Hydromethylthionine

updated

Hydromethylthionine (HMTM, formerly LMTM, TRx0237) is a reduced form of the methylthioninium moiety of methylene blue, promoted with the claim that phase 3 trials show it slows Alzheimer's disease. By the authors' own account, all three phase 3 Alzheimer's programmes missed their randomized primary comparison, and the positive findings are post hoc, non-randomized and sponsor-linked. It is not approved anywhere: TauRx reports a UK application under MHRA review, and no MHRA decision was found as of 11 October 2026.

Tau aggregation inhibitor (reduced methylthioninium)

  • No regulator approval was found: on 11 October 2026 no FDA-approved HMTM product or label turned up, and no MHRA decision was found.
  • In the first phase 3 trial (891 participants, 15 months), ADAS-Cog change versus control was -0.02 (p = 0.9834) and -0.43 (p = 0.9323) at the two high doses, and the authors call the primary analysis negative.
  • In LUCIDITY (598 participants), the authors report it was not possible to show significant differences on the co-primary endpoints at 52 weeks.
  • The positive results come from non-randomized subgroup, exposure-response, delayed-start and matched-external-control analyses by the sponsor, TauRx.
  • A 2026 systematic review of tau-targeted trials, whose authors declare no conflicts, says LMTM did not demonstrate consistent clinical benefit in phase 3.
  • Methylene blue (PROVAYBLUE) is a different molecular form with its own FDA approval for acquired methemoglobinemia; its results and label do not transfer to HMTM.

What is hydromethylthionine, and has any regulator approved it?

Hydromethylthionine (HMTM, formerly LMTM, TRx0237) is a reduced form of the methylthioninium moiety of methylene blue, given in trials as a mesylate salt. It is not approved anywhere in the sources found: no FDA product turned up, and no MHRA decision was found as of 11 October 2026.

Hydromethylthionine is PubChem CID 164695, C16H19N3S, molecular weight 285.4 g/mol, CAS 613-11-6, with leucomethylene blue, leuco-methylthioninium and reduced methylene blue among the listed synonyms. That record is the free base. The trial drug is the mesylate salt (HMTM, formerly LMTM, leuco-methylthioninium bis(hydromethanesulfonate)), and the 2016 phase 3 paper calls LMTM a stable reduced form of the methylthioninium moiety. Methylene blue itself is a different record, CID 6099, C16H18ClN3S, 319.9 g/mol, CAS 61-73-4.

The names are not interchangeable in the trials. One control arm took 4 mg LMTM twice a day, another took 8 mg/day of the drug, and the LUCIDITY control arm took methylthioninium chloride, which is methylene blue, 4 mg twice weekly.

On 11 October 2026 an openFDA Drugs@FDA query for hydromethylthionine returned no match and a DailyMed search returned no labels, so no FDA-approved HMTM product was found. Methylene blue, by contrast, has an FDA label: PROVAYBLUE (NDA204630, original approval 8 April 2016) is indicated for pediatric and adult patients with acquired methemoglobinemia. Methylene blue has its own page on this site, and its results and grade do not transfer to HMTM or the reverse.

PubChem: HMTM identity; no FDA-approved HMTM product found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 164695 (hydromethylthionine) and CID 6099 (methylene blue). openFDA Drugs@FDA query and DailyMed search for hydromethylthionine run on 11 October 2026. PROVAYBLUE (methylene blue) injection prescribing information, NDA204630, label effective 28 May 2025.

Hydromethylthionine: PubChem CID 164695, C16H19N3S, 285.4 g/mol, CAS 613-11-6, with leucomethylene blue, leuco-methylthioninium and "reduced methylene blue" among the listed synonyms. That record is the free base; the trial drug is the mesylate salt (HMTM, formerly LMTM, leuco-methylthioninium bis(hydromethanesulfonate)). Methylene blue is a different record, CID 6099, C16H18ClN3S, 319.9 g/mol, CAS 61-73-4. On 11 October 2026 an openFDA Drugs@FDA query for hydromethylthionine returned no match and a DailyMed search returned no labels, so no FDA-approved HMTM product was found. Methylene blue, by contrast, has an FDA label: PROVAYBLUE (NDA204630, original approval 8 April 2016) is indicated for pediatric and adult patients with acquired methemoglobinemia, carries a boxed warning for serotonin syndrome with serotonergic drugs and opioids, and is contraindicated in G6PD deficiency (label effective 28 May 2025).

Participants / model
Chemical identity, not a clinical population
Treatment
Defined free base; the mesylate salt is the trial drug
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search and label
Funding
not stated in the source

The mesylate salt's own PubChem record and CAS number are not given here. Methylene blue results do not transfer to HMTM or the reverse.

Read the original source
First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source

Where does the UK marketing application stand?

Only the company's account is available. TauRx says the MHRA accepted its application in July 2024 and, on 1 October 2025, that the MHRA requires some further information before a final determination. A TauRx FAQ page says HMTM is not approved or licensed by any medicines regulator. No MHRA decision was found as of 11 October 2026.

TauRx announced on 29 July 2024 that the MHRA accepted its UK marketing authorisation application for HMTM, a 4 mg oral tablet for MCI due to Alzheimer's disease and mild to moderate Alzheimer's disease, under the 150-day accelerated assessment procedure. A company statement dated 31 December 2024 said the MHRA had requested further information, with a response due in late February 2025, and that the company was also liaising with NICE. On 4 June 2025 TauRx said it had completed the request for information. On 1 October 2025 TauRx said it had received formal confirmation that the MHRA requires some further information before a final determination can be reached.

A TauRx FAQ page, undated but citing the March 2026 trial paper, says HMTM is investigational and not approved or licensed by any medicines regulator, that the application is under MHRA review, that HMTM received an MHRA Innovation Passport in 2022, and that a final decision is expected in 2026.

These are company statements, not MHRA records. No MHRA decision, approval or refusal was found on 11 October 2026.

UK filing, company statements only (latest dated 1 October 2025); no MHRA decision found

Regulatory record

TauRx Therapeutics. Company press releases dated 29 July 2024, 31 December 2024, 4 June 2025 and 1 October 2025, and an undated FAQ page, on the MHRA review of the HMTM marketing authorisation application.

TauRx announced on 29 July 2024 that the MHRA accepted its UK marketing authorisation application for HMTM, a 4 mg oral tablet for MCI due to Alzheimer's disease and mild to moderate Alzheimer's disease, under the 150-day accelerated assessment procedure. A statement dated 31 December 2024 said the MHRA had requested further information, with a response due in late February 2025, and that the company was also liaising with NICE. On 4 June 2025 TauRx said it had completed the request for information. On 1 October 2025 TauRx said it had received formal confirmation that the MHRA requires some further information before a final determination can be reached. A TauRx FAQ page (undated, but citing the March 2026 trial paper) says HMTM is investigational and not approved or licensed by any medicines regulator, that the application is under MHRA review, that it received an MHRA Innovation Passport in 2022, and that a final decision is expected in 2026. No MHRA decision, approval or refusal was found on 11 October 2026.

Participants / model
Not a clinical population
Treatment
Marketing authorisation application, 4 mg oral tablet
Follow-up
Statements dated 29 July 2024 to 1 October 2025, plus an undated FAQ
Study design
Company press releases and FAQ (the sponsor's own account); no regulator record found
Funding
Company statements by TauRx Therapeutics

These are company statements, not MHRA records.

Read the original source

What did the phase 3 trials find?

None showed benefit on its randomized primary comparison, by the authors' own account. The 891-participant trial was negative on both coprimary outcomes, the 800-patient trial's positive results come from non-randomized subgroups, and a 220-patient frontotemporal dementia trial found no difference between doses as randomized. LUCIDITY is covered in the next section.

In a 15-month randomized double-blind trial at 115 centres in 16 countries, 891 participants with mild to moderate Alzheimer's disease were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome. ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Funding was TauRx Therapeutics. The trial is registered as NCT01689246.

In an 18-month phase 3 trial in mild Alzheimer's disease, 800 patients were randomized to 100 mg or 4 mg twice a day. The paper says that before unblinding the statistical analysis plan was revised to compare two subgroups: the 100 mg twice a day monotherapy subgroup (n = 79) versus the 4 mg arm as randomized (n = 396), and 4 mg monotherapy (n = 76) versus 4 mg add-on (n = 297). The authors call these non-randomized cohort analyses, report that they were significant at the required p < 0.025 for ADAS-cog, ADCS-ADL, MRI atrophy and glucose uptake, and say a further suitably randomized trial is required. Subgroups defined by whether patients also took symptomatic Alzheimer's drugs are not randomized groups. The abstract does not report the original randomized primary comparison, which a later paper says showed no difference in either of the first two phase 3 trials. Authors are mostly TauRx staff and advisors.

In a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia, 200 mg/day and 8 mg/day did not differ significantly as randomized, and the authors report a biphasic exposure-response.

First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
Second phase 3: 800 patients; positive results only in non-randomized subgroups

Clinical trial

Wilcock GK et al. J Alzheimers Dis. 2018;61(1):435-457. PMID 29154277.

In an 18-month phase 3 trial in mild Alzheimer's disease, 800 patients were randomized to 100 mg or 4 mg twice a day. The paper says that before unblinding the statistical analysis plan was revised to compare two subgroups: the 100 mg twice a day monotherapy subgroup (n = 79) versus the 4 mg twice a day arm as randomized (n = 396), and 4 mg monotherapy (n = 76) versus 4 mg add-on (n = 297). The authors call these non-randomized cohort analyses, report that they were significant at the required p < 0.025 for ADAS-cog, ADCS-ADL, MRI atrophy and glucose uptake, and say a further suitably randomized trial is required. The abstract does not report the original randomized primary comparison, which a later paper says showed no difference in either of two phase 3 trials.

Participants / model
800 patients with mild Alzheimer's disease; 79 and 76 in the analyzed monotherapy subgroups
Treatment
Oral LMTM 100 mg or 4 mg twice a day
Follow-up
18 months
Study design
Post hoc cohort analysis of a randomized phase 3 trial
Funding
Authors are mostly TauRx staff and advisors; funding and conflict statements are not stated in the abstract

Subgroups defined by whether patients also took symptomatic Alzheimer's drugs are not randomized groups. The registry record that matches by title and enrollment is NCT01689233 (n = 800, results not posted); the abstract does not cite it.

Read the original source
Exposure-response and pharmacokinetic modeling: post hoc, sponsor-linked

Human study

Schelter BO et al. J Alzheimers Dis. 2019;72(3):931-946. PMID 31658058.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), with differences in cognitive decline and brain atrophy in patients above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses; the authors say benefit is predicted to be maximal at 16 mg/day as monotherapy. The same approach in a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia (200 mg/day versus 8 mg/day) found no significant difference between doses as randomized and a biphasic exposure-response. A 2026 population pharmacokinetic model from five phase 1 studies and the LUCIDITY trial (710 participants, 7,784 plasma measurements) reports U-shaped clearance, with plasma levels at 12 months 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions, and concludes this explains the unexpected activity of the low-dose control and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design.

Participants / model
1,686 patients in two Alzheimer's trials (samples from 1,162); 220 patients with behavioral variant frontotemporal dementia; 710 participants in the pharmacokinetic model
Treatment
Oral hydromethylthionine, 8 to 250 mg/day in the source trials
Follow-up
Up to 18 months in the source trials; 52 weeks in the bvFTD trial
Study design
Post hoc exposure-response analyses and population pharmacokinetic modeling
Funding
The 2026 model's authors are Certara employees and TauRx employees and officers, several named on HMT patents; funding for the 2019 paper is not stated in the abstract

These are explanations for why the randomized comparisons were null, from the sponsor and its contractor, not randomized evidence of benefit.

Read the original source

What did LUCIDITY, the third phase 3 trial, find?

It missed its co-primary endpoints. The authors report it was not possible to show significant differences on ADAS-cog11 or ADCS-ADL23 at 52 weeks and attribute this to symptomatic activity in the methylene blue control arm. Later differences in participants with mild cognitive impairment come from delayed-start comparisons.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant. All received 16 mg/day after 52 weeks in a modified delayed-start design.

The authors report that it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks, and they attribute this to symptomatic activity in the control arm. The ClinicalTrials.gov record NCT03446001 (completed, results posted) lists 52-week ADAS-cog11 least-squares mean changes of 1.71 on control and 1.34 on 16 mg/day, and ADCS-ADL23 changes of -0.77 and -0.62; registry test statistics are not given here.

In participants with MCI, ADAS-cog13 differences emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock. Plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories.

LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source

What do the sponsor's exposure-response and external-control analyses add?

Explanations, not randomized proof. Sponsor-linked analyses argue that the low-dose controls were active and that benefit peaks at low plasma levels, and a comparison with matched historical controls found differences the authors call consistent with clinical benefit. None of these is a randomized comparison.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), differences in cognitive decline and brain atrophy above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses. A 2026 population pharmacokinetic model from five phase 1 studies and LUCIDITY (710 participants, 7,784 plasma measurements) reports U-shaped clearance and concludes this explains the unexpected activity of the low-dose control, and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design. Its authors are Certara employees and TauRx employees and officers, several named on HMT patents.

Because the authors say a conventional placebo-controlled trial was precluded by loss of blinding, they compared HMTM 16 mg/day in LUCIDITY with propensity-score matched placebo controls from the Critical Path for Alzheimer's Disease database (127 pairs), matched ADNI natural-history controls (189 pairs) and a meta-analysis of placebo arms (control n = 1,805 to 8,567). Versus matched placebo, ADAS-Cog13 and whole brain volume at 78 weeks differed with p < 0.0001, CDR-SB differed at 104 weeks (p < 0.001), and the odds of progressing to a more advanced CDR-Global stage were lower (overall OR 0.31, MCI OR 0.15). Matched historical controls are not randomized, differ in assessment conditions, and cannot replace a randomized control. Authors include TauRx staff and the chief executive.

A 2026 systematic review of eleven direct tau-targeted trials, whose authors declare no conflicts, says LMTM did not demonstrate consistent clinical benefit in phase 3 trials. It is a short abstract-level judgment, and only one LMTM study is among the eleven.

Exposure-response and pharmacokinetic modeling: post hoc, sponsor-linked

Human study

Schelter BO et al. J Alzheimers Dis. 2019;72(3):931-946. PMID 31658058.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), with differences in cognitive decline and brain atrophy in patients above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses; the authors say benefit is predicted to be maximal at 16 mg/day as monotherapy. The same approach in a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia (200 mg/day versus 8 mg/day) found no significant difference between doses as randomized and a biphasic exposure-response. A 2026 population pharmacokinetic model from five phase 1 studies and the LUCIDITY trial (710 participants, 7,784 plasma measurements) reports U-shaped clearance, with plasma levels at 12 months 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions, and concludes this explains the unexpected activity of the low-dose control and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design.

Participants / model
1,686 patients in two Alzheimer's trials (samples from 1,162); 220 patients with behavioral variant frontotemporal dementia; 710 participants in the pharmacokinetic model
Treatment
Oral hydromethylthionine, 8 to 250 mg/day in the source trials
Follow-up
Up to 18 months in the source trials; 52 weeks in the bvFTD trial
Study design
Post hoc exposure-response analyses and population pharmacokinetic modeling
Funding
The 2026 model's authors are Certara employees and TauRx employees and officers, several named on HMT patents; funding for the 2019 paper is not stated in the abstract

These are explanations for why the randomized comparisons were null, from the sponsor and its contractor, not randomized evidence of benefit.

Read the original source
External-control analysis: matched historical controls, non-randomized, sponsor-linked

Human study

Schelter BO et al. J Prev Alzheimers Dis. 2026;13(6):100560. PMID 42000569.

Because the authors say a conventional placebo-controlled trial was precluded by loss of blinding, they compared HMTM 16 mg/day in the LUCIDITY trial with propensity-score matched placebo controls from the Critical Path for Alzheimer's Disease database (127 pairs), matched ADNI natural-history controls (189 pairs) and a meta-analysis of placebo arms (control n = 1,805 to 8,567). Versus matched placebo, ADAS-Cog13 and whole brain volume at 78 weeks (primary) differed with p < 0.0001, CDR-SB differed at 104 weeks (p < 0.001), and the odds of progressing to a more advanced CDR-Global stage were lower (overall OR 0.31, MCI OR 0.15). The authors conclude the results are consistent with clinical benefit.

Participants / model
127 pairs (HMTM versus CPAD placebo) and 189 pairs (versus ADNI) with MCI or mild to moderate Alzheimer's disease
Treatment
Oral HMTM 16 mg/day versus historical controls
Follow-up
78 and 104 weeks
Study design
Non-randomized external-control analysis
Funding
Authors include TauRx staff and the chief executive; the abstract does not give a conflict statement

Matched historical controls are not randomized, differ in assessment conditions, and cannot replace a randomized control.

Read the original source
Independent review, 2026: LMTM did not show consistent clinical benefit in phase 3

Systematic review and meta-analysis

Yin L et al. Pharmacol Res. 2026;232:108439. PMID 42702342.

The review identified eleven direct tau-targeted trials (seven monoclonal antibody studies, two vaccine trials, one antisense oligonucleotide study and one small-molecule aggregation inhibitor study) and says the small-molecule aggregation inhibitor LMTM did not demonstrate consistent clinical benefit in phase 3 trials, and that tau-targeted therapies overall showed limited and inconsistent translation into clinical benefit.

Participants / model
Participants in eleven tau-targeted trials in Alzheimer's disease and primary tauopathies
Treatment
Antibody, vaccine, antisense and small-molecule trials
Follow-up
not stated in the source
Study design
Systematic review with risk-of-bias assessment
Funding
The authors declare no conflicts

A short abstract-level judgment, and only one LMTM study is among the eleven.

Read the original source

Does hydromethylthionine slow Alzheimer's progression, as phase 3 trials are said to show?

Not by a randomized primary result. A PubMed search on 11 October 2026 returned 21 trial records, and none reports a randomized comparison in which the prespecified primary clinical endpoints separated HMTM from control. The claim rests on analyses the authors themselves label post hoc, non-randomized or awaiting confirmation, all from the sponsor.

Sellers say hydromethylthionine, a methylene blue derivative also called HMTM, LMTM or TRx0237, slows the progression of Alzheimer's disease, as shown in phase 3 trials. In the three phase 3 Alzheimer's programmes the randomized comparisons were negative or did not separate from control by the authors' own account, and the supportive findings are subgroup, exposure-response, delayed-start and matched-external-control analyses.

A PubMed search on 11 October 2026 for hydromethylthionine, LMTM, leuco-methylthioninium or TRx0237 with randomized, phase 3 or clinical trial terms returned 21 records, and all 21 titles were reviewed. Six report efficacy analyses of the phase 3 program, including the frontotemporal dementia trial; in each, the authors describe a null or inconclusive result on the prespecified randomized comparison or present a post hoc, subgroup or external-control analysis. The rest were reviews, a design paper, preclinical mouse and in vitro studies, the 2026 pharmacokinetic model, a hypoxaemia study in phase 3 participants and an unrelated imaging paper.

All trials were sponsored by TauRx, and no independent randomized trial of HMTM in Alzheimer's disease was found. No regulator decision supporting the claim was found as of 11 October 2026.

PubMed search, 11 October 2026: 21 records, no randomized primary-endpoint success, no independent trial

PubMed search record

PubMed search run on 11 October 2026: (hydromethylthionine[tiab] OR LMTM[tiab] OR "leuco-methylthioninium"[tiab] OR TRx0237[tiab]) AND (randomized[tiab] OR randomised[tiab] OR "phase 3"[tiab] OR "phase III"[tiab] OR clinical trial[pt]). 21 records returned.

The search returned 21 records, and all 21 titles were reviewed. Six report efficacy analyses of the phase 3 program, including the bvFTD trial: in each, the authors describe a null or inconclusive result on the prespecified randomized comparison or present a post hoc, subgroup or external-control analysis. The rest were reviews, a design paper, preclinical mouse and in vitro studies, the 2026 pharmacokinetic model, a prospective hypoxaemia study in phase 3 participants, and a non-HMTM imaging paper. No record reports a randomized comparison in which the prespecified primary clinical endpoints separated HMTM from control. All trials were sponsored by TauRx, and no independent randomized trial of HMTM in Alzheimer's disease was found.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Literature search
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source
First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source
External-control analysis: matched historical controls, non-randomized, sponsor-linked

Human study

Schelter BO et al. J Prev Alzheimers Dis. 2026;13(6):100560. PMID 42000569.

Because the authors say a conventional placebo-controlled trial was precluded by loss of blinding, they compared HMTM 16 mg/day in the LUCIDITY trial with propensity-score matched placebo controls from the Critical Path for Alzheimer's Disease database (127 pairs), matched ADNI natural-history controls (189 pairs) and a meta-analysis of placebo arms (control n = 1,805 to 8,567). Versus matched placebo, ADAS-Cog13 and whole brain volume at 78 weeks (primary) differed with p < 0.0001, CDR-SB differed at 104 weeks (p < 0.001), and the odds of progressing to a more advanced CDR-Global stage were lower (overall OR 0.31, MCI OR 0.15). The authors conclude the results are consistent with clinical benefit.

Participants / model
127 pairs (HMTM versus CPAD placebo) and 189 pairs (versus ADNI) with MCI or mild to moderate Alzheimer's disease
Treatment
Oral HMTM 16 mg/day versus historical controls
Follow-up
78 and 104 weeks
Study design
Non-randomized external-control analysis
Funding
Authors include TauRx staff and the chief executive; the abstract does not give a conflict statement

Matched historical controls are not randomized, differ in assessment conditions, and cannot replace a randomized control.

Read the original source
UK filing, company statements only (latest dated 1 October 2025); no MHRA decision found

Regulatory record

TauRx Therapeutics. Company press releases dated 29 July 2024, 31 December 2024, 4 June 2025 and 1 October 2025, and an undated FAQ page, on the MHRA review of the HMTM marketing authorisation application.

TauRx announced on 29 July 2024 that the MHRA accepted its UK marketing authorisation application for HMTM, a 4 mg oral tablet for MCI due to Alzheimer's disease and mild to moderate Alzheimer's disease, under the 150-day accelerated assessment procedure. A statement dated 31 December 2024 said the MHRA had requested further information, with a response due in late February 2025, and that the company was also liaising with NICE. On 4 June 2025 TauRx said it had completed the request for information. On 1 October 2025 TauRx said it had received formal confirmation that the MHRA requires some further information before a final determination can be reached. A TauRx FAQ page (undated, but citing the March 2026 trial paper) says HMTM is investigational and not approved or licensed by any medicines regulator, that the application is under MHRA review, that it received an MHRA Innovation Passport in 2022, and that a final decision is expected in 2026. No MHRA decision, approval or refusal was found on 11 October 2026.

Participants / model
Not a clinical population
Treatment
Marketing authorisation application, 4 mg oral tablet
Follow-up
Statements dated 29 July 2024 to 1 October 2025, plus an undated FAQ
Study design
Company press releases and FAQ (the sponsor's own account); no regulator record found
Funding
Company statements by TauRx Therapeutics

These are company statements, not MHRA records.

Read the original source

How is hydromethylthionine supposed to work?

It is classed as a small-molecule tau aggregation inhibitor, a stable reduced form of the methylthioninium moiety. The sponsor's exposure-response work argues that effects appear at low plasma levels, which is why the low-dose control arms are said to have been active; that argument comes from post hoc analyses, not randomized results.

A 2026 systematic review of tau-targeted trials classes LMTM as a small-molecule aggregation inhibitor, the only one among the eleven trials it reviewed. The 2016 phase 3 paper calls LMTM a stable reduced form of the methylthioninium moiety. Methylene blue (methylthioninium chloride) is a different molecular form, with its own chemical record and its own trial history.

The sponsor's analyses describe steep concentration-response at the 8 mg/day control dose (plasma 0.3 to 0.8 ng/mL) and no added benefit at the 4 to 21 ng/mL reached on high doses, and a 2026 model reports U-shaped clearance with 12-month plasma levels 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions. No half-life appears in the sources on this page.

Independent review, 2026: LMTM did not show consistent clinical benefit in phase 3

Systematic review and meta-analysis

Yin L et al. Pharmacol Res. 2026;232:108439. PMID 42702342.

The review identified eleven direct tau-targeted trials (seven monoclonal antibody studies, two vaccine trials, one antisense oligonucleotide study and one small-molecule aggregation inhibitor study) and says the small-molecule aggregation inhibitor LMTM did not demonstrate consistent clinical benefit in phase 3 trials, and that tau-targeted therapies overall showed limited and inconsistent translation into clinical benefit.

Participants / model
Participants in eleven tau-targeted trials in Alzheimer's disease and primary tauopathies
Treatment
Antibody, vaccine, antisense and small-molecule trials
Follow-up
not stated in the source
Study design
Systematic review with risk-of-bias assessment
Funding
The authors declare no conflicts

A short abstract-level judgment, and only one LMTM study is among the eleven.

Read the original source
First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
Exposure-response and pharmacokinetic modeling: post hoc, sponsor-linked

Human study

Schelter BO et al. J Alzheimers Dis. 2019;72(3):931-946. PMID 31658058.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), with differences in cognitive decline and brain atrophy in patients above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses; the authors say benefit is predicted to be maximal at 16 mg/day as monotherapy. The same approach in a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia (200 mg/day versus 8 mg/day) found no significant difference between doses as randomized and a biphasic exposure-response. A 2026 population pharmacokinetic model from five phase 1 studies and the LUCIDITY trial (710 participants, 7,784 plasma measurements) reports U-shaped clearance, with plasma levels at 12 months 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions, and concludes this explains the unexpected activity of the low-dose control and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design.

Participants / model
1,686 patients in two Alzheimer's trials (samples from 1,162); 220 patients with behavioral variant frontotemporal dementia; 710 participants in the pharmacokinetic model
Treatment
Oral hydromethylthionine, 8 to 250 mg/day in the source trials
Follow-up
Up to 18 months in the source trials; 52 weeks in the bvFTD trial
Study design
Post hoc exposure-response analyses and population pharmacokinetic modeling
Funding
The 2026 model's authors are Certara employees and TauRx employees and officers, several named on HMT patents; funding for the 2019 paper is not stated in the abstract

These are explanations for why the randomized comparisons were null, from the sponsor and its contractor, not randomized evidence of benefit.

Read the original source
PubChem: HMTM identity; no FDA-approved HMTM product found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 164695 (hydromethylthionine) and CID 6099 (methylene blue). openFDA Drugs@FDA query and DailyMed search for hydromethylthionine run on 11 October 2026. PROVAYBLUE (methylene blue) injection prescribing information, NDA204630, label effective 28 May 2025.

Hydromethylthionine: PubChem CID 164695, C16H19N3S, 285.4 g/mol, CAS 613-11-6, with leucomethylene blue, leuco-methylthioninium and "reduced methylene blue" among the listed synonyms. That record is the free base; the trial drug is the mesylate salt (HMTM, formerly LMTM, leuco-methylthioninium bis(hydromethanesulfonate)). Methylene blue is a different record, CID 6099, C16H18ClN3S, 319.9 g/mol, CAS 61-73-4. On 11 October 2026 an openFDA Drugs@FDA query for hydromethylthionine returned no match and a DailyMed search returned no labels, so no FDA-approved HMTM product was found. Methylene blue, by contrast, has an FDA label: PROVAYBLUE (NDA204630, original approval 8 April 2016) is indicated for pediatric and adult patients with acquired methemoglobinemia, carries a boxed warning for serotonin syndrome with serotonergic drugs and opioids, and is contraindicated in G6PD deficiency (label effective 28 May 2025).

Participants / model
Chemical identity, not a clinical population
Treatment
Defined free base; the mesylate salt is the trial drug
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search and label
Funding
not stated in the source

The mesylate salt's own PubChem record and CAS number are not given here. Methylene blue results do not transfer to HMTM or the reverse.

Read the original source

What safety signals did the trials report?

Trial safety signals look modest: gastrointestinal and urinary effects led discontinuations at high doses, dose-dependent hemoglobin reductions were the most common laboratory abnormality, and headache (1.5%) and diarrhoea (1.2%) were the most frequent adverse effects in LUCIDITY. Whether the methylene blue label warnings apply to HMTM is not established, and no pregnancy, interaction or long-term data were found beyond the trials.

In the 891-participant trial, gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation. Non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded. In LUCIDITY the most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

The methylene blue label (PROVAYBLUE) carries a boxed warning for serotonin syndrome with serotonergic drugs and opioids and is contraindicated in G6PD deficiency. Those warnings belong to methylene blue and were not tested for HMTM. Whether they apply to the reduced mesylate form at 8 to 16 mg/day is not established in the sources on this page, and the trial exclusion of drugs with methemoglobinemia warnings is the only related design feature found.

No pregnancy, drug interaction or long-term safety data for HMTM were found beyond the trials above.

First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source
PubChem: HMTM identity; no FDA-approved HMTM product found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 164695 (hydromethylthionine) and CID 6099 (methylene blue). openFDA Drugs@FDA query and DailyMed search for hydromethylthionine run on 11 October 2026. PROVAYBLUE (methylene blue) injection prescribing information, NDA204630, label effective 28 May 2025.

Hydromethylthionine: PubChem CID 164695, C16H19N3S, 285.4 g/mol, CAS 613-11-6, with leucomethylene blue, leuco-methylthioninium and "reduced methylene blue" among the listed synonyms. That record is the free base; the trial drug is the mesylate salt (HMTM, formerly LMTM, leuco-methylthioninium bis(hydromethanesulfonate)). Methylene blue is a different record, CID 6099, C16H18ClN3S, 319.9 g/mol, CAS 61-73-4. On 11 October 2026 an openFDA Drugs@FDA query for hydromethylthionine returned no match and a DailyMed search returned no labels, so no FDA-approved HMTM product was found. Methylene blue, by contrast, has an FDA label: PROVAYBLUE (NDA204630, original approval 8 April 2016) is indicated for pediatric and adult patients with acquired methemoglobinemia, carries a boxed warning for serotonin syndrome with serotonergic drugs and opioids, and is contraindicated in G6PD deficiency (label effective 28 May 2025).

Participants / model
Chemical identity, not a clinical population
Treatment
Defined free base; the mesylate salt is the trial drug
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search and label
Funding
not stated in the source

The mesylate salt's own PubChem record and CAS number are not given here. Methylene blue results do not transfer to HMTM or the reverse.

Read the original source

Why does hydromethylthionine sit in C tier?

C reflects a plausible mechanism and small, hard-to-interpret data. A filed drug awaiting a decision would ordinarily rate B, but three phase 3 programmes missed the randomized primary comparison by the authors' own account. D would overstate it, given modest trial safety signals and a UK decision that is still outstanding.

Three phase 3 programmes (891, 800 and 598 Alzheimer's participants, plus 220 with behavioral variant frontotemporal dementia) did not show separation from control on the randomized primary comparison as the authors report it. The positive findings come from subgroup, exposure-response, delayed-start and matched-external-control analyses by the sponsor, and the only independent summary says clinical benefit was not consistent.

The A grade for methylene blue rests on an approved drug used off label and does not carry over. Methylene blue and HMTM are different molecular forms with different trial histories. "Slows Alzheimer's progression" is not established by a randomized primary result, and no UK regulator decision was found.

First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
Second phase 3: 800 patients; positive results only in non-randomized subgroups

Clinical trial

Wilcock GK et al. J Alzheimers Dis. 2018;61(1):435-457. PMID 29154277.

In an 18-month phase 3 trial in mild Alzheimer's disease, 800 patients were randomized to 100 mg or 4 mg twice a day. The paper says that before unblinding the statistical analysis plan was revised to compare two subgroups: the 100 mg twice a day monotherapy subgroup (n = 79) versus the 4 mg twice a day arm as randomized (n = 396), and 4 mg monotherapy (n = 76) versus 4 mg add-on (n = 297). The authors call these non-randomized cohort analyses, report that they were significant at the required p < 0.025 for ADAS-cog, ADCS-ADL, MRI atrophy and glucose uptake, and say a further suitably randomized trial is required. The abstract does not report the original randomized primary comparison, which a later paper says showed no difference in either of two phase 3 trials.

Participants / model
800 patients with mild Alzheimer's disease; 79 and 76 in the analyzed monotherapy subgroups
Treatment
Oral LMTM 100 mg or 4 mg twice a day
Follow-up
18 months
Study design
Post hoc cohort analysis of a randomized phase 3 trial
Funding
Authors are mostly TauRx staff and advisors; funding and conflict statements are not stated in the abstract

Subgroups defined by whether patients also took symptomatic Alzheimer's drugs are not randomized groups. The registry record that matches by title and enrollment is NCT01689233 (n = 800, results not posted); the abstract does not cite it.

Read the original source
Exposure-response and pharmacokinetic modeling: post hoc, sponsor-linked

Human study

Schelter BO et al. J Alzheimers Dis. 2019;72(3):931-946. PMID 31658058.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), with differences in cognitive decline and brain atrophy in patients above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses; the authors say benefit is predicted to be maximal at 16 mg/day as monotherapy. The same approach in a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia (200 mg/day versus 8 mg/day) found no significant difference between doses as randomized and a biphasic exposure-response. A 2026 population pharmacokinetic model from five phase 1 studies and the LUCIDITY trial (710 participants, 7,784 plasma measurements) reports U-shaped clearance, with plasma levels at 12 months 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions, and concludes this explains the unexpected activity of the low-dose control and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design.

Participants / model
1,686 patients in two Alzheimer's trials (samples from 1,162); 220 patients with behavioral variant frontotemporal dementia; 710 participants in the pharmacokinetic model
Treatment
Oral hydromethylthionine, 8 to 250 mg/day in the source trials
Follow-up
Up to 18 months in the source trials; 52 weeks in the bvFTD trial
Study design
Post hoc exposure-response analyses and population pharmacokinetic modeling
Funding
The 2026 model's authors are Certara employees and TauRx employees and officers, several named on HMT patents; funding for the 2019 paper is not stated in the abstract

These are explanations for why the randomized comparisons were null, from the sponsor and its contractor, not randomized evidence of benefit.

Read the original source
LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source
External-control analysis: matched historical controls, non-randomized, sponsor-linked

Human study

Schelter BO et al. J Prev Alzheimers Dis. 2026;13(6):100560. PMID 42000569.

Because the authors say a conventional placebo-controlled trial was precluded by loss of blinding, they compared HMTM 16 mg/day in the LUCIDITY trial with propensity-score matched placebo controls from the Critical Path for Alzheimer's Disease database (127 pairs), matched ADNI natural-history controls (189 pairs) and a meta-analysis of placebo arms (control n = 1,805 to 8,567). Versus matched placebo, ADAS-Cog13 and whole brain volume at 78 weeks (primary) differed with p < 0.0001, CDR-SB differed at 104 weeks (p < 0.001), and the odds of progressing to a more advanced CDR-Global stage were lower (overall OR 0.31, MCI OR 0.15). The authors conclude the results are consistent with clinical benefit.

Participants / model
127 pairs (HMTM versus CPAD placebo) and 189 pairs (versus ADNI) with MCI or mild to moderate Alzheimer's disease
Treatment
Oral HMTM 16 mg/day versus historical controls
Follow-up
78 and 104 weeks
Study design
Non-randomized external-control analysis
Funding
Authors include TauRx staff and the chief executive; the abstract does not give a conflict statement

Matched historical controls are not randomized, differ in assessment conditions, and cannot replace a randomized control.

Read the original source
Independent review, 2026: LMTM did not show consistent clinical benefit in phase 3

Systematic review and meta-analysis

Yin L et al. Pharmacol Res. 2026;232:108439. PMID 42702342.

The review identified eleven direct tau-targeted trials (seven monoclonal antibody studies, two vaccine trials, one antisense oligonucleotide study and one small-molecule aggregation inhibitor study) and says the small-molecule aggregation inhibitor LMTM did not demonstrate consistent clinical benefit in phase 3 trials, and that tau-targeted therapies overall showed limited and inconsistent translation into clinical benefit.

Participants / model
Participants in eleven tau-targeted trials in Alzheimer's disease and primary tauopathies
Treatment
Antibody, vaccine, antisense and small-molecule trials
Follow-up
not stated in the source
Study design
Systematic review with risk-of-bias assessment
Funding
The authors declare no conflicts

A short abstract-level judgment, and only one LMTM study is among the eleven.

Read the original source
UK filing, company statements only (latest dated 1 October 2025); no MHRA decision found

Regulatory record

TauRx Therapeutics. Company press releases dated 29 July 2024, 31 December 2024, 4 June 2025 and 1 October 2025, and an undated FAQ page, on the MHRA review of the HMTM marketing authorisation application.

TauRx announced on 29 July 2024 that the MHRA accepted its UK marketing authorisation application for HMTM, a 4 mg oral tablet for MCI due to Alzheimer's disease and mild to moderate Alzheimer's disease, under the 150-day accelerated assessment procedure. A statement dated 31 December 2024 said the MHRA had requested further information, with a response due in late February 2025, and that the company was also liaising with NICE. On 4 June 2025 TauRx said it had completed the request for information. On 1 October 2025 TauRx said it had received formal confirmation that the MHRA requires some further information before a final determination can be reached. A TauRx FAQ page (undated, but citing the March 2026 trial paper) says HMTM is investigational and not approved or licensed by any medicines regulator, that the application is under MHRA review, that it received an MHRA Innovation Passport in 2022, and that a final decision is expected in 2026. No MHRA decision, approval or refusal was found on 11 October 2026.

Participants / model
Not a clinical population
Treatment
Marketing authorisation application, 4 mg oral tablet
Follow-up
Statements dated 29 July 2024 to 1 October 2025, plus an undated FAQ
Study design
Company press releases and FAQ (the sponsor's own account); no regulator record found
Funding
Company statements by TauRx Therapeutics

These are company statements, not MHRA records.

Read the original source

Studies and sources

PubChem: HMTM identity; no FDA-approved HMTM product found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 164695 (hydromethylthionine) and CID 6099 (methylene blue). openFDA Drugs@FDA query and DailyMed search for hydromethylthionine run on 11 October 2026. PROVAYBLUE (methylene blue) injection prescribing information, NDA204630, label effective 28 May 2025.

Hydromethylthionine: PubChem CID 164695, C16H19N3S, 285.4 g/mol, CAS 613-11-6, with leucomethylene blue, leuco-methylthioninium and "reduced methylene blue" among the listed synonyms. That record is the free base; the trial drug is the mesylate salt (HMTM, formerly LMTM, leuco-methylthioninium bis(hydromethanesulfonate)). Methylene blue is a different record, CID 6099, C16H18ClN3S, 319.9 g/mol, CAS 61-73-4. On 11 October 2026 an openFDA Drugs@FDA query for hydromethylthionine returned no match and a DailyMed search returned no labels, so no FDA-approved HMTM product was found. Methylene blue, by contrast, has an FDA label: PROVAYBLUE (NDA204630, original approval 8 April 2016) is indicated for pediatric and adult patients with acquired methemoglobinemia, carries a boxed warning for serotonin syndrome with serotonergic drugs and opioids, and is contraindicated in G6PD deficiency (label effective 28 May 2025).

Participants / model
Chemical identity, not a clinical population
Treatment
Defined free base; the mesylate salt is the trial drug
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search and label
Funding
not stated in the source

The mesylate salt's own PubChem record and CAS number are not given here. Methylene blue results do not transfer to HMTM or the reverse.

Read the original source
First phase 3: 891 participants, 15 months, negative on both coprimary outcomes

Randomized controlled trial

Gauthier S et al. Lancet. 2016;388(10062):2873-2884. PMID 27863809.

891 participants with mild to moderate Alzheimer's disease at 115 centres in 16 countries were assigned to LMTM 75 mg twice a day (268), 125 mg twice a day (266) or control, 4 mg LMTM twice a day to maintain blinding for urine and fecal discoloration (357). Patients on other Alzheimer's medicines were allowed. At week 65 the prespecified primary analyses showed no benefit on either coprimary outcome: ADAS-Cog change versus control was -0.02 (95% CI -1.60 to 1.56, p = 0.9834) at 75 mg twice a day and -0.43 (95% CI -2.06 to 1.20, p = 0.9323) at 125 mg twice a day, and ADCS-ADL change was -0.93 (95% CI -3.12 to 1.26, p = 0.8659) and -0.34 (95% CI -2.61 to 1.93, p = 0.9479). The authors call the primary analysis negative and say the results do not suggest benefit of LMTM as add-on treatment. Gastrointestinal and urinary effects were the most common adverse events at the two high doses and the most common reasons for discontinuation, non-clinically significant dose-dependent hemoglobin reductions were the most common laboratory abnormality, and amyloid-related imaging abnormalities were seen in less than 1% (8 of 885) of participants. Patients using medicines with methemoglobinemia warnings were excluded.

Participants / model
891 participants with mild to moderate Alzheimer's disease
Treatment
Oral LMTM tablets 75 or 125 mg twice a day versus 4 mg twice a day
Follow-up
15 months (primary analysis at week 65)
Study design
Randomized double-blind phase 3 trial
Funding
TauRx Therapeutics. The last author, Claude Wischik, is described on the company's FAQ page as its co-founder and chief executive

The control arm took a low dose of the same drug to keep urine discoloration blinded.

Read the original source
Second phase 3: 800 patients; positive results only in non-randomized subgroups

Clinical trial

Wilcock GK et al. J Alzheimers Dis. 2018;61(1):435-457. PMID 29154277.

In an 18-month phase 3 trial in mild Alzheimer's disease, 800 patients were randomized to 100 mg or 4 mg twice a day. The paper says that before unblinding the statistical analysis plan was revised to compare two subgroups: the 100 mg twice a day monotherapy subgroup (n = 79) versus the 4 mg twice a day arm as randomized (n = 396), and 4 mg monotherapy (n = 76) versus 4 mg add-on (n = 297). The authors call these non-randomized cohort analyses, report that they were significant at the required p < 0.025 for ADAS-cog, ADCS-ADL, MRI atrophy and glucose uptake, and say a further suitably randomized trial is required. The abstract does not report the original randomized primary comparison, which a later paper says showed no difference in either of two phase 3 trials.

Participants / model
800 patients with mild Alzheimer's disease; 79 and 76 in the analyzed monotherapy subgroups
Treatment
Oral LMTM 100 mg or 4 mg twice a day
Follow-up
18 months
Study design
Post hoc cohort analysis of a randomized phase 3 trial
Funding
Authors are mostly TauRx staff and advisors; funding and conflict statements are not stated in the abstract

Subgroups defined by whether patients also took symptomatic Alzheimer's drugs are not randomized groups. The registry record that matches by title and enrollment is NCT01689233 (n = 800, results not posted); the abstract does not cite it.

Read the original source
Exposure-response and pharmacokinetic modeling: post hoc, sponsor-linked

Human study

Schelter BO et al. J Alzheimers Dis. 2019;72(3):931-946. PMID 31658058.

A population pharmacokinetic analysis of samples from 1,162 of the 1,686 patients in the first two phase 3 trials reported steep concentration-response relationships at the 8 mg/day control dose (plasma levels of 0.3 to 0.8 ng/mL), with differences in cognitive decline and brain atrophy in patients above a plasma threshold, effect sizes halved as add-on, and no additional benefit at the 4 to 21 ng/mL produced by high doses; the authors say benefit is predicted to be maximal at 16 mg/day as monotherapy. The same approach in a 52-week phase 3 trial in 220 patients with behavioral variant frontotemporal dementia (200 mg/day versus 8 mg/day) found no significant difference between doses as randomized and a biphasic exposure-response. A 2026 population pharmacokinetic model from five phase 1 studies and the LUCIDITY trial (710 participants, 7,784 plasma measurements) reports U-shaped clearance, with plasma levels at 12 months 3 times (HMTM) and 5 times (methylthioninium chloride) above linear predictions, and concludes this explains the unexpected activity of the low-dose control and that comparable urine discoloration without therapeutic activity may not be possible in a standard placebo-controlled design.

Participants / model
1,686 patients in two Alzheimer's trials (samples from 1,162); 220 patients with behavioral variant frontotemporal dementia; 710 participants in the pharmacokinetic model
Treatment
Oral hydromethylthionine, 8 to 250 mg/day in the source trials
Follow-up
Up to 18 months in the source trials; 52 weeks in the bvFTD trial
Study design
Post hoc exposure-response analyses and population pharmacokinetic modeling
Funding
The 2026 model's authors are Certara employees and TauRx employees and officers, several named on HMT patents; funding for the 2019 paper is not stated in the abstract

These are explanations for why the randomized comparisons were null, from the sponsor and its contractor, not randomized evidence of benefit.

Read the original source
LUCIDITY: 598 participants; co-primary endpoints not met at 52 weeks

Randomized controlled trial

ClinicalTrials.gov record NCT03446001 (LUCIDITY; completed, results posted), read 11 October 2026, with the trial paper as reported. LUCIDITY adverse-event figures from PMID 41570392.

598 amyloid-PET positive participants (263 with mild cognitive impairment due to Alzheimer's disease, 335 with mild to moderate dementia) at 82 centres were randomized to HMTM 16 mg/day, HMTM 8 mg/day or methylthioninium chloride 4 mg twice weekly, intended as an inactive urinary colourant, with all receiving 16 mg/day after 52 weeks in a modified delayed-start design. The authors report it was not possible to show significant differences on the co-primary endpoints (ADAS-cog11 and ADCS-ADL23) at 52 weeks and attribute this to symptomatic activity in the control arm. In participants with MCI, differences in ADAS-cog13 emerged at 78 weeks (p = 0.0291) and 104 weeks (p = 0.0308) between early- and delayed-start 16 mg/day, in analyses the authors say were specified before the 24-month database lock; plasma neurofilament light fell more in the whole population at 52 weeks (p = 0.0291), and pTau217 was lower in MCI (p = 0.0165). The registry lists 52-week ADAS-cog11 least-squares mean changes of 1.71 (control) and 1.34 (16 mg/day) and ADCS-ADL23 changes of -0.77 and -0.62. The most frequent adverse effects were headache (1.5%) and diarrhoea (1.2%).

Participants / model
598 participants with MCI due to Alzheimer's disease or mild to moderate Alzheimer's dementia
Treatment
Oral HMTM 16 or 8 mg/day versus methylthioninium chloride 4 mg twice weekly; all on 16 mg/day after 52 weeks
Follow-up
52 weeks randomized, with delayed-start follow-up to 104 weeks
Study design
Randomized phase 3 trial with a modified delayed-start design, with registry results
Funding
Company-sponsored. Most authors are TauRx staff or advisors, and two disclose patents owned by WisTa Laboratories

The control arm took methylene blue, a different molecular form. Registry figures are given without test statistics here.

Read the original source
External-control analysis: matched historical controls, non-randomized, sponsor-linked

Human study

Schelter BO et al. J Prev Alzheimers Dis. 2026;13(6):100560. PMID 42000569.

Because the authors say a conventional placebo-controlled trial was precluded by loss of blinding, they compared HMTM 16 mg/day in the LUCIDITY trial with propensity-score matched placebo controls from the Critical Path for Alzheimer's Disease database (127 pairs), matched ADNI natural-history controls (189 pairs) and a meta-analysis of placebo arms (control n = 1,805 to 8,567). Versus matched placebo, ADAS-Cog13 and whole brain volume at 78 weeks (primary) differed with p < 0.0001, CDR-SB differed at 104 weeks (p < 0.001), and the odds of progressing to a more advanced CDR-Global stage were lower (overall OR 0.31, MCI OR 0.15). The authors conclude the results are consistent with clinical benefit.

Participants / model
127 pairs (HMTM versus CPAD placebo) and 189 pairs (versus ADNI) with MCI or mild to moderate Alzheimer's disease
Treatment
Oral HMTM 16 mg/day versus historical controls
Follow-up
78 and 104 weeks
Study design
Non-randomized external-control analysis
Funding
Authors include TauRx staff and the chief executive; the abstract does not give a conflict statement

Matched historical controls are not randomized, differ in assessment conditions, and cannot replace a randomized control.

Read the original source
Independent review, 2026: LMTM did not show consistent clinical benefit in phase 3

Systematic review and meta-analysis

Yin L et al. Pharmacol Res. 2026;232:108439. PMID 42702342.

The review identified eleven direct tau-targeted trials (seven monoclonal antibody studies, two vaccine trials, one antisense oligonucleotide study and one small-molecule aggregation inhibitor study) and says the small-molecule aggregation inhibitor LMTM did not demonstrate consistent clinical benefit in phase 3 trials, and that tau-targeted therapies overall showed limited and inconsistent translation into clinical benefit.

Participants / model
Participants in eleven tau-targeted trials in Alzheimer's disease and primary tauopathies
Treatment
Antibody, vaccine, antisense and small-molecule trials
Follow-up
not stated in the source
Study design
Systematic review with risk-of-bias assessment
Funding
The authors declare no conflicts

A short abstract-level judgment, and only one LMTM study is among the eleven.

Read the original source
UK filing, company statements only (latest dated 1 October 2025); no MHRA decision found

Regulatory record

TauRx Therapeutics. Company press releases dated 29 July 2024, 31 December 2024, 4 June 2025 and 1 October 2025, and an undated FAQ page, on the MHRA review of the HMTM marketing authorisation application.

TauRx announced on 29 July 2024 that the MHRA accepted its UK marketing authorisation application for HMTM, a 4 mg oral tablet for MCI due to Alzheimer's disease and mild to moderate Alzheimer's disease, under the 150-day accelerated assessment procedure. A statement dated 31 December 2024 said the MHRA had requested further information, with a response due in late February 2025, and that the company was also liaising with NICE. On 4 June 2025 TauRx said it had completed the request for information. On 1 October 2025 TauRx said it had received formal confirmation that the MHRA requires some further information before a final determination can be reached. A TauRx FAQ page (undated, but citing the March 2026 trial paper) says HMTM is investigational and not approved or licensed by any medicines regulator, that the application is under MHRA review, that it received an MHRA Innovation Passport in 2022, and that a final decision is expected in 2026. No MHRA decision, approval or refusal was found on 11 October 2026.

Participants / model
Not a clinical population
Treatment
Marketing authorisation application, 4 mg oral tablet
Follow-up
Statements dated 29 July 2024 to 1 October 2025, plus an undated FAQ
Study design
Company press releases and FAQ (the sponsor's own account); no regulator record found
Funding
Company statements by TauRx Therapeutics

These are company statements, not MHRA records.

Read the original source

Why is Hydromethylthionine in C tier?

C reflects a plausible mechanism and small, hard-to-interpret data. Three phase 3 programmes did not separate from control on the randomized primary comparison as the authors report it, the positive findings are sponsor analyses that are post hoc or non-randomized, and the trial safety signals look modest.