TREM2: A Promising but Elusive Target in Neurodegeneration

Alector’s anti-TREM2 antibody just failed in an Alzheimer’s Phase 2 trial, while Novartis bets big on their TREM2 antibody (VHB937) for ALS and Alzheimer's. Can other contenders seize the lead?

Neurodegenerative diseases like Alzheimer’s disease (AD) and ALS are relentless, leaving patients and families grasping for hope. Enter TREM2, a target that has sparked immense interest, and no small amount of frustration, in the scientific and biotech communities.

Despite its promise, translating TREM2 biology into clinical success has proven elusive. But why? One reason is that, let’s be honest, we do not fully understand what TREM2 does or how it does it. We know some of it, but not the whole story. Let’s explore what we know about how TREM2 works, the drugs in the pipeline targeting it, and some of the lessons learned so far.

What is TREM2?

TREM2 (Triggering Receptor Expressed on Myeloid Cells 2) is a transmembrane glycoprotein receptor primarily found on myeloid cells throughout the body, including microglia, the brain’s resident immune cells. Microglia act as housekeepers, clearing debris, responding to inflammation, and supporting neuronal health.

The known endogenous ligands of TREM2 include phospholipids, bacterial lipopolysaccharide (LPS), and ApoE.

While TREM2 is a single receptor, its transmembrane region associates with the adaptor proteins DNAX-activation protein 12 (DAP12; also known as TYRO protein tyrosine kinase-binding protein) and DAP10 (also known as haematopoietic cell signal transducer).

TREM2 surface expression declines on activation of myeloid cells with Toll-like receptor ligands, such as LPS, or with inflammatory cytokines, such as IFNγ. This phenomenon is believed to be caused by the cleavage of membrane-bound TREM2 by proteases, that release soluble TREM2 (sTREM2).

The exact mechanism of why this cleavage takes place is unknown. Some believe that cleaving TREM2 (and thus increasing sTREM2) is a mechanism to attenuate neuroinflammatory signals. Inhibiting this cleavage, and thus reducing levels of released sTREM2, would be expected to increase TREM2-related signals.

The role of sTREM2 and the intracellular TREM2 in the pathogenesis of AD and other neurodegenerative disorders is not well understood.

TREM2 genetic variants associated with neurodegeneration

Variants in the TREM2 gene are linked to an increased risk of several neurodegenerative disorders, including AD, amyotrophic lateral sclerosis (ALS), and frontotemporal dementia (FTD).

The most extensively studied variant is R47H, a partial loss-of-function mutation that raises AD risk by three- to five-fold, comparable to the effect of APOE4, the strongest genetic risk factor for sporadic AD. Homozygous carriers face a significantly higher risk compared to heterozygous.

The R47H variant impairs TREM2’s ability to form complexes with APOE, LDL, and other lipoproteins, disrupting microglial clearance of amyloid-beta. This leads to amyloid plaque accumulation, a hallmark of AD.

Notably, the odds ratio for AD risk with R47H homozygosity is a staggering 97.1 (95% CI, 23.5–401.1). Its allelic frequency varies across populations, from nearly absent in some groups to 0.7% among Icelanders and 1.4% in Ashkenazi Jews.

R47H homozygotes with AD tend to experience an earlier disease onset, with a mean diagnosis age 6.4 years earlier (95% CI, 2.7–10.1) compared to AD patients who do not carry TREM2 variants.

Other variants also contribute to neurodegenerative risk. For example:

• H157Y, a rare coding variant, has been linked to a significantly increased AD risk in the Han Chinese population.

• Variants such as R62H, T96K, and D87N confer lower AD risks and are population-specific.

• Loss-of-function mutations in TREM2 are known to cause Nasu-Hakola disease, a rare disorder marked by early-onset dementia and bone cysts.

Consequently and very quickly, a new star on the therapeutic horizon of AD was born, with several pharmaceutical companies developing TREM2-modulating antibodies, antisense oligonucleotides, and small molecules.

Based on this, maintaining TREM2 activity, specifically transmembrane TREM2 rather than soluble TREM2 (sTREM2), appears essential, right?

So we just need to activate TREM2, right?

Given that AD-associated TREM2 variants cause loss-of-function effects, pharmacological agents that promote TREM2 expression, signalling and activity might be beneficial, at least in patients carrying TREM2 variants. Right?

Not so fast.

Chronic overactivation of TREM2 could potentially trigger excessive neuroinflammatory responses, which might worsen neurodegeneration, as some animal studies suggest. This indicates there may be a 'Goldilocks' level of TREM2 activation, where too little or too much could both be harmful.

Additionally, research has shown that eliminating TREM2 in earlier stages of AD could reduce amyloid plaque burden, yet the opposite effect is observed in later stages, where TREM2 deletion exacerbates plaque accumulation. This highlights the importance of timing in TREM2-targeted interventions, as the stage of AD may significantly influence therapeutic outcomes.

Despite these unknowns, many companies have ventured into clinical development with their TREM2-targeted candidates. Let’s go over the TREM2 pipeline.

The TREM2 Pipeline: Hits and Misses

Denali DNL919

The first company taking a TREM2-targeting molecule into clinical development was Denali who, in collaboration with Takeda, developed DNL919, a anti-TREM2 antibody conjugated to an anti-TfR-antibody designed to cross the blood-brain barrier efficiently and activate TREM2.

I discussed antibody-drug conjugates (including DNL919) in an earlier entry in August 2024:

In January 2022, the FDA placed a clinical hold on DNL919 due to undisclosed preclinical toxicology findings. Denali then shifted its focus to Europe, initiating a Phase 1 clinical trial in the Netherlands in July 2022 (NCT05450549). This trial aimed to evaluate the safety, tolerability, pharmacokinetics, and target engagement of single ascending doses of DNL919 in 80 healthy participants.

Denali announced on August 8, 2023, that it discontinued the development of DNL919. The decision followed findings that, while DNL919 affected multiple microglial biomarkers such as CSF1R, SPP1, IL1RA, IP10, MIP1b, and MCP-1, indicating target engagement (however, we don’t know the direction of the change, i.e., increase or decrease in these biomarkers), it also caused moderate, reversible anemia (a known side effect of many TfR-mediated shuttles).

So, as far as we know, the development was halted because of the TfR-related problems, not necessarily due to targeting TREM2.

Alector/AbbVie: AL002

AL002, an agonistic monoclonal antibody that binds to TREM2 and activates its signaling and prevents TREM2’s cleavage. was initially evaluated in the Phase 1 INVOKE trial, which included a single-ascending-dose phase with 69 healthy volunteers and a multiple-ascending-dose phase with 30 AD patients divided into three dose cohorts.

The antibody was well tolerated, with no serious adverse events reported up to 12 weeks post-dosing. Treatment demonstrated dose-dependent effects on two key pharmacodynamic markers in cerebrospinal fluid (CSF):

1. sTREM2 levels decreased within 2 days of treatment and remained suppressed for 30 days in the highest dose cohort, indicating reduced cleavage and, possibly, enhanced TREM2 activity.

2. CSF1R levels increased, further supporting the activation of TREM2 by the antibody.

The trial concluded in December 2020, and detailed results were recently published, shedding light on the potential mechanism of action and safety profile of AL002. (Long et al., 2024).

Based on this evidence of target engagement, Alector proceeded with the Phase 2 trial (INVOKE-2), enrolling 328 participants with early AD to evaluate monthly infusions of AL002 at 15, 40, or 60 mg/kg versus placebo. The primary endpoint was the Clinical Dementia Rating-Sum of Boxes (CDR-SB), with the treatment phase running 48 to 96 weeks.

In late 2021, reports surfaced of amyloid-related imaging abnormalities (ARIA) in ApoE4 homozygotes, causing serious neurological events like seizures and mental status changes. This led Alector to halt dosing in ApoE4 homozygotes and exclude them from the trial. After this adjustment, the frequency of ARIA dropped, though ARIA-E occurred in 29% of participants and ARIA-H in 27%, primarily in ApoE4 carriers. Most ARIA cases were asymptomatic, resolving within four months of onset. Recruitment concluded in September 2023, with the trial completing in September 2024.

On November 25, 2024, Alector announced that INVOKE-2 failed to meet its primary endpoint or any secondary endpoints, including AD biomarkers and amyloid PET (press release). Despite evidence of target engagement and microglia activation, no amyloid plaque clerance or clinical benefits were observed. The company halted the trial’s open-label extension.

AL002’s failure suggests that TREM2 activation is not useful to clear amyloid or to exert clinical benefits in early AD. There are remaining questions, though:

Was the TREM2 activation too weak to have a meaningful impact? Could stronger activation have yielded better results? Or perhaps the opposite is true: was the TREM2 activation overly intense, leading to harmful effects (perhaps not detected by the set of biomarkers used by Alector)? These are critical questions that remain unanswered in the quest to optimize TREM2-targeting therapies.

Vigil Neuroscience

Vigil is developing two molecules targeting TREM2.

Iluzanebart (VGL101): a human monoclonal antibody targeting TREM2 that activates its signaling, currently in Phase 2 trials for adult-onset leukoencephalopahy with axonal spheroids and pigmented glia (ALSP). A rare and fatal neurodegenerative disease, ALSP is caused by mutations in the gene for the glial protein Colony Stimulating Factor 1 Receptor (CSF1R), which shares a common downstream signaling pathway with TREM2. The rationale for VGL101 is that TREM2 activation should compensate for loss of CSFR1 function and rescue glia function.

Vigil licensed VGL101 from Amgen, which had published preclinical work on TREM2 antibodies that activated signaling and abrogated defects in microglia carrying the rare R47H TREM2 gene variant, a risk factor for AD (Cheng et al., 2018).

The mechanism of action of iluzanebart is therefore the same as AL002, albeit the target population is different.

In 2023, Vigil reported Phase 1 data from 136 healthy volunteers, demonstrating that its TREM2-targeting therapy was safe and well-tolerated at the highest doses tested. Repeat dosing showed clear signs of target engagement, with dose-dependent reductions in CSF sTREM2 and increases in soluble CSF1R.

Building on these results, a Phase 2 (IGNITE) open-label trial began in December 2022, enrolling 20 patients with ALSP and CSF1R mutations. At the American Academy of Neurology meeting in April 2024, Vigil shared interim results: in six patients treated with 20 mg/kg for six months, biomarkers of microglial activation (soluble CSF1R and osteopontin) increased, while NfL levels showed a downward trend. There were no treatment-related serious adverse events or discontinuations, though the company did not disclose data on sTREM2 levels at this stage.

VG-3927 is Vigil’s second TREM2-targeting asset and the first small-molecule TREM2 agonist in clinical development. In October 2023, the first participant was dosed in a Phase 1 placebo-controlled, single- and multiple-ascending-dose safety study in healthy volunteers.

At the July 2024 AAIC, Vigil presented single-dose data, reporting that VG-3927 was well-tolerated, with only mild to moderate adverse events that resolved spontaneously and no serious adverse events. The pharmacokinetics supported daily dosing, and significant reductions in CSF sTREM2 at the three highest doses confirmed target engagement.

In a press release, Vigil stated that early results from 2 completed multiple-dose cohorts, where participants received daily dosing for two weeks, also showed safety and lower CSF sTREM2 levels. The company plans to include a cohort of AD patients in the study to evaluate biomarker responses and intends to release full trial data in early 2025.

Vigil aims to improve the treatment paradigm by developing an oral small-molecule TREM2 agonist, VG-3927, which could offer a significant advantage over IV-administered antibodies by enhancing patient convenience and accessibility.

Novartis VHB937

Novartis’ VHB937 is the latest contender entering clinical trials. There is not much public information about VHB937 other than it is an stabilizing and activating antibody targeting TREM2 by increasing TREM2 cell surface expression by reduction of shedding.

While no Phase 1 trial appeared in clinical trial registries, Novartis just announced that VHB937 recently initiated a Phase 2 trial in ALS (N=225) and plans to launch a Phase 2 AD trial in 2025.

Most information about VHB937 comes from their preclinical data presented at scientific and medical meetings including AD/PD 2024 in Lisbon or the American Academy of Neurology meeting.

Challenges and Safety Concerns

• ARIA Without Amyloid Clearance: AL002’s INVOKE-2 trial revealed ARIA-like changes despite no reduction in brain amyloid plaque. This raises questions about the relationship between TREM2 activation and vascular safety.

• Clinical Holds: Vigil faced temporary clinical holds by the FDA on both its antibody and small-molecule programs, underscoring the safety complexities of modulating microglia.

• Mechanistic Ambiguities: How do we balance TREM2 activation to enhance clearance of pathological proteins without tipping the scale toward excessive inflammation or systemic side effects? Is CSF sTREM2 an appopriate biomarker of target engagement of microglial transmembrane TREM2?

• Combination with Approved Anti-Amyloid Antibodies: With the recent approval of anti-amyloid antibodies lecanemab and donanemab, the role of TREM2-targeted therapies needs to be redefined. Could combination be synergistic, or sequential treatment be implemented? Could sTREM2 be used a biomarker of microglial activation?

Conclusions

TREM2 represents a tantalizing target for neurodegeneration, bridging innate immunity and brain health. But as Denali, Alector, and others have shown, targeting TREM2 comes with significant challenges: mechanistically, clinical, and safety-related. Success may lie in precision: selecting the right patients, fine-tuning doses, and better understanding the downstream effects of microglial activation.

As we await results from the next wave of trials, one thing is clear: the story of TREM2 is far from over. Will it deliver on its promise, or will it remain an elusive target leaving companies TREMbling? Only time will tell.

Readers - How do you think some of these challenges related to TREM2 could be solved? Please share your thoughts, questions, and experiences in the comments below!

---

Thanks for reading. If you found this entry interesting, please, share it so that others can receive more like this in their inbox.

---

Views expressed here are my own and not necessarily those of my employer. All data mentioned and discussed are publicly available.