Background

THN391 is a humanized monoclonal antibody to the blood clotting protein fibrin. THN391 was designed to block fibrin-mediated neuroinflammation without interfering with clotting functions. It recognizes an inflammatory epitope that is present on fibrin, but not its soluble precursor fibrinogen. This epitope activates innate immune cells, including microglia and macrophages, via their complement receptor 3 or 4; the antibody prevents immune activation. The lead clinical applications are Alzheimer’s and diabetic eye disease. The antibody is given by intravenous infusion; in diabetes patients it is injected into the eye.

In Alzheimer’s, vascular disease, and other conditions, a leaky blood-brain barrier allows fibrinogen to enter the brain. There, it is cleaved into fibrin, which deposits in blood vessels and parenchyma and causes chronic inflammation (e.g., Mendiola et al., 2023). Epidemiological and animal studies link fibrin deposits, or high CSF and plasma fibrinogen levels, to neurodegeneration and the risk for dementia (reviewed in Kantor et al., 2023). Genetically, two APP mutations in the Aβ sequence cause hereditary cerebral amyloid angiopathy and stroke, Dutch E693Q and Iowa D694N; both were reported to increase Aβ42 binding affinity to fibrinogen and lead to increased fibrin deposition (Cajamarca et al., 2020).

In AD mice, fibrinogen induced spine elimination and cognitive deficits by complement receptor-mediated microglia activation. In brain tissue from AD patients, focal fibrinogen deposits were associated with loss of dendritic spines, even in areas lacking amyloid plaques (Merlini et al., 2019).

THN391 is a humanized version of the mouse monoclonal antibody 5B8, which was developed to react to the P2 fibrin epitope γ377-395. The antibody has a LALA mutation to reduce Fc effector region. In mouse models of AD, 5B8 entered the CNS and bound to parenchymal fibrin. It reduced microglial activation and neurodegeneration around plaques, without affecting plaque or microglia number (Ryu et al., 2018). The antibody was anti-inflammatory and neuroprotective in animal models of multiple sclerosis, as well.

Findings

In April 2023, Phase 1 began with a single- and multiple-ascending-dose safety study. Conducted in the Netherlands, it enrolled 96 healthy volunteers to receive single doses of 0.3, 1, 3 or 10 mg/kg, multiple doses of 3 mg/kg, or placebo. In April 2025, the company reported top-line results (press release), saying the antibody had been safe, without serious adverse events. Treatment did not affected coagulation, nor induce anti-drug antibodies. Pharmacokinetics were dose-proportional, with a 50-day half-life suitable for monthly dosing. The trial ended in June 2025; Therini plans to present Phase 1 data at the July 2025 Alzheimer’s Association International Conference.

In March 2025, a Phase 1b dose-finding study began enrolling 15 people with early AD. Participants must also have cerebral small-vessel disease and hypertension, Type 2 diabetes, or high blood lipids. They will receive three infusions over two months of antibody or matched placebo. At least three doses are to be tested. The primary outcomes are adverse events and drug concentrations in serum and CSF. Secondary endpoints are antidrug antibodies and blood coagulation parameters. Safety measures will include brain MRIs, electrocardiograms, mental health monitoring, and symptomatic testing. The trial is running at four sites in the Netherlands and U.K. until June 2026.