

Roche's new oral pill fenebrutinib cut MS relapses by up to 59% and matched its own blockbuster infusion in progressive disease. But a troubling death imbalance in clinical trials has analysts watching the FDA's next move closely.
Imagine spending your entire afternoon in a clinic, hooked to an IV, just to keep your immune system from attacking your own brain. That's the reality for millions of people living with multiple sclerosis. Now Roche says it can replace that infusion bag with something that fits in your palm.
The Swiss pharma giant just dropped Phase III results for fenebrutinib, an oral pill designed to treat both relapsing and primary progressive forms of MS. The data is impressive. The safety questions are real. And the commercial stakes are enormous.
MS treatments have come a long way, but most of the best options still involve needles. Self-injectables like Avonex require weekly shots. Infusions like Roche's own blockbuster Ocrevus (ocrelizumab) demand clinic visits every six months, each one eating up hours of a patient's day between the drip, the observation period, and the commute.
The burden is more than just inconvenience. Flu-like symptoms, injection-site reactions, and infusion reactions are common side effects. Many patients eventually switch or quit their therapy altogether, often because they simply can't tolerate the routine anymore.
Oral options already exist (think fingolimod and dimethyl fumarate), but here's the catch: they generally aren't as potent as the heavy-hitting infusions. Patients with aggressive disease often have no choice but to sit in that chair. What's been missing is an oral pill that punches as hard as an infusion. That's exactly the gap Roche is trying to fill.
Fenebrutinib belongs to a class of drugs called BTK inhibitors (Bruton's tyrosine kinase inhibitors, for the curious). Think of BTK as a molecular switch inside certain immune cells. Flip it on, and those cells go haywire, attacking the protective coating around nerves. Fenebrutinib keeps that switch turned off.
What makes it unusual is how it works. Most BTK inhibitors form permanent bonds with their target enzyme, like supergluing a light switch in the off position. Fenebrutinib takes a gentler approach: it binds, does its job, then lets go. This reversible mechanism could mean fewer off-target side effects, though that theory still needs long-term data to back it up.

Generate Biomedicines just raised $400 million in the largest biotech IPO since 2024, betting that AI can design better drugs than humans. With a Phase 3 asthma program, a generative biology platform, and the Moderna-incubator Flagship Pioneering behind it, this is the biggest test yet of whether AI drug design can survive public markets.


Join thousands of biotech professionals who start their day with our free, daily briefing.
Perhaps the biggest selling point is that the drug can actually cross into the brain. Many MS therapies work in the bloodstream but struggle to reach the central nervous system, where the real damage happens. Fenebrutinib was specifically designed to penetrate the blood-brain barrier, targeting both B cells and microglia (the brain's resident immune cells) at the source of the disease.
Roche ran three pivotal Phase III trials, and the headline results are genuinely strong.
In relapsing MS, the FENhance 1 and FENhance 2 trials compared fenebrutinib against teriflunomide (an existing oral therapy) in nearly 1,500 patients. Fenebrutinib cut the annualized relapse rate by 51% in FENhance 1 and 59% in FENhance 2. To put that in perspective, combined results translate to roughly one relapse every 17 years. For a disease that can steal your ability to walk, see, or think clearly, that's a meaningful difference.
The drug also slashed markers of brain inflammation. Active lesions (the bright spots on MRI scans that signal new damage) dropped by 71% to 78% compared to teriflunomide. Chronic disease burden, measured by a different type of lesion, fell by 76% to 83%.
For primary progressive MS, the FENtrepid trial compared fenebrutinib head-to-head against Ocrevus, Roche's own blockbuster. The pill met its primary endpoint of non-inferiority, meaning it worked at least as well as the infusion. It actually showed a numerical 12% reduction in disability progression risk. A post-hoc analysis found an even larger 22% reduction on a composite endpoint. That's not just "as good as the infusion"; it hints at potentially better.
Now for the part Roche would rather you not linger on.
Across the relapsing MS trials, eight patients died on fenebrutinib compared to one on teriflunomide. The primary progressive trial showed a similar imbalance: seven deaths on fenebrutinib versus one on Ocrevus. Investigators reviewed every case and found no consistent pattern in timing or cause, deeming all deaths unrelated to the drug.
That explanation hasn't fully satisfied analysts, and it shouldn't. Sanofi's rival BTK inhibitor, tolebrutinib, hit a regulatory wall over liver safety signals; the FDA issued a complete response letter for it over similar concerns. Fenebrutinib's liver enzyme elevations were comparable to the control groups, but one case of Hy's Law (a red flag for serious liver injury) popped up in each arm of the relapsing trials.
Regulators will scrutinize this data carefully. A death imbalance doesn't automatically sink a drug, especially when individual cases lack a clear link. But it does give the FDA something to chew on, and it could mean a longer review timeline or additional safety monitoring requirements.
Roche finds itself in a fascinating strategic position: competing against its own product. Ocrevus generated $6.5 billion in sales over just the first nine months of 2025. It's one of the company's top growth drivers. And now Roche is essentially telling the market, "We have something better."
The logic makes sense when you zoom out. Ocrevus faces patent expiration in 2029, with biosimilar competition looming right behind it. Novartis is already stealing share with Kesimpta, a self-administered injection that patients can use at home. Rather than watch its MS franchise erode, Roche is trying to replace Ocrevus before someone else does.
The company plans to file for regulatory approval following the FENhance 1 readout, with submissions expected in the first half of 2026.
Roche isn't the only company chasing a BTK inhibitor for MS. Novartis has remibrutinib in Phase III for relapsing MS. Sanofi's tolebrutinib, despite its FDA complete response letter for relapsing and progressive forms, managed to secure approval in the UAE for a rare subtype called non-relapsing secondary progressive MS.
The oral MS segment already holds roughly 36% to 50% of the total market depending on which analyst you ask, and that share is growing. Patients prefer pills. Doctors prefer compliance. Payers prefer anything that keeps patients out of infusion centers.
If fenebrutinib can clear the regulatory bar (a legitimate "if" given the safety signals), it would become the first oral, brain-penetrant BTK inhibitor approved for both relapsing and primary progressive MS. That's a first-in-class label in two indications simultaneously, the kind of positioning that makes commercial teams salivate.
Fenebrutinib's efficacy data is legitimately exciting. Cutting relapses by more than half, matching or beating the world's best-selling MS infusion, and doing it all with a pill you take at home: that's the kind of advance patients have been waiting for.
But the death imbalance casts a shadow that won't go away until regulators weigh in. Roche is betting billions that the benefit outweighs the risk. For the nearly 2.9 million people worldwide living with MS, the answer to that question isn't just financial. It's personal.
Roche deployed over 3,500 NVIDIA GPUs to build pharma's largest AI computing setup, and it's not just a flex. The move signals that GPU-accelerated drug discovery is graduating from pilot programs to full-scale enterprise deployment, with billions of dollars and years of patient waiting on the line.