

Five patient dropouts in a 28-person Huntington's disease trial cost Annexon Biosciences a third of its market value in a single day. The selloff exposes just how fragile investor confidence remains in CNS drug development, even when the underlying data tells a more nuanced story.
Five out of 28 patients stopped taking an experimental Huntington's disease drug. That's it. That's what vaporized roughly a third of Annexon Biosciences' market cap.
The stock cratered in a brutal punishment for a company that, by most clinical standards, had a pretty reasonable safety profile. But in the world of central nervous system (CNS) drug development, where failure is the default and investor nerves are perpetually frayed, "pretty reasonable" doesn't always cut it.
Welcome to biotech, where five people leaving a study can cost you billions.
Annexon's drug, ANX005, is an antibody that blocks a protein called C1q. Think of C1q as the trigger on an immune system chain reaction called the classical complement pathway. In Huntington's disease, this chain reaction goes haywire, chewing up brain cells like a Pac-Man that forgot to stop eating. ANX005 is designed to take the trigger out of Pac-Man's hands.
The Phase 2 trial enrolled 28 patients with early-stage Huntington's and gave them IV infusions every two weeks for about six months. Of those 28, five dropped out before the finish line. Two left for reasons unrelated to the drug: one got COVID, one simply withdrew consent.
The other three are where things get interesting.
All three experienced autoimmune-related side effects: one developed lupus, another got a form of noninfectious pneumonia, and a third had a blood disorder called hemolytic anemia. Every one of these cases resolved after the drug was stopped. And crucially, all three patients had elevated levels of antinuclear antibodies (ANA) at baseline, a pre-existing marker that suggests their immune systems were already a bit trigger-happy before ANX005 entered the picture.
Patients with normal ANA levels? Zero autoimmune problems. Zero discontinuations.
On paper, this looks manageable. The company identified a clear risk factor, has already added enhanced ANA screening to future trials, and plans to adjust dosing for at-risk patients. No one died. No serious infections occurred. Most patients tolerated the drug well enough to finish treatment.

A tiny UK biotech just posted positive Phase 2a data for a lung disease drug that takes the opposite approach to Insmed's billion-dollar Brinsupri. The safety looks clean, the dosing schedule is quarterly, and the efficacy signal is intriguing. Is there room for two players in NCFB?


Join thousands of biotech professionals who start their day with our free, daily briefing.
But here's the uncomfortable math: a 17.8% dropout rate in a 28-person trial is noisy data. In a small study, every patient who leaves punches a hole in your statistical power. Investors aren't just worried about safety; they're worried about whether Annexon can run a bigger, placebo-controlled Phase 2/3 trial without the same retention headaches.
Analysts were mixed but mostly constructive. H.C. Wainwright cut its price target from $30 to $20 but kept a Buy rating. Firms like Needham, Wells Fargo, Jefferies, and Bank of America largely maintained positive ratings, while consensus targets still implied significant upside. The selloff, in other words, was driven more by fear than by fundamentals.
That said, fear is a perfectly rational emotion in CNS drug development.
To understand why investors hit the eject button so fast, you need to understand what Huntington's disease does to drug companies. It destroys them.
The overall success rate for drugs entering clinical trials for HD is roughly 3.5%. Three and a half percent. You'd have better odds at a blackjack table. The attrition rate from Phase 2 to Phase 3 alone is over 80%.
The wreckage is everywhere. Roche and Ionis spent years developing tominersen, an antisense therapy that actually lowered the toxic huntingtin protein in patients' spinal fluid. The Phase 3 trial was stopped early because the drug made some patients worse, not better. More frequent dosing led to worsened motor and cognitive outcomes compared to placebo. Biochemical success, clinical disaster.
Wave Life Sciences tried a more targeted approach with allele-specific antisense drugs. They barely moved the needle on protein levels and quietly shut the programs down. Novartis pulled the plug on branaplam after patients developed nerve damage. UniQure's gene therapy, AMT-130, had to pause enrollment after neurological complications from brain surgery.
This isn't a field where investors give you the benefit of the doubt. It's a field where they've been burned so many times that smoke from a campfire looks like a forest fire.
Buried beneath the dropout headlines, the clinical data told a more encouraging story. Among the 23 patients who completed treatment, ANX005 achieved complete blockade of the classical complement pathway in both blood and cerebrospinal fluid. That's strong target engagement.
On the efficacy side (which was exploratory, not the main goal of this trial), more than half of participants showed stabilization or improvement on motor, cognitive, and functional measures. Among patients with excess complement activity at baseline, roughly 75% improved on the composite disease rating scale.
Those numbers aren't proof the drug works. The trial was open-label, meaning everyone knew they were getting the drug, and there was no placebo group for comparison. But for a Phase 2 safety study in a disease where everything seems to fail, they're the kind of signals that keep a program alive.
Annexon isn't alone in chasing Huntington's. More than 20 companies are actively developing therapies, and the competitive landscape is getting busier by the quarter. UniQure's AMT-130 gene therapy has Breakthrough Therapy designation and is targeting a regulatory filing in Q3 2026. Alnylam started first-in-human dosing of an RNA interference drug in late 2024. Skyhawk Therapeutics is pushing an oral pill that could lower huntingtin protein without a needle or a neurosurgeon.
The market for HD therapies is projected to grow at roughly 14% annually through 2036, driven almost entirely by the hope that someone, somewhere, will crack the disease-modification code. Only symptomatic drugs (think: treating the shaking, not the cause) have made it to market so far. The disease-modifying cupboard remains bare.
Annexon's approach is different from most competitors. Instead of trying to lower the toxic huntingtin protein directly, it targets neuroinflammation through complement inhibition. That could make ANX005 complementary to (pun intended) HTT-lowering therapies rather than a direct rival.
Annexon has cash to operate into the second half of 2026, thanks to a $125 million financing round in late 2023. The company's pipeline extends well beyond Huntington's: ANX005 is in a pivotal Phase 3 trial for Guillain-Barré syndrome, and a related eye drug (ANX007) is headed into Phase 3 for geographic atrophy.
For the HD program specifically, Annexon is planning a new Phase 2/3 placebo-controlled trial with improved autoimmune screening baked into the protocol. That trial will need to answer the question this Phase 2 couldn't: does blocking complement actually slow disease progression in a rigorous, blinded setting?
The stock drop stings. But the real test isn't what the market did on one day. It's what the data do in the next trial. In Huntington's disease, where a 3.5% success rate is the norm, just getting to the next trial is a small victory in itself.
Biogen closed its $5.6 billion Apellis acquisition and immediately killed most of the company's research programs, keeping only the two products already making money. It's a pattern that keeps repeating in big pharma M&A, and it says a lot about what acquirers actually value.