Roivant's JAK1/TYK2 Pill Wins First Dermatomyositis Nod & Spyre Deprioritizes TL1A Drug After RA Miss

A disease that has gone decades with nothing but steroids just got its first targeted pill — and the FDA stapled a boxed warning to it.

Plus Spyre is deprioritizing its rheumatoid arthritis drug after yesterday's miss, Tezspire cleared Phase 3 in eosinophilic esophagitis, and CStone put up lung cancer numbers that need a second read. Let's get into it.

Welcome to The Pharma Closeout for Thursday, August 27th. I'm Alex Mercer.

And I'm Maya Patel. Busy day at the agency. Let's take it apart.

The FDA approved LISRAYA — brepocitinib — thirty milligrams once daily, for adults with dermatomyositis. Roivant and Priovant announced it today, and it's prescribable in the U.S. as of this afternoon. But the number that decides this launch isn't an efficacy number. It's the boxed warning. Serious infections, mortality, malignancy, MACE, thrombosis — the full JAK class label, in a disease where the incumbent treatment is chronic high-dose prednisone.

Then say the comparator out loud, because that framing cuts the other way. Fifty-five percent of patients on LISRAYA hit the composite of moderate-or-better improvement on the Total Improvement Score plus minimal-or-no steroid use at fifty-two weeks, against thirty percent on placebo. You're asking a rheumatologist to weigh a boxed warning against a decade of prednisone. Those risk profiles are not symmetrical, and nobody in that clinic needs it explained to them.

Give me the taper.

Among patients on seven-point-five milligrams a day or more at baseline, sixty-two percent got down to minimal or no steroids by week fifty-two, versus thirty-eight percent on placebo. That's a full year of separation, not a snapshot. And the effect on the primary endpoint showed up by week four and kept building.

Which lands harder when you know what it's replacing. Priovant's own release describes decades of treatment resting on chronic steroids, non-specific immunomodulators, and IVIG — nothing aimed at the underlying biology. They call LISRAYA the first and only targeted therapy approved for dermatomyositis, and it cleared after Priority Review and Orphan Drug designation. VALOR ran in the New England Journal in March, with skin-specific secondary endpoints in JAMA Dermatology this month.

Here's the line most people will skim past. The indication carries no restriction on disease activity, clinical presentation, or prior treatment — and VALOR enrolled patients across a wide array of background regimens, including none at all. That's a label you can use first, not fourth. In a rare autoimmune disease, that breadth is worth more than any point estimate on the primary endpoint.

OK, but I want to press on treating the warning as priced in. Roivant's identity is the Vant model — stand it up, build it, hand it off. This is Priovant, a company with no launch history, putting a specialty product in front of rheumatologists and dermatologists who read JAK safety language conservatively. Endpoints framed today as the *beginning* of a road to a potential blockbuster. Beginnings are cheap.

If the warning were the binding constraint, I'd take that — but look at how they built the launch. Limited specialty pharmacy distribution. A dedicated Patient Access Liaison walking each prescription through coverage. Eligible patients paying a month. That is not a company hoping for volume. That is a company that already assumed every script would be fought for one at a time.

So the constraint isn't the label. It's finding the patients.

It's finding them and getting them covered. Those are different problems and they both bite.

Then here's the competitive read. If you're running immunology strategy, the asset today isn't dermatomyositis — it's the precedent. Priovant just proved dual JAK1/TYK2 inhibition clears FDA review in a rare autoimmune indication with a boxed warning attached. And brepocitinib is in late-stage development in non-infectious uveitis, cutaneous sarcoidosis, and lichen planopilaris. Same molecule, same regulatory logic, three more shots.

And we find out tomorrow how much of that they can fund. Roivant's investor call is eight a.m. Eastern. I want pricing and I want the covered-lives assumption — because in a disease this rare, reimbursement decides whether that broad label ever gets used at all.

Staying with immunology's cousin for a second — oncology. CStone Pharmaceuticals posted 2026 interim results today, and the lead asset did something people are going to quote badly. CS2009 is a PD-1/VEGF/CTLA-4 trispecific. At thirty milligrams per kilogram in first-line non-small cell lung cancer with high PD-L1 expression, the company reported a one hundred percent objective response rate.

On patients. Eleven of eleven is a hypothesis wearing a result's clothes. The number I'd actually underwrite is the cold-tumor activity — twenty percent response in heavily pretreated metastatic colorectal cancer, with a ninety-three point three percent disease control rate. Colorectal is where PD-1 blockade goes to die. If the CTLA-4 arm is doing real work, that's where it shows up first.

And they're taking it straight at the standard — first-wave global Phase 3s by year end, including first-line lung against pembrolizumab plus chemo. Meanwhile the commercial side is carrying weight: pralsetinib in-market sales volume rose almost five hundred percent year over year through July, following reimbursement listing in January.

Which changes what CStone is. A clinical-stage story funded by a commercial engine negotiates with a global partner from a very different chair than a clinical-stage story alone.

On the regulatory side — Amgen and AstraZeneca said today that Tezspire hit all primary and key secondary endpoints in the Phase 3 Crossing trial in eosinophilic esophagitis. We've been tracking this one since late July. What's new: both doses improved histologic remission and swallowing difficulty versus placebo at week twenty-four, and the effect held through fifty-two.

Third indication for a drug already approved in severe asthma and chronic rhinosinusitis with nasal polyps. People in the U.S. have this disease, per the partners, and prevalence is climbing.

And it walks straight into Dupixent, which per Fierce became the first medicine specifically indicated here. The distinction that decides the label is where each one acts — Tezspire upstream at TSLP, Dupixent downstream on IL-4 and IL-13. In a population where roughly half fail first-line therapy, upstream is a real argument rather than a me-too.

So this is a share story.

Not yet. They released no effect sizes — none. Until those land at a medical meeting, anyone modeling share shift is modeling a press release.

Also worth noting on the regulatory front — Verrica completed enrollment in COVE-2, its first pivotal Phase 3 for YCANTH in common warts. Topline in the first quarter of 2027, and the interim power analysis said no additional subjects were needed.

Which makes it a clean binary, and those are rare. Q1 2027 either opens an indication with no FDA-approved prescription therapy today, or it closes the label-expansion thesis outright. And Corvus published final Phase 1/1b soquelitinib data in Blood — median overall survival in the two-hundred-milligram cohort with one to three prior therapies. Peer review buys you credibility with investigators. It doesn't move a registration Phase 3 by a single day.

Second story — Spyre Therapeutics. Per Endpoints, the company is *deprioritizing* SPY072, its anti-TL1A antibody, in rheumatoid arthritis, after the Phase 2 miss we covered yesterday.

The verb is the tell. Companies defending a near-miss don't say deprioritizing — companies reallocating do.

So where does the capital go instead?

Straight at the readout that actually tests this molecule. Per Endpoints, Spyre's SKYLINE data in ulcerative colitis lands in September. TL1A's rationale was always built on bowel disease biology, not joint biology — the RA sub-study was a basket-trial add-on, not the thesis.

And the read-through runs past one ticker — there's real capital sitting in TL1A programs at Merck, Roche, Sanofi, and Teva.

Narrower than that. This is a setback for TL1A *outside* IBD. Nothing in the RA readout touches the bowel disease rationale — different tissue biology, separate evidence base. What it does do is end the free ride: every RA expansion arm in the class now has to justify itself on its own data instead of on class enthusiasm.

So the open thread is September — the SKYLINE data, and whether the mechanism holds where it's supposed to be strongest.

Exactly the discipline. One indication failing is a data point. A class fails when it fails where it should have worked — and September is when we find out.

Looking ahead — Roivant's call tomorrow at eight a.m. Eastern is the near-term catalyst. Price, distribution economics, and whatever they'll say about the three follow-on indications.

And two dated markers further out. CStone expects alignment with the FDA and China's CDE on the CS2009 global Phase 3 protocol in the fourth quarter, with two rapid oral presentations at ESMO in October. Verrica's topline lands in Q1 2027. Those are the dates that turn today's press releases into something you can actually underwrite.

The thing I'll be sitting with tonight is that taper curve — sixty-two percent off high-dose prednisone at a year. That number is going to get repeated in rheumatology clinics long after the stock reaction fades. Have a good evening.

And that is your Pharma Closeout for Thursday, August 27th — LISRAYA's approval and the boxed-warning trade underneath it, Spyre deprioritizing its RA drug after the miss, Tezspire's Phase 3 win, and CStone's trispecific. Four stories, four very different bets. Follow the show on Apple Podcasts, Spotify, or wherever you listen — a quick rating or share helps other listeners find us. Tomorrow's briefing lands on its own.