Sabirnetug Phase 2 Failure Ends Sole Asset as Approved Amyloids Raise Bar

Acumen Discontinues Alzheimer's Drug After Phase 2 Failure

Acumen Pharmaceuticals has discontinued the development of its sole pipeline asset, sabirnetug (ACU193), an investigational amyloid therapy for early Alzheimer’s disease, following its failure to meet primary and key secondary endpoints in the Phase 2 ALTITUDE-AD study. The monoclonal antibody showed no meaningful difference compared to placebo in cognitive and functional improvement at 18 months. This news led to an 80% plummet in the company's stock, prompting Acumen to implement cash-saving strategies and explore strategic options, including leveraging its enhanced brain delivery technology.

• The Phase 2 ALTITUDE-AD study, evaluating sabirnetug in patients with early Alzheimer’s, did not support its hypothesis, showing no meaningful difference from placebo in cognitive and functional improvement over 18 months. The drug also failed to meet any key secondary endpoints, indicating a lack of efficacy in the target population.
• Sabirnetug was generally well-tolerated in the study, though amyloid-related imaging abnormalities-edema (ARIA-E), a treatment-related brain swelling, were observed in approximately 10% of participants receiving the drug. This safety profile, while generally acceptable, did not outweigh the lack of efficacy.
• Following the trial's failure, Acumen has terminated all clinical development of sabirnetug, including an open-label extension. The company plans to implement cash conservation measures and explore strategic options, including its enhanced brain delivery technology, which it previously partnered with JCR Pharmaceuticals to combine with J-Brain Cargo for next-gen CNS therapeutics.

Sabirnetug's ALTITUDE-AD Failure: Efficacy and Safety Outcomes

Recent studies in early Alzheimer's disease have evaluated amyloid-targeting monoclonal antibodies for slowing cognitive and functional decline. These trials measured global, cognitive, and biomarker outcomes while monitoring amyloid-related imaging abnormalities and infusion-related reactions.

• Clarity AD – lecanemab 10 mg/kg biweekly: In the Asian region cohort of 294 individuals with early Alzheimer's disease, there was a slowing of decline with lecanemab in the CDR-SB at 18 months compared to placebo (adjusted mean difference: -0.349; 95 % confidence intervals: -0.773, 0.076; 24 % slowing of decline). Safety in Asian subjects showed ARIA-H (lecanemab:14.4 %; placebo:16.2 %), ARIA-E (lecanemab:6.2 %; placebo:1.4 %), and infusion-related reactions (lecanemab:12.3 %; placebo:1.4 %).

• Clarity AD Core + OLE safety – lecanemab 10 mg/kg biweekly: In 1612 participants with at least one dose of lecanemab, the most common adverse events in the lecanemab group (> 10%) were infusion-related reactions (24.5%), ARIA with hemosiderin deposits (ARIA-H) microhemorrhages (16.0%), COVID-19 (14.7%), ARIA with edema (ARIA-E; 13.6%), and headache (10.3%). ARIA-E generally occurred within 3-6 months of treatment, was more common in ApoE e4 carriers (16.8%) and most common in ApoE ε4 homozygous participants (34.5%).

• EMERGE and ENGAGE – aducanumab high-dose 10-mg/kg: Among 1029 patients with at least 1 postbaseline MRI scan, 425 of 1029 patients (41.3%) experienced ARIA, with serious cases occurring in 14 patients (1.4%). ARIA-edema (ARIA-E) was the most common adverse event (362 of 1029 [35.2%]), and 263 initial events (72.7%) occurred within the first 8 doses of aducanumab; 94 participants (26.0%) with an event exhibited symptoms.

• Study 201 Phase 2 core and OLE – lecanemab: There was a low incidence of ARIA-E (<10%), with <3% symptomatic cases. ARIA-E was generally asymptomatic, mild-to-moderate in severity, and occurred early (<3 months), with incidence higher in apolipoprotein E4 (ApoE4) homozygous carriers.

The Competitive Landscape for Amyloid Beta Oligomer Targeting

Sabirnetug (ACU193) is a humanized monoclonal antibody, raised against globular species of soluble AβO, that has over 650-fold greater binding affinity for AβOs over monomers and appears to have relatively little binding to amyloid plaque. Other monoclonal antibodies trialled for early symptomatic Alzheimer disease target soluble aggregated amyloid-beta species including oligomers and protofibrils.
Lecanemab is a humanized IgG1 monoclonal antibody binding with high affinity to protofibrils of amyloid-beta (Aβ) protein and preferentially targets soluble aggregated Aβ species (oligomers, protofibrils), with activity at insoluble fibrils.

For the intervention models of aducanumab and crenezumab trials, The knowledge base does not have sufficient information on this aspect.

Overcoming Challenges in Alzheimer's Disease Drug Development

Traditional therapies offer only symptomatic relief without modifying disease progression, while recently approved anti-amyloid monoclonal antibodies, donanemab and lecanemab, represent a shift toward disease-modifying treatment in patients with mild cognitive impairment (MCI) or mild AD with confirmed amyloid pathology. Despite statistically significant slowing of cognitive and functional decline, antibody effects were below the threshold of clinically meaningful change during the period they were studied, and use requires careful patient selection, biomarker confirmation, and adherence to safety protocols.

• Modest and uncertain clinical benefit: Antibodies as a drug class attenuated worsening on CDR-SB and ADAS-Cog by very small effect sizes, with cognitive effects of small effect sizes despite biomarker improvement. Lecanemab reduced clinical worsening by 27% when compared to placebo (mean difference of -0.45; 95% CI: -0.67 to -0.23), which fell short of the minimal clinically significant difference that is frequently mentioned for this patient population. Donanemab modestly reduced cognitive decline versus placebo: ADAS-Cog13 (MD, -1.86), CDR-SB (MD, -0.36), MMSE (MD, 0.64), and iADRS (MD, 3.19), with long-term outcomes and comparative effectiveness remaining uncertain.

• High burden of ARIA and infusion-related events: Antibodies increased risk of ARIA-E and ARIA-H by a very large and moderate effect size, respectively, with lecanemab-treated participants having higher incidence of amyloid-related imaging abnormalities such as edema and microhemorrhages (about 21.5%) than the placebo group (about 9%). Pooled ARIA incidence was 19% (95% CI: 16%-23%) and pooled symptomatic ARIA incidence was 3% (95% CI: 2%-4%), IRRs occurred in 26% (95% CI: 19%-34%), discontinuation due to AEs was 8% (95% CI: 5%-11%), and discontinuation due to ARIA occurring in 5% (95% CI: 3%-7%) of patients in RWE studies. In SCarlet RoAD OLE with up-titration to 1200 mg every 4 weeks (Q4W), 47 (30.5%) participants had an amyloid-related imaging abnormalities - edema (ARIA-E) MRI finding, and 51 (33.1%) had an ARIA - hemorrhage MRI finding over median duration of treatment of 2.9 years (152.9 weeks).

• APOE ε4 genotype-driven risk requiring screening and monitoring: Lecanemab-related ARIA demonstrates a clear ApoE4 gene-dose effect, with RR 1.45 for heterozygotes and 3.54 for homozygotes vs noncarriers, supporting routine ApoE4 genotyping before treatment. Risk of ARIA-E with donanemab was markedly increased (RR, 12.39), especially among APOE ε4 homozygotes, with infusion reactions (RR, 11.90) and discontinuations (RR, 3.22) also more frequent. Most ARIA-E findings were asymptomatic and manageable by MRI monitoring and dose intervention, but prompt recognition and management of ARIA are critical for improving treatment tolerability.

• Single-target approaches in multi-etiology disease: Multi-etiology dementia was the norm rather than the exception, with trial-engaged groups altogether having a mean of 2.54 pathologic features/person, whereas the MCI/D trial-engaged group had a mean of 3.2 pathologic features/person. Pure on-target pathology was only seen in 10, 23, and 29% of the engaged participants of V, MCI/D and P trials respectively, as all interventions were designed to target only a single pathologic feature.

• Limited symptomatic efficacy and tolerability of established agents: Donepezil demonstrated small but significant improvement on the primary measure of cognition but there was no change on the primary measure of global function, with the dual primary efficacy endpoint not reached and little change from baseline in CDR-SB. More donepezil-treated subjects (18.4%) discontinued treatment due to adverse events than placebo-treated subjects (8.3%).

• Implementation, access and system capacity constraints: Current interventions face limitations in accessibility, cost, and long-term efficacy, with limited infusion capacity, lack of reimbursement, and patient concentration in tertiary centers highlighting the need for referral networks. The rapid increase in treatment demand exceeded outpatient infusion capacity, resulting in progressively longer waiting times — up to 20 weeks 9 months after program initiation — while IRRs occurred in 41 (37%) under active surveillance and ARIA developed in 20 (19%): ARIA-E/H in 6 (5.7%) and isolated ARIA-H in 14 (13.3%).

Sabirnetug's Discontinuation: A Reality Check for Aβ Oligomer Targeting

The recent discontinuation of Acumen Pharmaceuticals' sabirnetug (ACU193) following its failure in the Phase 2 ALTITUDE-AD study marks a significant setback in the pursuit of disease-modifying therapies for early Alzheimer's disease. Sabirnetug, a monoclonal antibody designed to selectively target soluble amyloid beta oligomers (AβOs), represented a promising approach, aiming to address the 'most toxic species' of Aβ while potentially mitigating the amyloid-related imaging abnormalities (ARIA) often seen with plaque-clearing agents. Its Phase 1 INTERCEPT-AD study had shown encouraging signs, including dose-dependent target engagement, reductions in CSF pTau181, VAMP2, and neurogranin, and even some amyloid plaque clearance. These biomarker shifts were consistent with engaging disease biology and reaching the central compartment.

However, the Phase 2 results, showing no meaningful difference in cognitive and functional improvement compared to placebo at 18 months, highlight a critical disconnect between biomarker modulation and clinical efficacy. This outcome casts a shadow on the AβO-selective targeting hypothesis as a standalone strategy for clinical benefit, especially when compared to the modest but consistent cognitive slowing observed with agents like lecanemab and donanemab, which primarily target amyloid protofibrils or plaques. The failure underscores the profound complexity of Alzheimer's pathophysiology and the challenge of translating early mechanistic insights into tangible patient outcomes.

For Acumen, the implications are severe, necessitating a strategic re-evaluation and a potential pivot towards leveraging its 'enhanced brain delivery technology' as a platform rather than a product-focused approach. This event serves as a stark reminder of the high-risk nature of AD drug development, particularly for companies with a concentrated pipeline. It also reinforces the need for the broader pharmaceutical industry to critically assess the predictive value of early biomarkers and to consider multi-modal therapeutic strategies that address the multifaceted pathology of AD, including tau pathology, neuroinflammation, and vascular components, beyond just amyloid clearance. The path forward in AD drug development will likely demand even more rigorous clinical validation and a nuanced understanding of how different amyloid species contribute to cognitive decline.