Roche has moved its investigational anti-amyloid antibody trontinemab into Alzheimer's prevention. The design of PrevenTRON, a randomised, double-blind, placebo-controlled phase 3 study, was presented for the first time at the Alzheimer's Association International Conference in London on 14 July 2026, in a Featured Research Session flagged in advance by the company on 7 July. As presented, the trial plans to enrol roughly 1,600 cognitively unimpaired people aged 55 to 80 who have biomarker evidence of Alzheimer's pathology and are judged at high risk of clinical decline, with Roche's Elecsys plasma pTau217 test used to identify candidates. The primary endpoint is time to clinical progression, defined as a confirmed Clinical Dementia Rating global score above zero; dosing starts with an induction period of 3.6 mg/kg every four weeks before less frequent maintenance. The study carries the ClinicalTrials.gov identifier NCT07717411; this newspaper was unable to retrieve the full registry record at the time of writing, and the design figures above are those presented at the conference.
Trontinemab is a bispecific antibody engineered to cross the blood brain barrier using transferrin receptor 1 transport. The human evidence behind it is biomarker evidence, not clinical benefit. In the phase 1b/2a Brainshuttle AD study, in people who already had symptoms, mean amyloid plaque removal of 107 centiloids was reported after 28 weeks at 3.6 mg/kg in 12 participants, as summarised in the published conference abstract for the company's symptomatic phase 3 programme. Those two symptomatic trials, TRONTIER 1 and TRONTIER 2, began in September and November 2025 and measure change on the Clinical Dementia Rating Sum of Boxes at 18 months, with primary completion estimated for June 2028 on the registry. Neither has reported.
PrevenTRON joins two other prevention programmes in people without symptoms. AHEAD 3-45, Eisai's lecanemab study, measures change on the Preclinical Alzheimer Cognitive Composite 5 at 216 weeks in its A45 arm and amyloid on PET at 216 weeks in its A3 arm. Eli Lilly's TRAILBLAZER-ALZ 3 used plasma p-tau217 to select participants, screening 63,124 people to randomise 2,196 with a time to event primary outcome, an approach close to the one Roche has chosen. Roche's TRAVELLER prescreening pathway is intended to do the same job of routing people towards trials on the basis of a blood test and brief cognitive assessment rather than an upfront amyloid scan.
The counterpoint is straightforward: a trial design is a plan, not a result. No anti-amyloid antibody has yet been shown to delay or prevent the onset of symptoms in people who do not have them, and amyloid-related imaging abnormalities remain a known risk of the drug class that has to be weighed differently in people who feel well. Whether earlier treatment changes the course of the disease is precisely the question PrevenTRON has been built to answer, and it has not answered it.