Background

This humanized monoclonal IgG1 antibody was being developed in a partnership between Alector and AbbVie. It binds the microglial receptor TREM2 and activates signaling, increasing phosphorylation of TREM2's downstream effector Syk, and inducing microglia proliferation. AL002a, a variant of AL002 that recognizes mouse TREM2, reportedly doubled the number of CD11b-positive microglia in cortex and hippocampus 72 hours after injection in APP/PS1 mice. Microglia expressed more pro-inflammatory and repair genes, and nearby amyloid deposits were nearly halved (Dec 2016 conference news). The antibody produced similar results in 5XFAD mice after 72 hours.

Repeated injections of 50 mg/kg AL002a into 4-month-old 5XFAD mice for 14 weeks reportedly raised the number of CD11b-positive microglia by 50 percent. The number of microglia around amyloid plaques doubled, and amyloid load was cut in half. Treated mice performed like wild type in the radial arm water maze and novel object recognition tasks (May 2019 conference news).

An anti-human TREM2 antibody, AL002c, was tested in 5XFAD mice carrying human TREM2 genes. Three months of weekly injection with 30 mg/kg AL002c activated microglia, reduced neurotoxicity and inflammatory signaling, and normalized behavior in an elevated maze. Treatment did not alter Aβ plaque load (news on Wang et al., 2020). AL002c is the preclinical variant of AL002.

Findings

In November 2018, AL002 started INVOKE, a Phase 1 trial at six sites in the U.S., Australia, and the U.K. In its single-ascending-dose phase, 69 healthy adults received a single infusion of one of nine doses between 0.003 to 60 mg/kg or placebo; in the multiple-ascending dose phase, 30 AD patients were enrolled in three dose cohorts. Outcomes included safety, tolerability, pharmacokinetics, pharmacodynamics, and immunogenicity. Some results of the single-dose portion of the study have been published (June 2020 news). The antibody was well tolerated, and no serious adverse events were noted up to 12 weeks after dosing. Treatment caused dose-dependent changes in two pharmacodynamic markers in CSF: soluble TREM2 was decreased two days after treatment, while a fragment of CSF1R, a receptor expressed only by microglia in the brain, was increased. The trial was completed in December 2020, and full trial results were subsequently published (Long et al., 2024). The half-life of AL002 in blood was 8 to 9 days, and CSF entry ranged from 0.17 to 0.28 percent after 13 days. The reduction in CSF sTREM2 persisted for 30 days in the highest dose group, and was accompanied by changes in other markers of microglia activity. The multiple dose study in AD patients was terminated early due to the COVID-19 pandemic. Pharmacokinetic and target engagement studies after multiple dosing were completed in non-human primates, and informed dose selection for Phase 2.

In January 2021, the Phase 2 INVOKE-2 study started enrolling 328 people with early AD to compare monthly infusions of 15, 40, or 60 mg/kg of AL002 to placebo on the CDR-SB as a primary outcome. The treatment phase ran for 48 to 96 weeks, through the end of 2023, at 90 sites in North and South America, Europe, Australia, and New Zealand. In late 2021, Alector reported that amyloid-related imaging abnormalities (ARIA) were observed in INVOKE-2, with some serious neurological adverse events in people with two copies of the ApoE4 allele. The company discontinued dosing E4 homozygotes in the trial, and amended the protocol to exclude them (Feb 2022 press release). After exclusion of homozygotes, no additional serious adverse events related to ARIA were reported (Aug 2022 press release, Jan 2023 company presentation, slides 23-32). The company reported more safety data at the July 2023 AAIC in Amsterdam (Aug 2023 conference news). Three serious ARIA-like events in E4 homozygotes included focal neurological deficits, seizures, and mental status changes requiring hospitalization. All resolved with AL002 stoppage and corticosteroid treatment. After E4 homozygotes were excluded from the trial, the overall frequency and severity of ARIA was reduced, although ARIA-E still occurred in 29 percent of participants, and ARIA-H in 27 percent. ARIA occurred more in E4 carriers than non-carriers. It mainly appeared in the first three months of treatment, was asymptomatic, and resolved within 4 months. MRI results from two patients who developed ARIA-H showed signs of microhemorrhage appearing 3 and 5 months after the last AL002 dose. Recruitment for this trial was completed in September 2023. It finished in September 2024. A one-year open label extension was fully enrolled and expected to run through 2025.

On November 25, 2024, Alector announced that INVOKE-2 had failed to meet its primary endpoint of slowing decline on the CDR-SB (press release). Treatment did not improve any secondary endpoints, AD biomarkers, or amyloid PET, despite evidence of sustained target engagement and microglia activation. The company stopped the long-term extension of this trial.

For all trials of AL002, see clinicaltrials.gov.