Phase 1b Trial Shows Short-Lived Anti-PD-L1 Antibody IBC-Ab002 Is Well Tolerated in Early Alzheimer's Disease
Key Insights
IBC-Ab002, an engineered short-lived anti-PD-L1 (search) antibody with a ~4-day half-life, was well tolerated across all tested doses in a phase 1b trial of 40 participants with early symptomatic Alzheimer's disease (search).
Immune-related adverse events occurred in 33.3% of active participants, mostly mild or moderate; no treatment-related serious adverse events or amyloid-related imaging abnormalities (ARIA) were observed.
Exploratory CSF analyses at the highest dose (30 mg/kg) showed directional reductions in neurogranin (search) (−23.1%), total tau (search) (−12.4%), and pTau181 (search) (−11.0%) relative to placebo, though none reached statistical significance.
A first-in-human phase 1b trial of IBC-Ab002, a newly engineered short-lived anti-PD-L1 (search) antibody, has demonstrated an acceptable safety and tolerability profile in participants with early symptomatic Alzheimer's disease (search) (AD), according to results published in Nature Medicine. The study provides the first clinical evidence that intermittent immune checkpoint blockade may offer a novel therapeutic approach for neurodegeneration, distinct from both anti-amyloid immunotherapies and oncology applications of PD-(L)1 inhibitors.
The IBC-01-01 trial (NCT05551741), conducted across 11 sites in the UK, Israel, and the Netherlands between April 2023 and December 2025, enrolled 40 participants across five dose-escalation cohorts. Participants were randomized 3:1 to receive intravenous IBC-Ab002 (1, 3, 6, 15, or 30 mg/kg) or matching placebo, with a single-dose observation period followed by three additional doses administered every 12 weeks over approximately 48 weeks.
Study Population and Design
The overall study population had a mean age of 69.1 years (range: 51–81), was 57.5% female, and predominantly white (92.5%). Based on the Clinical Dementia Rating–Global Score (CDR-GS), 75% of participants had a score of 0.5, consistent with mild cognitive impairment due to AD, while 25% had a score of 1.0, consistent with mild AD dementia. The median Mini-Mental State Examination (MMSE) score at screening was 25 (range: 20–28), and 77.5% were apolipoprotein E (APOE) ε4 carriers. All participants were required to have an amyloid-positive and tau-positive (A+T+) CSF biomarker profile.
Safety and Tolerability
Treatment-emergent adverse events (TEAEs) were reported in all 40 participants, reflecting intensive longitudinal safety monitoring. TEAEs considered related to the study drug were reported in 73.3% of active participants and 60.0% of placebo participants. The most common adverse events among active participants were fatigue (26.7%), headache (23.3%), and infusion-related reactions (20.0%), all of which were grade 1 or 2 in severity.
Three serious adverse events (SAEs) occurred: two in the placebo group (seizure and syncope) and one in an active participant treated with 15 mg/kg (multiple fractures after a mechanical fall). No treatment-related SAEs were observed.
Immune-related adverse events (irAEs), predefined as adverse events of special interest, were observed in 33.3% of active participants and 10.0% of placebo participants (odds ratio 4.5, 95% CI 0.5–40.7). Approximately half of these events were identified through routine laboratory monitoring rather than presenting clinically. Asymptomatic thyroid dysfunction was the most common irAE, observed in four participants across cohorts 2, 3, and 5. Three resolved spontaneously and one developed persistent hypothyroidism managed with thyroid hormone replacement therapy.
The only severe irAE (CTCAE grade 3) was an asymptomatic elevation in liver enzymes in a cohort 4 participant after the third dose, consistent with a mixed hepatocellular and cholestatic pattern with no change in total bilirubin; it resolved spontaneously. This participant discontinued the study drug per protocol-defined stopping rules.
Brain MRI performed at weeks 7 and 48 revealed no new study-drug-related abnormalities, and no cases of amyloid-related imaging abnormalities (ARIA) were observed—a finding consistent with the mechanism of action of IBC-Ab002, which does not target vascular amyloid.
Pharmacokinetics and Target Engagement
IBC-Ab002 was engineered with two types of Fc modifications: one to eliminate antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity, and a second to reduce FcRn binding and accelerate systemic clearance. Noncompartmental and population pharmacokinetic analyses (30 participants, 756 observations) confirmed a short mean elimination half-life of approximately 4 days across all dose cohorts (range: 3.5–4.6 days), with a low volume of distribution (~3.5–4 L) and rapid clearance (~0.7–0.8 L/day).
Serum exposure increased proportionally with dose across the 1–30 mg/kg range, and serum concentrations fell below the lower limit of quantification before each subsequent infusion. No accumulation was observed with repeated every-12-week dosing. At the highest dose (30 mg/kg), mean Cmax was approximately 670 μg/mL.
Peripheral target engagement, measured by PD-L1 (search) receptor occupancy (RO) on circulating T cells, was high (80–100%) across all dose levels when serum concentrations exceeded approximately 1 μg/mL. RO returned to baseline within approximately 2 months after each dose in the two highest dose cohorts, consistent with the intended intermittent exposure profile.
Treatment-emergent anti-drug antibodies (ADAs) were detected in 63.3% of active participants, with no appreciable effect on Cmax or AUC at any dose level.
Peripheral Immune Activation
A single dose of IBC-Ab002 induced a dose-dependent increase in the frequency of circulating PD-1+ICOS+ CD4+ memory T cells, a population representing recently activated T cells released from PD-L1 (search)-mediated suppression. This response peaked approximately 7 days after administration and was transient, returning close to baseline before each subsequent dose at week 12—consistent with the design rationale that intermittent dosing is required to initiate an immune response with each administration.
Exploratory CSF Biomarker Findings
End-of-study lumbar puncture was performed in 19 of 40 participants (47.5%), including five of six active participants receiving the highest dose (30 mg/kg) and six of ten placebo participants. In these participants, directional reductions relative to pooled placebo were observed in neurogranin (search) (−23.1%), total tau (search) (−12.4%), and pTau181 (search) (−11.0%). None of these changes reached statistical significance, as expected given the limited sample size. Changes were directionally consistent across the three markers within each treated participant.
No detectable effects were observed in plasma biomarkers over the 48-week treatment period, which may reflect the narrower dynamic range and slower kinetics of plasma versus CSF biomarkers, dilution of CNS-derived analytes in the peripheral compartment, and limited statistical power.
Cognitive Assessments and Next Steps
Longitudinal cognitive function was assessed using the MMSE and the Cognitive Functional Composite (CFC). No signs of rapid or unexpected cognitive deterioration were observed in any active treatment group; however, the study was designed to evaluate safety and was not powered for clinical efficacy endpoints. An adequately powered evaluation of cognitive endpoints is planned for an upcoming phase 2 study.
Mechanistic Rationale
The approach differs fundamentally from anti-amyloid immunotherapies in that it does not depend on direct amyloid binding or plaque clearance. Preclinical studies suggest that the observed CNS effects reflect an immune-mediated process driven by recruitment of peripheral immune repairing cells that can remove damage-associated molecular patterns and senescent microglia, reducing local inflammation. Because IBC-Ab002 acts primarily through the peripheral immune system, its activity is not expected to depend on antibody penetration into or persistence within the brain.
The combination of a short-lived antibody with intermittent administration is expected to contribute to a favorable safety profile in neurodegeneration. The brief systemic exposure and 12-week dosing interval would be expected to limit the duration and severity of irAEs relative to oncology anti-PD-(L)1 regimens, although this will require confirmation in larger studies.
The safety, pharmacokinetic, and central and peripheral pharmacodynamic signals observed at the highest dose provide a basis for further clinical development of this newly engineered anti-PD-L1 (search) therapy in AD. Because this strategy does not rely on directly targeting a specific proteinopathy, it may have broader applicability across other chronic neurodegenerative conditions in which neuroinflammation contributes to disease progression.