Feature|Articles|October 30, 2026

Examining Trontinemab in Early Alzheimer Disease: The Phase 3 TRONTIER 1 and 2 Trials

Listen

0:00 / 0:00

Key Takeaways

Zeleciment basivarsen is designed to knock down nuclear DMPK RNA via TfR1-mediated delivery to skeletal muscle and CNS, addressing upstream spliceopathy rather than symptom-specific pathways.
HARMONIA randomizes ~150 genetically confirmed DM1 patients (CTG repeat >100) to IV 6.8 mg/kg Q8W versus placebo, with a 48-week controlled period plus 24-week all-active extension.
Week-49 change in 5xSTS is the primary endpoint, supported by vHOT, QMT, 10-Meter Walk/Run, MDHI, clinician/patient-reported outcomes, and broad exploratory CNS domain measures.
Key exclusions include congenital DM1, major recent surgery (pacemaker/ICD allowed), and recent GLP-1/incretin use; baseline ambulation and independent 5xSTS performance are required.
Phase 1/2 ACHIEVE MAD data at 6.8 mg/kg Q8W showed early DMPK RNA knockdown, splicing improvement, sustained functional gains through 12 months, and no serious treatment-related TEAEs reported.

The phase 3 TRONTIER 1 and 2 trials will test whether trontinemab's faster brain entry and rapid amyloid clearance translate into slower clinical decline in early Alzheimer disease.

Welcome to NeurologyLive's Clinical Trial in Focus. Every month, an ongoing clinical trial in the landscape of neurology is featured, highlighting the design of the study, the targeted patient population, the enrollment criteria, the primary and secondary end points, and its potential implications for clinical care. This month's spotlight is the twin phase 3 TRONTIER 1 (NCT07169578) and TRONTIER 2 (NCT07170150) trials assessing trontinemab (Roche), an investigational bispecific antiamyloid antibody engineered to cross the blood-brain barrier, in individuals with early symptomatic Alzheimer disease (AD).^1-3

Mechanism of Action

Trontinemab combines an amyloid-β–binding antibody with a transferrin receptor 1 (TfR1) shuttle module, an approach Roche calls Brainshuttle.⁴ The molecule uses a bispecific "2+1" format, pairing a bivalent antiamyloid antibody with a single, monovalent, low-affinity Fab fragment that binds TfR1.^1,5 TfR1 normally ferries iron across the blood-brain barrier; by engaging it, trontinemab is transported into the brain by receptor-mediated transcytosis rather than relying on passive diffusion.⁴

The amyloid-binding portion is derived from gantenerumab, Roche's earlier antiamyloid antibody.⁵ That lineage is relevant to interpreting TRONTIER. In the phase 3 GRADUATE I (NCT03444870) and II (NCT03443973) trials, subcutaneous gantenerumab (Roche) lowered amyloid burden vs placebo at 116 weeks but did not significantly slow clinical decline on CDR-SB, and only 28.0% and 26.8% of treated participants reached amyloid-negative status.⁶ ARIA with edema (ARIA-E) occurred in 24.9% of gantenerumab-treated participants.

The therapeutic hypothesis behind trontinemab is that more efficient brain delivery achieves faster, deeper, and more uniform plaque removal at lower systemic doses than conventional antibodies, which may in turn reduce the risk of ARIA.^4,5 TRONTIER is designed to test whether that pharmacologic advantage yields a clinically meaningful slowing of disease progression.

“Alzheimer’s disease represents one of the greatest challenges in healthcare today and tackling it requires early detection and effective therapeutics,” Levi Garraway, MD, PhD, chief medical officer and head of global product development at Roche, said in a statement.⁴ “Trontinemab is designed to target a key driver of Alzheimer’s disease biology more effectively in the brain. Combining new treatment avenues with advanced diagnostics may enable earlier and potentially more effective intervention. With plans for Phase III trials in both early symptomatic and preclinical Alzheimer’s disease, we are advancing science with the goal of delaying —and ultimately preventing—progression of this devastating condition.”

Study Design

First presented at the 2025 Alzheimer's Association International Conference (AAIC), held July 27-30, in Toronto, Canada, by Janice Smith, PhD, global development lead for trontinemab at Roche, TRONTIER 1 and 2 are identically designed, global, randomized, double-blind, placebo-controlled, parallel-group phase 3 studies.^1,4,5 They evaluate the efficacy, safety, tolerability, pharmacokinetics, and pharmacodynamics of intravenous trontinemab in participants with early AD and confirmed amyloid pathology. Each trial aims to enroll approximately 800 participants, for a combined total of approximately 1600 across 18 countries.^5,7 Dosing began in September 2025.⁷

Participants are randomly assigned to trontinemab 3.6 mg/kg or placebo by intravenous infusion. The 72-week treatment period is divided into 2 phases: a 6-month induction phase with infusions every 4 weeks (weeks 0, 4, 8, 12, 16, 20, and 24), followed by a maintenance phase with less frequent dosing, reported as every 3 months.^7,8 An open-label extension is offered after the double-blind period. The induction-maintenance structure reflects the rapid plaque clearance observed in phase 1b/2a, with the maintenance phase intended to sustain amyloid control while reducing infusion burden; data from the Brainshuttle AD open-label extension informed this regimen.⁵

The primary end point is the change from baseline in CDR-SB after 18 months of treatment.⁴ Secondary end points include change in the Mini-Mental State Examination (MMSE) through week 72, time to increase in the Clinical Dementia Rating–Global Score (CDR-GS), the Alzheimer's Disease Assessment Scale–Cognitive Subscale 13 (ADAS-Cog13), and the Alzheimer's Disease Cooperative Study–Activities of Daily Living (ADCS-ADL), along with safety, pharmacokinetics, and the incidence of antidrug antibodies.^1,2,7 Pharmacodynamic effects will be assessed with amyloid and tau positron emission tomography (PET), magnetic resonance imaging, and fluid biomarkers. Amyloid PET is scheduled at screening, at week 12 or 28, and at week 72, and tau PET at baseline and week 72.⁸

Recruitment is supported by TRAVELLER, a master prescreening study that evaluates potential participants with the International Shopping List Test and Roche's Elecsys plasma phosphorylated tau 217 (p-tau217) assay before referral to TRONTIER 1, TRONTIER 2, or Roche's preclinical PrevenTRON trial, without requiring prior amyloid confirmation.^4,5 A blood-based prescreening step is intended to reduce screening burden at clinical sites.

Inclusion Criteria

Eligible participants are adults aged 50 to 90 years who weigh 150 kg or less.⁸ They must have probable AD dementia or mild cognitive impairment due to AD, corresponding to clinical stage III or IV of the Alzheimer's clinical syndrome, with a screening MMSE score of 22 or higher and a CDR-GS of 0.5 or 1.0.^2,3 These thresholds restrict enrollment to the early symptomatic stage in which antiamyloid therapy has shown clinical benefit and in which change on CDR-SB can be detected over 18 months.

Amyloid pathology must be confirmed on amyloid PET; a cerebrospinal fluid (CSF) phosphorylated tau 181/amyloid-β42 ratio may be used as an alternative if PET is unavailable.³ Participants must also have a participant- or informant-reported history of gradual-onset, progressive cognitive decline over the year before screening, and a Repeatable Battery for the Assessment of Neuropsychological Status Delayed Memory Index (RBANS DMI) score of 85 or lower, which helps ensure objective episodic memory impairment at baseline.

Additional requirements include adequate visual and auditory acuity to complete neuropsychological testing, the availability of a study partner, and the willingness and ability to complete all study procedures, including MRI, clinical genotyping, and PET imaging or CSF collection as applicable. Genotyping supports ARIA risk assessment, because apolipoprotein E ε4 (APOE ε4) carrier status is a known risk factor for ARIA with antiamyloid antibodies.

Exclusion Criteria

Individuals are excluded if they have evidence of any condition other than AD that may affect cognition, which helps ensure that observed decline reflects AD pathology rather than a competing diagnosis. A history or presence of clinically significant cerebrovascular disease is also exclusionary.³Participants with a history of severe, clinically significant central nervous system trauma, defined as causing a persistent neurologic deficit or structural brain damage, or a history or presence of a clinically significant intracranial mass, are ineligible.

Roche added superficial siderosis and probable cerebral amyloid angiopathy at baseline to the exclusion criteria for ongoing and future trontinemab trials after a fatal cerebral macrohemorrhage in a participant with both findings in the expanded Brainshuttle AD cohort.⁷ These exclusions align with broader efforts to reduce hemorrhagic risk with amyloid-lowering therapy. The publicly posted registry summaries list key criteria only; additional protocol-defined criteria apply.

Prior Data

The rationale for TRONTIER comes from the ongoing phase 1b/2a Brainshuttle AD study (NCT04639050). Part 1 used a staggered, parallel-group design with 4 dose cohorts (0.2, 0.6, 1.8, and 3.6 mg/kg). At least 10 participants per cohort were randomly assigned 4:1 to trontinemab or placebo, given intravenously every 4 weeks for 7 doses.⁹

In an interim analysis of the Part 1 3.6 mg/kg cohort, mean amyloid PET reduction was 89 centiloids at 12 weeks (n = 13) and 107 centiloids at 28 weeks (n = 12). By week 25 in the same cohort, CSF total tau, phosphorylated tau 181, and neurogranin were reduced by 30%, 34%, and 29%, respectively. In the expanded 3.6-mg/kg cohort, 91% of participants (49 of 54) fell below the 24-centiloid amyloid positivity threshold after 28 weeks, and 72% (39 of 54) reached levels below 11 centiloids.⁵ Investigators also reported changes in plasma p-tau217.

Across the 1.8 and 3.6-mg/kg cohorts combined, ARIA-E occurred in fewer than 5% of participants (4 of 149), and all cases were radiographically mild.^5,10 For context, ARIA-E rates of 13% with lecanemab and 24% with donanemab have been reported in their pivotal programs, although cross-trial comparisons are limited by differences in populations, dosing, and MRI monitoring.⁵ One participant in the 1.8-mg/kg cohort with baseline superficial siderosis and probable cerebral amyloid angiopathy died of a cerebral macrohemorrhage, prompting the exclusion criteria changes described above.⁷

At AAIC 2026, Roche reported 1-year open-label extension data showing continued amyloid reduction from baseline. Infusion-related reactions occurred in 7 of 81 participants (8.6%) assigned to the every-12-week maintenance schedule.¹⁰ These phase 1b/2a data are biomarker and safety findings from a small, early-phase study and were not powered to detect clinical benefit; TRONTIER is the first test of whether trontinemab's amyloid clearance slows cognitive and functional decline.

Trontinemab's development also includes PrevenTRON, a phase 3 trial of approximately 1600 cognitively unimpaired individuals aged 55 to 80 years with elevated plasma p-tau217, which uses time to clinical progression as its primary end point.⁵

Click here to view more Clinical Trial in Focus.