
What this does not tell you: Twelve weeks is long enough to see blood sugar and scan changes. It is far too short to see anything about ageing, heart attacks and strokes, or cancer. Those are the areas where this drug family has had its safety problems. The Alzheimer trial was a 29 person test run, and it missed its main measure. Nothing in a person speaks to lifespan. The mouse program's second run found harm to females when the drug was started late.
The state of the research
The National Institute on Aging runs a mouse lifespan program called the Interventions Testing Program. It tests each compound at three labs at the same time: The Jackson Laboratory, the University of Michigan and UT Health San Antonio. The mice are a mixed breed, so a result is less likely to be a quirk of one strain. Males and females are both tested and judged separately. The three labs' results are then pooled. Every result gets published, whether or not the compound worked. It runs on public money from the National Institute on Aging, with no drug company money. Mitoglitazone is the case where the mouse program's habit of retesting did its job. Started at 7 months in the 2021 run, it raised typical (median) male lifespan 8.7 percent. It did not move survival to old age. That already marked it as something that stopped earlier deaths, not something that stretched the oldest lives. Started at 18 months in the 2022 run, the food only reached 126 of the target 300 ppm. It did nothing for males. It shortened female lifespan, by 5.8 percent at the median across all three labs. The harm still held when the lab with unusually long-lived untreated females was left out. Pioglitazone, its parent drug, did the same harm to females in the same run. The honest reading: one good male result at one dose and starting age. One failed retest, with harm to females. And a dose shortfall in the retest, so nobody knows what a full late-start dose would have done.
The human work stopped in 2013, and the company has not said why. Both mid-stage trials posted results, which is to its credit. Neither was followed by a large final-stage trial. A private company holding the patent had every money reason to carry on if the data were good. The Parkinson disease story gets repeated a lot. It rests entirely on a 2016 paper in animals and cells. No trial in patients exists. A page that says it is "in phase 2 for Parkinson disease" is wrong.
The evidence, in charts
Fig. 1a · how many people
770people, across 4 studies in people cited here
| Study | People | Share |
|---|---|---|
| Clinical Pharmacology and Therapeutics 2013 | 356 | 46% |
| ClinicalTrials.gov 2013 | 356 | 46% |
| Current Alzheimer Research 2014 | 29 | 4% |
| ClinicalTrials.gov 2014 | 29 | 4% |
Counts are the number of people each study on this page says it signed up. That includes studies that only watched people and gave nobody the compound. Studies still signing people up are left out, and only studies cited on this page are counted.
Fig. 1b · evidence mix
4 of 8 sources here are studies in people. The rest are not.
- Study in people · done in people450%
- Animal study · done in animals338%
- Lab (cells in a dish) · cells or tissue113%
Counted from the list of sources on this page. It counts only what we cite, so there may be more studies out there. A review counts once, however many studies it sums up.
Fig. 2 · evidence over time
The studies cover 4 different years, 2013 to 2026, counted from the list of sources on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Not approved
AU
Not approved
CA
Not approved
Not approved in any of the four countries shown. This is what the law says, not a verdict on the science. See Legal status below for the full text and any notes.
Fig. 4 · how much does what
Nobody has tested different amounts in people
This chart would show how the effect changes as the amount goes up. For this compound, nobody has measured that in people. So we leave it empty. We will not borrow an animal chart and pretend it works the same in people.
Fig. 5 · what the molecule is
- Kind of compound
- Small molecule
- Molecule weight
- Not on file
- Half-life
- None on file
- Chain length
- None on file
No amino acid sequence is on file for Mitoglitazone, which is expected: a small molecule is not built from amino acids.
The studies we cite
- Study in people2013Clinical proof-of-concept study with MSDC-0160, a prototype mTOT-modulating insulin sensitizer
It lowered blood sugar about as well as the older drug it was meant to improve on, with less water retention. 356 people with type 2 diabetes were split by chance between three doses of mitoglitazone, pioglitazone, or a dummy pill, for twelve weeks, and 258 finished. Fasting blood sugar fell about the same on both real drugs. The blood changes that signal water retention were about half the size. Metabolic Solutions Development Company paid for it.
Phase 2b, 12 weeks, 258 completers of 356 randomised to 50, 100 or 150 mg mitoglitazone, pioglitazone 45 mg or placebo. Fasting glucose fell equally on the two active drugs; HbA1c reductions at 100 and 150 mg were not different from pioglitazone; haematocrit, red cell and haemoglobin changes indicating fluid retention were 50 percent smaller; HMW adiponectin rose less (p < 0.0001). Sponsored by Metabolic Solutions Development Company.
- Study in people2014An evaluation of MSDC-0160, a prototype mTOT modulating insulin sensitizer, in patients with mild Alzheimer's disease
The Alzheimer study missed what it set out to measure. 29 people with mild Alzheimer disease and no diabetes were split by chance, 16 onto the drug and 13 onto a dummy pill, for twelve weeks. The main brain scan comparison showed no difference between the groups. A second way of reading the same scans suggested brain sugar use held steady on the drug and fell on the dummy pill. That is a side look at a failed main result, in 29 people. The maker ran it.
29 non-diabetic people with mild Alzheimer disease randomised to 150 mg daily (n = 16) or placebo (n = 13) for 12 weeks. Primary FDG-PET analysis referenced to pons or whole brain: no significant difference. Exploratory analysis referenced to cerebellum: metabolism maintained on drug and declined on placebo in cingulate, parietal and temporal regions. Sponsor-run.
- Study in people2013Phase 2B dose-ranging study of mitoglitazone in type 2 diabetes
This is the registry entry for the diabetes trial above. 356 people enrolled, it finished in December 2011, and results were posted in March 2013. The main thing it was built to measure was the change in fasting blood sugar at twelve weeks. The published paper describes what came of it. The company behind the drug, Metabolic Solutions Development Company, ran it.
Registry record: 356 enrolled, completed December 2011, results posted March 2013, sponsor Metabolic Solutions Development Company. Primary endpoint change in fasting plasma glucose at week 12.
- Study in people20143-month study of MSDC-0160 on brain glucose utilization, cognition and safety in Alzheimer disease
The registry entry for the Alzheimer study, and also the end of the road. 29 people enrolled, it completed in May 2013 and results were posted in November 2014. Nothing followed it. No further Alzheimer trial of this drug is registered, and despite the talk about Parkinson disease, no Parkinson trial of it is registered either.
Registry record: 29 enrolled, completed May 2013, results posted November 2014. No follow-on trial in Alzheimer disease is registered. No MSDC-0160 trial in Parkinson disease is registered.
- Animal study2026Extension of lifespan by epicatechin, halofuginone and mitoglitazone in male but not female genetically heterogeneous mice
In mice. In the 2021 run of the NIA mouse lifespan program, mitoglitazone was fed from 7 months of age. The typical male lived 863 days against 794, an 8.7 percent gain. Survival to old age rose 2.9 percent, which could be chance. There were 149 treated males and 298 untreated ones. Treated males were heavier at 12, 18 and 24 months. Females dipped 2.3 percent at the median, not clearly. Public NIA money paid for it.
Interventions Testing Program 2021 cohort, treatment from 7 months. Male median lifespan 863 versus 794 days (+8.7 percent, log-rank p = 0.015); 90th percentile survival +2.9 percent, not significant (p = 0.19); 149 treated males against 298 controls. Treated males were heavier at 12, 18 and 24 months. Females: median -2.3 percent, p = 0.07. NIA funded.
- Animal study2026Astaxanthin, meclizine, mitoglitazone, pioglitazone, alpha-ketoglutarate, mifepristone, methotrexate, and atorvastatin-telmisartan do not increase lifespan in UM-HET3 mice
In mice. The 2022 run retested mitoglitazone from 18 months of age. The food reached only 126 ppm of a planned 300 ppm. Males gained nothing clear. Across all three labs, females died sooner: median down 5.8 percent, survival to old age down 7.8 percent. With the lab where untreated females lived unusually long left out, the harm still held. Pioglitazone did the same harm to females. Public NIA money paid for it.
2022 cohort retest of mitoglitazone started at 18 months (target 300 ppm, achieved 126 ppm). Males: median +7.8 percent, p = 0.18, not significant. Females pooled across three sites: median -5.8 percent (p = 0.001), 90th percentile -7.8 percent (p = 0.006); with the site whose control females lived unusually long removed, median -0.7 percent (p = 0.037) and 90th percentile -6.45 percent (p = 0.024). Pioglitazone showed the same female harm. NIA funded.
- Animal study2016Mitochondrial pyruvate carrier regulates autophagy, inflammation, and neurodegeneration in experimental models of Parkinson's disease
In mice, worms and nerve cells in a dish. The drug protected the nerve cells that make dopamine in mice given a toxin that mimics Parkinson disease. It did the same in a worm model and in human brain cells in a dish. It turned down a growth signal called mTOR and helped cells clear their own junk. This paper is where the Parkinson interest comes from. No Parkinson trial in people has followed.
MSDC-0160 protected dopaminergic neurons in MPTP-treated mice, a C. elegans alpha-synuclein model and cultured human midbrain neurons, reduced mTOR signalling and normalised autophagy. Animal and cell data; the source of the Parkinson interest. No human Parkinson trial has followed.
- Lab (cells in a dish)2013Identification of a mitochondrial target of thiazolidinedione insulin sensitizers (mTOT): relationship to newly identified mitochondrial pyruvate carrier proteins
In a dish and in fruit flies. The maker's own lab found where this drug family binds inside mitochondria: a two-part gate, MPC1 and MPC2, that lets pyruvate in. In fruit flies bred with less of that gate, the drug stopped working and their blood sugar rose.
Photo-crosslinking proteomics identified MPC1 and MPC2 as the mitochondrial complex that binds thiazolidinediones; knockdown in Drosophila blocked drug action and raised haemolymph glucose. Company-authored mechanism paper.
Common questions
What is the Mitoglitazone dosage?
No human dose established. Amounts that were given in studies sit in Human evidence on this page. None of them is a personal dose. 2 completed trials, 385 people given it
Is Mitoglitazone FDA-approved?
Investigational, not approved. Mid-stage studies in type 2 diabetes (2011) and Alzheimer disease (2013) finished, with results posted. No final-stage study registered. The longer lifespan was in male mice started on it at 7 months old. A repeat test started at 18 months found no benefit in males and shorter lives in females.
Did mitoglitazone extend lifespan in the ITP?
In one run, in males started at 7 months, typical lifespan rose 8.7 percent (863 against 794 days). Survival to old age did not clearly change. When retested from 18 months in the next run, males gained nothing and females lived shorter lives, with the median down 5.8 percent across all labs. The harm to females held when the numbers left out a lab whose untreated mice lived unusually long.
Has mitoglitazone been tested in people?
Yes. In 356 people with type 2 diabetes, it lowered blood sugar as much as pioglitazone over 12 weeks, with about half the water retention. In 29 people with mild Alzheimer disease, it missed its main brain scan measure over 12 weeks. The maker ran both, between 2008 and 2013. No large final-stage trial followed.
Is it in trials for Parkinson disease?
No. The Parkinson data come from mice, worms (C. elegans) and nerve cells in a dish, in a 2016 paper. ClinicalTrials.gov lists no MSDC-0160 trial in Parkinson disease.
How is it different from pioglitazone?
It is a close chemical cousin. It keeps the action on the mitochondria's pyruvate gate and mostly drops the PPAR-gamma action. In the diabetes trial, that meant about the same drop in blood sugar with roughly half the signs of water retention. In the mouse program's 2022 run, both drugs shortened female lifespan when started late. So the shared mitochondria action, not PPAR-gamma, is the more likely cause of that harm. The authors say the cause is not clear.
Latest reading on Mitoglitazone
Blog posts about Mitoglitazone, newest first.