Beyond Amyloid: Why Tau Is the Next Frontier in Alzheimer’s Therapy

Clinical Review · Treatment Watch · October 2026

Anti-amyloid antibodies have slowed clinical decline by roughly one quarter to one third over 18 months in trials of people with early, symptomatic Alzheimer’s disease, but they have not stopped decline or reversed established impairment. Tau pathology is strongly associated with the location and progression of cognitive deficits, making it an important therapeutic target. Continued decline despite amyloid removal may also reflect established neuronal injury, coexisting brain pathologies, and aspects of Alzheimer’s biology that remain incompletely understood.

As of October 2026, no tau-targeting drug is approved for Alzheimer’s disease. Diranersen (BIIB080), an investigational treatment administered into the cerebrospinal fluid, produced substantial reductions in measured tau biomarkers and encouraging clinical signals in the Phase 2 CELIA trial. CELIA did not meet its prespecified primary endpoint, however, and the reported clinical findings require confirmation in Phase 3. The results apply to research in early, symptomatic Alzheimer’s disease—not prevention in otherwise well APOE ε4 carriers. This review summarizes the evidence and outlines what it means for people living with or at risk for Alzheimer’s disease, including carriers of the APOE ε4 allele.

Why anti-amyloid therapies do not cure Alzheimer’s dementia

Lecanemab and donanemab are a genuine milestone: they were the first drugs shown in Phase 3 trials to remove amyloid plaque and slow clinical decline. But slowing is not stopping. Treated patients in both pivotal trials still declined, only more slowly than those on placebo. (For a detailed look at the two drugs, see Do Leqembi and Kisunla Meaningfully Help? and Lecanemab vs Donanemab: Why the Amyloid Target Matters.)

CDR-SB is an 18-point scale of cognition and daily function; higher is worse. ARIA-E is amyloid-related imaging abnormality with edema (brain swelling).

Five biological and clinical observations help explain why removing amyloid has not been a cure.

1. Amyloid is an early process, not the whole disease. Amyloid accumulation is an important early process in Alzheimer’s disease, beginning years to decades before symptoms. Tau pathology, neurodegeneration, and other coexisting brain pathologies may contribute to continued decline despite amyloid removal; the relative contribution of each remains under investigation. In studied cohorts, tau PET location and burden correspond closely to patterns of atrophy and the cognitive domains affected, and baseline tau PET predicts subsequent cognitive change more strongly than amyloid PET does (Ossenkoppele et al., Brain 2016; Bejanin et al., Brain 2017; Ossenkoppele et al., JAMA Neurology 2021). This association is consistent with a central role for tau, but it does not establish that tau alone drives decline once amyloid is removed.

2. The pivotal trials are consistent with a role for tau. In TRAILBLAZER-ALZ 2, participants with low/medium baseline tau showed greater slowing than the overall population, which included people with high tau. This pattern suggests that anti-amyloid treatment may be most effective earlier in the disease, although the trial was not designed to establish why.

3. Established injury has not been shown to reverse. By the stage of mild cognitive impairment, substantial synaptic and neuronal loss has typically occurred. Amyloid clearance has not been shown to restore established neuronal injury or reverse clinical impairment.

4. Most older brains carry more than one disease. Autopsy studies show that Alzheimer’s pathology in older adults usually coexists with vascular injury, TDP-43 proteinopathy (LATE), or Lewy bodies (Kapasi, DeCarli and Schneider, Acta Neuropathologica 2017). An anti-amyloid drug does not treat these contributors.

5. Safety limits how hard amyloid can be pushed. ARIA is the dose-limiting risk, and as the table shows, it is highest in APOE ε4 homozygotes. That restricts eligibility and treatment intensity for the very group at greatest genetic risk.

The next target: tau

Tau is among the most active therapeutic targets after amyloid. Within the disease-targeting segment of the Alzheimer’s drug pipeline, tau-directed agents have grown from about 6% to about 20% of agents since 2017, while amyloid-targeted agents have fallen from about 33% to about 20% (Cummings et al., Alzheimer’s & Dementia: TRCI 2026). Across the full 2026 pipeline of 158 drugs, 15 (9%) target tau-related processes and 25 (16%) target amyloid-related processes.

What tau does. Tau is a protein, encoded by the MAPT gene, that stabilizes microtubules, the internal transport tracks of neurons. In Alzheimer’s disease, tau becomes abnormally phosphorylated, detaches from microtubules, misfolds and aggregates into neurofibrillary tangles. Misfolded tau “seeds” can pass from neuron to neuron along connected circuits, which is thought to underlie the stereotyped spread described by Braak staging. Tangle burden is closely associated with regional synaptic and neuronal loss. As UCSF neurologist Adam Boxer has summarized the long-standing hypothesis, “amyloid might be the match, but tau is really the fire” (Boston Globe, July 17, 2026).

Five ways to attack tau. Each strategy intervenes at a different step.

Lessons from the failures. The first generation of tau antibodies targeted the N-terminus of tau and did not slow decline, even when they engaged their target; gosuranemab, for example, lowered unbound N-terminal tau in CSF without clinical benefit (Teng et al., JAMA Neurology 2022; Shulman et al., Nature Aging 2023; Florian et al., Brain 2023). One proposed explanation is that most pathological tau sits inside neurons, where antibodies have limited reach, and N-terminal fragments may not be the species that spread disease. These failures shifted attention toward intracellular tau lowering and toward antibodies against the mid-domain and microtubule-binding region (MTBR).

Is a tau medication ready for use?

No. As of October 2026, no tau-targeting therapy is approved by the FDA or any major regulator, and none can be prescribed outside a clinical trial. The most advanced candidate is diranersen, which has completed Phase 2.

Diranersen (BIIB080): biomarker effects and a preliminary clinical signal

Diranersen is an antisense oligonucleotide developed by Ionis and Biogen. It binds MAPT messenger RNA and triggers its degradation, lowering production of all tau forms. It is given by lumbar puncture (intrathecal injection) every three or six months. In the earlier Phase 1b study, it lowered cerebrospinal fluid (CSF) tau by up to about 60% and reduced tau PET signal (Mummery et al., Nature Medicine 2023).

The Phase 2 CELIA trial enrolled 416 adults aged 50 to 80 (mean age 68) with amyloid-confirmed mild cognitive impairment or mild dementia due to Alzheimer’s disease; none had received anti-amyloid therapy. About 69% were APOE ε4 carriers, including 23% homozygotes. The prespecified primary endpoint was a dose-response relationship on CDR-SB at 76 weeks. Results were presented at the Alzheimer’s Association International Conference (AAIC) in London on July 14, 2026.

Source: Biogen press release and AAIC 2026 presentation, July 14, 2026 (sponsor-reported; not peer reviewed). All figures are secondary comparisons with placebo from a trial that missed its primary endpoint; most differences were nominally significant only. ADCS-ADL-MCI showed no separation from placebo.

Three biomarker and safety observations are encouraging. First, CSF total tau fell by 50% to 65% from baseline at every dose. Second, in the tau PET substudy (n = 131), tau PET signal decreased from baseline in all evaluated brain regions at every dose, evidence of target engagement. Third, ARIA has not emerged as an apparent safety signal in the reported CELIA results, consistent with the drug’s mechanism. The full safety profile, including the effects of repeated intrathecal administration, remains to be established. The most frequent adverse events reported were procedural pain, post–lumbar puncture syndrome, and transient confusional states, most of which resolved within a week.

Four caveats are equally important for a careful reader:

• The primary endpoint was missed. CELIA’s prespecified primary analysis tested whether higher doses produce greater slowing on CDR-SB; they did not. The clinical findings above are secondary and require confirmation.
• The dose relationship was inverse. Higher doses lowered tau more but were associated with less clinical slowing. Possible explanations include chance, differences between dose groups, or a nonlinear treatment effect. An optimal level of tau reduction is a hypothesis that Phase 3 will need to test, not a conclusion of this trial.
• Function did not clearly improve. The standalone ADCS-ADL-MCI scale showed no separation from placebo at 18 months, although the functional domains of CDR-SB favored diranersen.
• The data are not yet peer reviewed. Results have been presented at a scientific meeting and in sponsor communications but, at the time of writing, have not appeared in a peer-reviewed journal. A complete, dose-by-dose adverse-event analysis has not yet been published.

Who these results apply to. CELIA enrolled people with MCI or mild dementia due to Alzheimer’s disease who had not previously received anti-amyloid therapy. Its results cannot presently be extended to asymptomatic APOE ε4 carriers, to moderate or advanced dementia, or to diranersen combined or sequenced with an anti-amyloid antibody such as lecanemab.

Other tau candidates

• Etalanetug (E2814, Eisai) is an MTBR-directed antibody with FDA Fast Track designation. It is being tested with lecanemab in people with dominantly inherited Alzheimer’s disease in the DIAN-TU Tau NexGen Phase 2/3 trial, with primary completion expected in April 2028. The trial is an important test of combination therapy, but its dominantly inherited population is not interchangeable with typical late-onset, APOE ε4-associated disease.
• Bepranemab (UCB) missed its primary endpoint (CDR-SB at 80 weeks) in the 466-person TOGETHER trial but slowed tau PET accumulation by 33% to 58% versus placebo (Maguire et al., AAIC 2025 abstract) and slowed decline on the key secondary cognitive measure, ADAS-Cog14, by 21% to 25% (UCB, CTAD 2024). In a subgroup analysis that UCB described as post hoc, participants who had high baseline tau and were APOE ε4 carriers showed no benefit from high-dose bepranemab on almost all clinical endpoints, whereas participants with low baseline tau or without ε4 showed larger effects. Subgroup findings of this kind are hypothesis-generating; they do not establish that ε4 carriers cannot benefit from tau-directed therapy.
• Posdinemab (Johnson & Johnson) was stopped after the Phase 2b Auτonomy trial failed to slow clinical decline.
• Hydromethylthionine (HMTM, TauRx) is an oral tau-aggregation inhibitor that is not approved. Its UK marketing authorisation application has been under review by the Medicines and Healthcare products Regulatory Agency (MHRA) since 2024; in October 2025 the MHRA requested further information, and as of July 2026 the company reported that it was still awaiting a decision. In the peer-reviewed report of the Phase 3 LUCIDITY trial, HMTM did not differ significantly from the control arm on its co-primary clinical endpoints (ADAS-Cog and ADCS-ADL) at 52 weeks (Wischik et al., J Prev Alzheimers Dis 2026). The control arm received a low dose of a related compound (methylthioninium chloride) to mask urine discoloration, and the sponsor attributes the null result to symptomatic activity in that arm. Claims of benefit rest on delayed-start analyses and sponsor-authored comparisons with external control data (Schelter et al., J Prev Alzheimers Dis 2026), which cannot substitute for a randomized placebo comparison.

For now, the only way to receive a tau-targeting therapy is through a clinical trial. Patients can search ClinicalTrials.gov or the Alzheimer’s Association TrialMatch service with their physician.

What scientists are saying about the near future

The expert response has been cautious optimism: tau lowering now has biomarker evidence of target engagement and a preliminary clinical signal, but Phase 3 must confirm it. Four themes recur in published commentary and conference discussion.

1. A first randomized clinical signal. Lead investigator Cath Mummery of University College London said the results suggest that “reducing tau pathology can translate into clinically meaningful benefit.” Laura Nisenbaum of the Alzheimer’s Drug Discovery Foundation described them as “the first data from a randomized trial” to show a tau-targeting drug producing both biomarker effects and a clinical signal. These are interpretations of secondary findings from a trial that missed its primary endpoint; independent confirmation is still required.

2. Dose is the central open question. The Foundation highlighted “the dose-response discordance between the biomarker and clinical endpoints” and said that identifying the optimal level of tau reduction will be central to designing Phase 3.

3. Combination therapy is a likely future direction. Erik Roberson of the University of Alabama at Birmingham, whose laboratory showed that tau reduction protects mouse models, expects that “many patients will end up on combination therapies targeting both amyloid and tau,” and stressed that diranersen is “a Phase 2 result… not an approved drug.” Whether diranersen’s first Phase 3 trial will test it alone or alongside anti-amyloid therapy has not been detailed publicly in the sources reviewed here.

4. The timeline is years, not months. Biogen plans to launch a larger trial in 2027. Allowing for enrollment, 18 months of treatment and regulatory review, an approved tau therapy before about 2030 would be unusually fast. This is my estimate, not a company projection.

What this means for APOE ε4 carriers

The tau story matters especially for APOE ε4 carriers, for three reasons.

• APOE ε4 amplifies tau. Human imaging shows that ε4 carriers accumulate more tau than non-carriers at similar amyloid levels (Therriault et al., JAMA Neurology 2020), and ε4 worsens tau-mediated neurodegeneration in experimental models (Shi et al., Nature 2017).
• ARIA has not been an apparent signal with tau lowering so far. Because ARIA risk with current anti-amyloid antibodies is highest in ε4 homozygotes, a therapy without ARIA would be particularly valuable for this group. ARIA has not emerged as an apparent safety signal in the reported CELIA results, but long-term safety has not been established. Some researchers have proposed that APOE ε4 homozygosity represents a distinct genetic form of Alzheimer’s disease (Fortea et al., Nature Medicine 2024). That proposal concerns homozygotes rather than all ε4 carriers, and it remains a research framework, not a basis for distinct tau-treatment recommendations.
• Benefit is not guaranteed. In a post hoc subgroup analysis of the bepranemab trial, ε4 carriers with high baseline tau showed no apparent benefit from high-dose treatment (UCB, CTAD 2024). About 69% of CELIA participants were ε4 carriers (23% homozygotes), but results by APOE genotype have not yet been reported in detail and should be a priority question for Phase 3.

Blood tests such as plasma p-tau217 and newer tangle-specific markers such as MTBR-tau243 are promising tools for identifying people at the right biological stage for tau trials, although they are not yet used to select patients for routine tau treatment. Earlier intervention, before tau spreads widely, is a shared research goal.

Bottom line

For the full landscape of approved and emerging options, see Alzheimer’s Treatments and Emerging Therapies: A Guide for APOE ε4 Carriers.

References

Peer-reviewed publications

1. van Dyck CH, et al. Lecanemab in early Alzheimer’s disease. N Engl J Med. 2023;388:9-21.
2. Sims JR, et al. Donanemab in early symptomatic Alzheimer disease: the TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA. 2023;330:512-527.
3. Ossenkoppele R, et al. Tau PET patterns mirror clinical and neuroanatomical variability in Alzheimer’s disease. Brain. 2016;139:1551-1567.
4. Bejanin A, et al. Tau pathology and neurodegeneration contribute to cognitive impairment in Alzheimer’s disease. Brain. 2017;140:3286-3300.
5. Ossenkoppele R, et al. Accuracy of tau positron emission tomography as a prognostic marker in preclinical and prodromal Alzheimer disease: a head-to-head comparison against amyloid positron emission tomography and magnetic resonance imaging. JAMA Neurol. 2021;78:961-971. doi:10.1001/jamaneurol.2021.1858
6. Kapasi A, DeCarli C, Schneider JA. Impact of multiple pathologies on the threshold for clinically overt dementia. Acta Neuropathol. 2017;134:171-186.
7. Mummery CJ, et al. Tau-targeting antisense oligonucleotide MAPTRx in mild Alzheimer’s disease: a phase 1b, randomized, placebo-controlled trial. Nat Med. 2023;29:1437-1447.
8. Teng E, et al. Safety and efficacy of semorinemab in individuals with prodromal to mild Alzheimer disease. JAMA Neurol. 2022;79:758-767.
9. Shulman M, et al. TANGO: a placebo-controlled randomized phase 2 study of efficacy and safety of the anti-tau monoclonal antibody gosuranemab in early Alzheimer’s disease. Nat Aging. 2023;3:1591-1601. doi:10.1038/s43587-023-00523-w
10. Florian H, et al. Tilavonemab in early Alzheimer’s disease: results from a phase 2, randomized, double-blind study. Brain. 2023;146:2275-2284.
11. Therriault J, et al. Association of apolipoprotein E ε4 with medial temporal tau independent of amyloid-β. JAMA Neurol. 2020;77:470-479.
12. Shi Y, et al. ApoE4 markedly exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. Nature. 2017;549:523-527.
13. Fortea J, et al. APOE4 homozygosity represents a distinct genetic form of Alzheimer’s disease. Nat Med. 2024;30:1284-1291.
14. Maguire RP, Byrnes W, Barton ME, et al. Tau positron emission tomography results from TOGETHER, a double-blind, placebo-controlled Phase II study of bepranemab in prodromal–mild Alzheimer’s disease. Alzheimers Dement. 2025;21(Suppl 5):e099279 (AAIC 2025 conference abstract). doi:10.1002/alz70859_099279
15. Schneider LS, Llibre-Guerra JJ, Clifford DB, et al. DIAN-TU-001 Trial (Tau NexGen) rationale and enrollment experience. Alzheimers Dement. 2025;21(Suppl 5):e105298 (AAIC 2025 conference abstract). doi:10.1002/alz70859_105298
16. Cummings JL, Zhou Y, Yang Y, et al. Alzheimer’s disease drug development pipeline: 2026. Alzheimers Dement (N Y). 2026;12(2):e70251. doi:10.1002/trc2.70251
17. Fleisher AS, Munsie L, Mancini M, et al. Ceperognastat in early symptomatic Alzheimer disease: a randomized clinical trial. JAMA. 2026;336(12):1017-1028. doi:10.1001/jama.2026.12768
18. Wischik CM, Stefanacci R, Bentham P, et al. Clinical, imaging and blood biomarker outcomes in a Phase 3 clinical trial of tau aggregation inhibitor hydromethylthionine mesylate in mild cognitive impairment and mild to moderate dementia due to Alzheimer’s disease. J Prev Alzheimers Dis. 2026;13(3):100480. doi:10.1016/j.tjpad.2026.100480
19. Schelter BO, Shiells H, Lo S, et al. Assessment of clinical and neuroimaging efficacy of treatment targeting tau pathology in mild cognitive impairment and mild to moderate Alzheimer’s disease with hydromethylthionine mesylate using external control data. J Prev Alzheimers Dis. 2026;13(6):100560. doi:10.1016/j.tjpad.2026.100560

Conference data and reports (not yet peer reviewed)

• Biogen. Biogen presents Phase 2 CELIA data at AAIC demonstrating meaningful clinical outcomes and robust tau reduction with diranersen in early Alzheimer’s disease [press release]. GlobeNewswire; July 14, 2026. globenewswire.com
• UCB. UCB presents encouraging data on bepranemab in early Alzheimer’s disease in Phase 2a study at CTAD 2024 [press release]. October 31, 2024. ucb.com
• Medscape. First evidence that targeting tau may slow cognitive decline. July 21, 2026. medscape.com
• Alzheimer’s Association. Phase 2 results from the CELIA clinical trial of the tau-targeting therapy diranersen [news release]. July 14, 2026. alz.org
• Alzheimer’s Drug Discovery Foundation. Diranersen results at AAIC provide first evidence that targeting tau may slow cognitive decline [statement]. July 14, 2026. alzdiscovery.org
• University of Alabama at Birmingham, Department of Neurology. Landmark Alzheimer’s trial shows tau-lowering therapy slows cognitive decline. July 15, 2026. uab.edu
• Boston Globe. What Biogen’s anti-tau Alzheimer’s drug means for treatment. July 17, 2026 (source of the Adam Boxer quotation). bostonglobe.com
• Fierce Biotech. J&J’s Alzheimer’s hope fades as anti-tau antibody posdinemab flops in Phase 2. Accessed October 2026. fiercebiotech.com
• TauRx Therapeutics. Frequently asked questions for medical professionals (HMTM regulatory status and publications). Accessed October 2026. taurx.com
• ClinicalTrials.gov. Dominantly Inherited Alzheimer Network Trial (DIAN-TU-001) master protocol. NCT01760005. Accessed October 2026. clinicaltrials.gov/study/NCT01760005

This article is for educational purposes only and does not constitute medical advice, diagnosis, or treatment. It is intended to supplement, not replace, the physician–patient relationship. Always consult a qualified healthcare provider regarding any medical condition or treatment decision.