Patients with early-stage Alzheimer’s disease saw their cognitive decline slow by 26 percent over roughly 76 weeks when treated with an experimental drug that targets tau protein, according to results from a mid-stage clinical trial. The drug, known as diranersen or BIIB080, was tested in the phase 2 CELIA study on adults aged 50 to 80 with mild cognitive impairment or mild dementia due to Alzheimer’s. The findings, presented at the Alzheimer’s Association International Conference in July 2026, represent the first time a tau-targeting therapy has shown both a measurable reduction in tau pathology and a corresponding cognitive benefit in this patient population.
Why a tau-targeting Alzheimer’s drug changes the research calculus
For more than two decades, Alzheimer’s drug development concentrated almost exclusively on clearing amyloid-beta plaques from the brain. Approved amyloid-targeting antibodies such as lecanemab have shown modest clinical benefits, but their effect sizes have left many clinicians and patients wanting more. The CELIA trial results shift attention to tau, the other hallmark protein of Alzheimer’s, which forms tangled fibers inside neurons and tracks more closely with the severity of cognitive symptoms than amyloid does.
The 26 percent slowing was measured on the CDR-SB scale, a standard tool that clinicians use to assess memory, orientation, judgment, and daily functioning. A key question the data raise is whether the cognitive benefit stems directly from tau clearance or from some other mechanism. If the benefit is truly driven by tau reduction, then individual patients who showed the largest drops in cerebrospinal fluid or PET-measured tau at week 76 should also show the greatest cognitive improvement in any future patient-level re-analysis of the CELIA dataset. That correlation has not yet been publicly demonstrated, and confirming it would strengthen the biological case for tau-targeted treatment far beyond what a single topline number can establish.
For the roughly 6.9 million Americans living with Alzheimer’s, a second viable drug target could eventually mean combination therapies that attack the disease from two directions. That prospect is still years away from clinical reality, but the CELIA data give it a concrete scientific foundation that did not exist before this trial reported out.
What the CELIA trial data show and who produced them
The trial is formally registered under the identifier NCT05399888 and carries the full title “A Study to Learn About the Safety of BIIB080 Injections and Whether They Can Improve Symptoms of Participants With Mild Cognitive Impairment Due to Alzheimer’s Disease (AD) or Mild AD Dementia Between 50 to 80 Years of Age.” Its registry record history shows the study was first posted in May 2022, and the prespecified endpoints, including CDR-SB, have remained consistent since that original version.
Cath Mummery, a neurologist at University College London, presented the early results at AAIC. According to a UCL news release, Mummery described the findings as the first study to show reduction in tau pathology alongside a cognitive benefit in patients with early Alzheimer’s disease. Her involvement reflects the trial’s international academic footprint; the UCL Queen Square Institute of Neurology and the NIHR University College London Hospitals Biomedical Research Centre both supported the research infrastructure.
The drug itself, described in a preclinical report, is an antisense oligonucleotide delivered by intrathecal injection, meaning it is administered directly into the spinal fluid. Unlike monoclonal antibodies that clear amyloid from outside neurons, antisense oligonucleotides work inside cells by binding to messenger RNA and reducing the production of a target protein. In this case, BIIB080 reduces the production of tau protein itself, aiming to prevent new tangles from forming rather than clearing existing ones.
The study population included participants with mild cognitive impairment due to Alzheimer’s or mild Alzheimer’s dementia, a relatively narrow window in the disease’s progression. Enrolling patients at this stage is deliberate: too early and cognitive decline is difficult to measure over a trial’s duration, too late and neuronal damage may be irreversible regardless of treatment. By focusing on this early symptomatic phase, the CELIA investigators sought a balance between measurable decline and remaining brain resilience.
Within this group, patients were randomized to receive diranersen or placebo over approximately a year and a half, with periodic assessments of cognition, daily functioning, and biomarkers. The topline result that drew attention at AAIC was the 26 percent slowing in CDR-SB decline compared with placebo, accompanied by reductions in tau markers in cerebrospinal fluid and brain imaging. Those parallel changes are what make the signal biologically persuasive, even before full data release.
Gaps in the evidence and what to watch for next
The 26 percent figure comes from topline results, not a full peer-reviewed publication with patient-level data. Several pieces of information that researchers and clinicians need to evaluate the drug’s real-world promise have not yet been released publicly.
First, the complete statistical analysis plan and individual patient CDR-SB trajectories are not available beyond the conference presentation summary. Without those data, independent statisticians cannot assess whether the benefit was evenly distributed across the study population or concentrated in a subgroup. A drug that works well for 30 percent of patients and does nothing for the rest would still produce a positive average effect, but it would raise complex questions about how to select the right candidates in everyday practice.
Second, durability remains unknown. The CELIA trial followed patients for roughly 76 weeks, which is long enough to detect a difference in decline but not long enough to know whether the curves continue to diverge, plateau, or converge over time. Longer-term extension data, if collected and shared, will be essential to understand whether diranersen meaningfully delays progression to more severe stages of dementia.
Safety is another unresolved issue. Intrathecal administration requires lumbar punctures at regular intervals, which can be uncomfortable and carry procedural risks. Antisense oligonucleotides can also provoke inflammation or off-target effects in the nervous system. While the conference report did not flag major safety red flags, a detailed accounting of adverse events, discontinuation rates, and any dose-related toxicities will be critical for regulators and prescribers.
There is also the question of how tau-targeting fits alongside existing amyloid therapies. The CELIA study enrolled patients who met criteria for Alzheimer’s disease, but the extent to which participants had previously received or were concurrently taking amyloid-lowering drugs has not been fully described. Future trials may need to test diranersen on top of standard amyloid therapy to see whether the benefits are additive, redundant, or potentially synergistic.
Regulatory pathways will depend heavily on the completeness and consistency of the forthcoming data. A single phase 2 trial, even a positive one, is rarely sufficient for full approval in a complex neurodegenerative condition. Larger phase 3 studies with diverse populations, longer follow-up, and more granular outcome measures will almost certainly be required. Those trials will also need to address practical questions, such as optimal dosing intervals, the feasibility of repeated lumbar punctures in older adults, and how to monitor tau levels in routine care.
For patients and families, the CELIA results offer cautious optimism rather than an immediate change in treatment options. A 26 percent slowing of decline does not reverse dementia, but if confirmed and sustained, it could translate into extra months or years of preserved independence in daily life. The possibility that tau could join amyloid as a validated therapeutic target also opens the door to combination regimens, much as cancer care often relies on multiple drugs that hit different pathways.
For the field, the most important next steps are transparent data sharing and rigorous peer review. Detailed publication of the CELIA trial, including biomarker correlations and subgroup analyses, will allow the broader scientific community to test the robustness of the findings and refine hypotheses about who benefits most. Only then will it be clear whether diranersen marks the beginning of a new era of tau-directed therapy or a promising but limited advance in a still-evolving treatment landscape.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
