GSK Bet $700M on a Brain Theory. It Just Lost Twice.

TD Cowen analysts noted that the failure effectively ends the GSK-Alector collaboration entirely. Both partnered candidates are now dead. That $700 million upfront payment? Gone. The $1.5 billion in milestones? Never coming.

The Brain's Immune System: A Beautiful Idea That Won't Cooperate

To understand why this matters beyond one partnership, you need to understand what these drugs were trying to do.

Your brain has its own immune cells called microglia. Think of them as the brain's cleanup crew: they patrol for damage, gobble up toxic protein clumps, and maintain order. In Alzheimer's patients, this crew seems to malfunction. They stop clearing amyloid plaques effectively. They become inflamed rather than helpful.

The theory behind Alector's pipeline was simple: wake up the cleanup crew. One approach (the GSK partnership) focused on raising progranulin levels to support neuronal health. Another approach targeted a receptor called TREM2, which acts like an "on switch" for microglia, telling them to surround plaques and eat them.

Alector tested that TREM2 idea separately with a drug called AL002, partnered with AbbVie. In the Phase 2 INVOKE-2 trial (381 patients, results in late 2024), AL002 showed strong signs of engaging its target. Microglia were clearly responding. But clinical progression? Unchanged. Patients didn't get better. The drug hit the lock but couldn't open the door.

So now the scorecard reads: progranulin elevation failed twice, TREM2 activation failed once. Three separate trials. Three misses. All targeting the brain's immune system.

Why This Keeps Happening

Alzheimer's drug development is where good ideas go to die. The disease is devastatingly complex, and what works in mouse brains rarely translates to human ones.

The microglial approach faces a particularly thorny problem: timing. Research suggests TREM2 benefits might depend heavily on disease stage. Too early, and you might disrupt normal processes. Too late, and the damage is irreversible. It's like trying to catch a wave; you need perfect positioning, and we still don't know where that is.

There's also the question of whether engaging a target is the same as fixing the disease. AL002 proved it could flip the microglial switch. Nivisnebart presumably raised progranulin levels. But translating biological activity into slower cognitive decline remains an enormous gap that nobody has bridged with these mechanisms.

What's Left Standing

Alector isn't dead yet. The company has roughly $256 million in cash, enough to fund operations through 2027. It's pivoting toward earlier-stage assets, including an anti-amyloid candidate called AL137 and a blood-brain barrier platform that could attract new partnerships.

For GSK, neuroscience was never a core priority (that honor goes to oncology, respiratory, and HIV). But the company has been quietly building an early-stage neuro portfolio through deals with ABL Bio (£38.5 million upfront for blood-brain barrier shuttle technology), Muna Therapeutics (target discovery from post-mortem brain samples), and Vesalius Therapeutics.

The Alector write-off stings, but GSK isn't abandoning the space entirely. It presented data at AAIC 2025 linking its shingles vaccine to potentially reduced dementia risk, which could open a prevention angle that bypasses the treatment graveyard altogether.

The Bigger Picture

Neuroscience's most-watched therapeutic hypothesis just took its biggest hit. The idea that you can modulate the brain's immune system to fight Alzheimer's remains scientifically plausible; the genetics are real, the biology is compelling. But three consecutive Phase 2 failures make it clear that plausibility and clinical proof are separated by a canyon.

For patients, the wait continues. For investors, the lesson is familiar: in Alzheimer's, the only thing more expensive than a failed drug is the next one you try.

Alector says it will present full PROGRESS-AD data at a future medical meeting. Until then, a $700 million partnership has produced exactly zero approved drugs and one very expensive education in how little we still understand about the aging brain.

A Patient Died in a Schizophrenia Trial. Now the FDA Wants Answers.

The FDA slammed the brakes on Newron Pharmaceuticals' Phase 3 schizophrenia trial after a patient died at a non-U.S. site. The company says the drug isn't to blame, but the agency wants proof. For one of the few novel approaches to treatment-resistant schizophrenia, the clock is ticking.