The phase 2 CELIA trial of diranersen (formerly BIIB080), an antisense oligonucleotide given by lumbar puncture to lower production of tau protein, did not meet its primary endpoint, according to the sponsors' own statements and the trial's registry record. CELIA (NCT05399888) randomised 416 people aged 50 to 80 with mild cognitive impairment or mild dementia due to Alzheimer's disease, all with confirmed amyloid pathology and none previously treated with an anti-amyloid drug, to placebo or one of three intrathecal regimens (60 mg every 24 weeks, 115 mg every 24 weeks, 115 mg every 12 weeks) over a 76-week placebo-controlled period. The prespecified primary measure was a dose response for change from baseline to week 76 on the Clinical Dementia Rating Sum of Boxes (CDR-SB). Biogen announced on 14 May 2026 that this endpoint was not met, and said it would take the drug into registrational development anyway. Detailed results were presented at the Alzheimer's Association International Conference in London on 14 July 2026 by Catherine Mummery of University College London.
On biology, the reported effect was large and consistent. Biogen and Ionis report mean reductions in cerebrospinal fluid total tau of roughly 50 to 65 percent from baseline across all three doses, and reductions in tau measured by PET across evaluated brain regions in an imaging substudy of 131 participants, while tau on PET rose in the placebo group. On the clinical measures, the strongest numbers came from the lowest dose: in the 60 mg every six months group (n=60) compared with placebo (n=115), the companies report slowing of decline of 0.54 points, or 26 percent, on CDR-SB, plus 42 percent on ADAS-Cog13, 50 percent on MMSE, 30 percent on modified iADRS and 23 percent on ADCOMS at 18 months. Alzforum's conference report puts the two higher doses at 14 percent and 9 percent slowing on CDR-SB, and states that none of the between-group differences reached statistical significance. On the functional scale ADCS-ADL-MCI, the same report says the 60 mg group showed no effect compared with placebo.
The registry is the document readers can check against the framing. It describes CELIA as a phase 2, randomised, double-blind, placebo-controlled, dose-ranging study, not a registrational one; the trial began in August 2022 and remains active in a long-term extension. Alzforum records that the study was designed for about 735 participants across 54 sites in North America, Europe, Japan and Australia and closed enrolment at 416, which cuts the statistical power available to detect the dose response the primary endpoint required. The only peer-reviewed CELIA publication so far is a baseline characteristics abstract from the Biogen team in Alzheimer's and Dementia, published in December 2025, describing who was enrolled: mean age 68 years, 51 percent female, 60 percent at the mild cognitive impairment stage. The efficacy and safety results themselves exist as company releases and conference slides.
On tolerability, the sponsors describe most adverse events as mild or moderate and non-serious, with procedural pain, post-lumbar puncture syndrome and confusional state the most frequent, most confusional events occurring within days of dosing and resolving within a week. The May topline release also noted a higher incidence of serious adverse events at the highest dose studied. More than 90 percent of participants who completed the placebo-controlled period elected to continue into the extension. Diranersen is investigational and not approved by any regulator; it is available only in trials, and only by intrathecal injection. Independent commentators including the Alzheimer's Drug Discovery Foundation have called the biomarker effects notable while saying the results leave open questions about the right dose and the design of any confirmatory study.