Argenx Had the Best Day and Worst Day of Its Life (At the Same Time)

The core problem is that Sjögren's is a shape-shifter. Some patients deal mostly with brutal dryness in their eyes and mouth. Others battle fatigue and pain. Still others have systemic inflammation attacking their organs. Lumping all these patients into one trial and measuring them with a single ruler is like grading a cooking competition, a dance-off, and a spelling bee with the same scorecard. The endpoints just don't capture what matters for each patient.

Placebo arms also tend to improve substantially in Sjögren's trials, which makes it even harder for an active drug to separate itself from sugar pills. It's a disease that has humbled some of the smartest companies in biotech, and argenx is now the latest name on that wall.

The Celiac Silver Lining

Now for the good news, which got almost completely buried under the Sjögren's wreckage.

argenx also reported positive Phase 2 results for a different drug called FB102 in celiac disease. And the data actually look pretty compelling. The study enrolled 126 adults with confirmed celiac disease who had been symptom-free on a strict gluten-free diet for at least a year. Researchers then deliberately challenged them with gluten for eight weeks (a controlled torture test, essentially) to see whether FB102 could protect their intestines from damage.

The primary endpoint measured something called villus height-to-crypt depth ratio, which is basically a microscopic snapshot of gut health. Healthy intestinal villi look like tall, waving fingers that absorb nutrients. Gluten destroys them in celiac patients, flattening them out. FB102-treated patients had significantly better preservation of their gut architecture versus placebo, hitting statistical significance with a p-value of 0.0176.

Here's what makes FB102 interesting: it's not the same type of drug as Vyvgart. While Vyvgart works by blocking the FcRn receptor (a recycling system that keeps harmful antibodies circulating longer than they should), FB102 is an anti-CD122 antibody. It targets the receptor for IL-2 and IL-15, two immune signals that drive the T-cell attack at the heart of celiac disease. The idea is to calm down the rogue immune cells that treat gluten like a foreign invader, without wiping out the regulatory T-cells that keep your immune system in check.

For a disease where the only current "treatment" is obsessively avoiding gluten (and even that doesn't fully prevent damage for many patients), a drug that actually protects the gut could be transformative. argenx plans to advance FB102 into Phase 3.

One Company, Two Very Different Stories

The market's reaction tells you everything about how investors weigh bad news versus good news. The Sjögren's failure removed a potential growth driver for Vyvgart, argenx's flagship franchise that's already approved in three indications: generalized myasthenia gravis (gMG), chronic inflammatory demyelinating polyneuropathy (CIDP) in Japan (as VYVGART HYTRULO/VYVDURA, a subcutaneous formulation), and primary immune thrombocytopenia (ITP) in Japan. Losing Sjögren's shrinks the total addressable market for the FcRn platform, and that's what analysts zeroed in on.

The celiac data, meanwhile, were viewed as encouraging but too early-stage to compensate. A Phase 2 win is still at least a few years and one more large trial away from an approval. Investors tend to penalize today's failure more than they reward tomorrow's promise, and October 8 was a textbook example.

But zoom out a bit. argenx still has a growing Vyvgart franchise with expanding labels (including a recently broadened gMG indication covering seronegative patients), plus active programs in myositis, lupus nephritis, and systemic sclerosis. The celiac program gives the company a differentiated asset outside the FcRn space. The pipeline isn't broken; one branch just got pruned.

The Bigger Lesson

Autoimmune drug development is brutally hard, and Sjögren's might be the hardest of the hard. The field desperately needs better trial designs: smarter patient selection, more sensitive endpoints, and creative ways to account for a disease that presents differently in almost every person who has it.

For argenx, the path forward means learning everything it can from the UNITY data after the database is locked, doubling down on indications where Vyvgart already works, and betting that FB102 can deliver in a Phase 3 celiac trial. It was a painful day, but the company's story isn't over. It just got more complicated.

Roche Is Hiring a Chinese Startup to Invent Its Next Cancer Drug

Roche signed a deal worth up to $1 billion with Defand Therapeutics, a two-year-old Chinese startup with zero clinical trials, to discover molecular glue cancer drugs. It's Roche's fourth China partnership since August, and it signals a radical shift in how big pharma thinks about drug discovery.