Roche has announced its plans to start a pivotal Phase III trial for Alzheimer’s therapy trontinemab, slated to later this year.
The Swiss pharmaceutical giant made the announcement at the AD/PD 2025 International Conference on Alzheimer’s and Parkinson’s Diseases which took place from 1-5 April in Vienna, Austria.
Although Roche did not disclose the details of the Phase III trial design. Luka Kulic, Vice President and Head of Early Development in Neuroscience & Rare Diseases at Roche, noted in a press briefing at the conference that the study will not evaluate the therapy as a combination at this stage. However, he did not rule out the possibility of evaluating trontinemab as a combination treatment for Alzhiemer’s in further studies.
Phase II trial data for trontinemab
Trontinemab is an investigational Brainshuttle bispecific 2+1 amyloid-beta antibody specifically engineered for enhanced access to the brain to enable rapid reduction of amyloid in people with Alzheimer’s disease.
Roche presented preliminary results from the 114 participants enrolled in the Phase Ib/IIa Brainshuttle AD study (NCT04639050). The study evaluated multiple doses (1.8mg/kg and 3.6mg/kg) administered intravenously every four weeks, where rapid and significant amyloid lowering was observed.
The data indicates mean reductions from baseline in amyloid positron emission tomography (PET) of 78 centiloid at the 1.8mg/kg dose and 96 centiloid at the 3.6mg/kg dose after 28 weeks. Notably, the 3.6mg/kg dose demonstrated around 60% amyloid reduction by eight weeks (Day 50) post two doses of treatment.
By the conclusion of the 28-week treatment period, 81% of participants in the 3.6mg/kg group fell below the amyloid positivity threshold of 24 centiloid. The presentation also pointed out that there were no nonresponders to treatment observed at the 28-week follow-up.
In addition to the primary amyloid PET endpoint, a second presentation on interim biomarker results for trontinemab showed improvements across several fluid biomarkers in both cerebrospinal fluid (CSF) and plasma.
A CSF analysis at 25 weeks with the 3.6mg/kg dose revealed an 84% increase in the Aβ42/40 ratio, a 29% decrease in CSF phosphorylated tau 181 (p-tau181), a 21% reduction in neurogranin, and a 22% drop in total tau. Plasma biomarkers mirrored these trends, with especially notable reductions of 36% in p-tau181 and 51% in p-tau217 at the 3.6mg/kg dose after six months of treatment.
How does trontinemab compare to others in the field?
Kulic noted that only indirect comparisons could be made at this stage, owing to the absence of head-to-head trial data. Professor Jeffery Cummings, another panelist at the Rochemedia briefing, highlighted trontinemab’s ability to lower the amyloid to the threshold of undetectable “more rapidly than other molecules”.
Low incidence of ARIAs
Amyloid‐related imaging abnormalities (ARIA) are commonly found in those patients with Alzheimer’s disease, who are being treated with amyloid‐targeted monoclonal antibody treatments.
The risk of ARIAs was a major safety issue for Eli Lilly’s Alzheimer’s disease med Kisunla (donanemab). In March, the European Medicines Agency’s Committee for Medicinal Products for Human Use (CHMP) recommended against a marketing authorization for the drug’s use in early symptomatic Alzheimer’s disease, citing that he benefits of the treatment “were not large enough” to outweigh the risk of ARIAs.
Roche’s trontinemab showed a favorable safety profile in Phase Ib/IIa study, with a low incidence of ARIA. Only three cases of ARIA-E (oedema/effusion) were reported in the 1.8mg/kg cohort (3.9%), with no cases observed in the higher 3.6mg/kg dose group as of the November 2024 data cutoff.
Kulic noted that the low incidence of ARIA could potentially be due to the Brainshuttle technology. Noting that compared to standard antibodies which are usually administered systematically and in large doses to cross the blood-brain barrier reach the brain. Brainshuttle technology bypasses vascular amyloid pathology, whose involvement usually occurs in cases of systematic antibodies and might contribute to the development of ARIAs.
