Zenbexus Approval Tests a New Endpoint: What BMS Still Owes the FDA

The FDA just approved a Bristol Myers cancer drug without seeing a single survival curve — and handed the entire myeloma field a new playbook while doing it.

Meanwhile Capricor pulled a Duchenne filing back from the dead, Taiho and Cullinan stopped a Phase 3 early because the data got too good to keep blinded, and a failed obesity biotech just sold its own stock listing. Busy Friday. Let's get into it.

Welcome to The Pharma Closeout for Friday, August fourteenth. I'm Alex Mercer.

And I'm Maya Patel.

We've been tracking the Zenbexus decision since it hit the August calendar, and it landed. Here's the number that carries it: forty-one percent minimal residual disease-negative complete response, against twenty-one for the standard comparator. That doubling is the entire approval. Not progression-free survival — that trial is still running. Bristol Myers has an accelerated approval for iberdomide, branded Zenbexus, with Darzalex and dexamethasone, in myeloma patients past their first line. Fierce Pharma reports it's the first FDA approval for a CELMoD, and per the company, the first new myeloma drug cleared on residual disease rather than a longer-term measure.

That doubling is real. But the number I'd put in front of people is four percent.

Say more.

Four percent of patients in Excaliber-RRMM had prior anti-CD38 exposure. Four. The FDA wrote that limitation straight into the label. So the drug is approved for second-line and beyond, in a world where second-line patients are increasingly Darzalex-experienced from the front line. The population they studied is the population that's disappearing.

OK, I'll take part of that — the first indication is thinner than the headline implies. But you're describing an indication problem and Bristol Myers is solving a franchise problem. Revlimid has been eroding since exclusivity went in twenty twenty-two. CELMoDs have been the designated replacement engine in every investor deck they've put out. William Blair models U.S. sales above a billion by early twenty thirty-one, which means the street isn't underwriting this launch — it's underwriting Excaliber-Maintenance, post-transplant, newly diagnosed. That's where the volume lives.

Then the honest read is that today isn't really about Zenbexus.

Go on.

It's about the endpoint — and that cuts in a direction Bristol Myers doesn't control. The FDA issued draft guidance in January on residual disease supporting accelerated approval in myeloma, after an advisory committee endorsed the endpoint unanimously in twenty twenty-four. Zenbexus is the first new molecule to actually cash that check. Which hands every sponsor in the field a template: one Phase 3, earlier endpoint gets the accelerated nod, progression-free survival converts it later. That's Project FrontRunner running in the wild, and Bristol Myers just proved it works.

That's the tell.

And here's the part nobody's pricing. The confirmatory obligation is progression-free survival — expected this year — from the same trial that's already been unblinded for the residual disease analysis. So the sponsor who opened the door is also the sponsor who could close it. If that curve comes back soft, the first drug approved on this endpoint becomes the first cautionary tale about it. The precedent isn't created by the approval. It's created by whether the approval survives.

And that's the competitive read. If you run strategy in myeloma — or any heme malignancy with a validated surrogate — the trial-design assumption you carried into January is now obsolete. One pivotal study can carry both filings. That compresses timelines for anyone willing to build to it, and it strands anyone who committed to a two-trial program eighteen months ago. The question isn't whether Zenbexus sells. It's whose protocol gets rewritten next quarter because of it.

On the regulatory side — Capricor, and this is one we've been tracking since the advisory committee. That panel voted nine to three that the evidence did not support effectiveness in Duchenne-related cardiomyopathy. What's new today: CEO Linda Marbán told analysts the FDA is willing to review an amendment. Capricor will file for what she called a refined indication built on the primary endpoint — upper limb function. Current action date is August twenty-second, and they expect an extension once the amendment lands.

Stock went up seventy-one percent on that.

And still trades well below where it sat before the panel. So the market is pricing a lifeline, not a rescue — and I'd argue even the lifeline needs reading carefully.

Why? They're retreating to the endpoint they actually powered for. That's a clean regulatory move.

It is, procedurally. Hope-3's primary endpoint was upper limb function; cardiomyopathy was secondary. They took a secondary endpoint to a panel and lost on it. Falling back to the primary is defensible. But go to the meeting transcript. A biostatistician from Iowa said he was not convinced there's a benefit in upper limb — only that he felt stronger about it than about ejection fraction. A neurologist called the data very fragile. That's the fallback endpoint getting hedged support, not clean support.

So the better argument still didn't land.

The better of two weak arguments. The amendment adds open-label extension data plus new analyses of existing data. Whether longer follow-up converts a fragile signal into evidence is the whole question — and there's a real scenario where the FDA takes the filing, extends the date, and still says no. A willingness to review is not a signal of intent.

Also worth noting on the regulatory front — Taiho and Cullinan, Wednesday. Zipalertinib plus chemo hit in first-line EGFR exon insertion lung cancer, and investigators halted the study at an early data check because the results were clearly positive. No detailed numbers. What we do have is William Blair's Matt Phipps noting the trial was powered for a forty percent reduction in relative risk of progression or death at final analysis — and stopping early implies the observed effect ran bigger than that.

Which is a genuinely strong result, and I'd still say it doesn't settle the market. Rybrevant and Zegfrovy are already there. Cantor's Li Watsek made the point that differentiation lands on safety, not efficacy — and she's right, because ArriVent is running firmonertinib as monotherapy against chemo. If that reads out clean, they have a tolerability story a chemo combination structurally cannot match.

ArriVent shares dipped as much as seven percent anyway.

Which tells you the market read it as an efficacy race. Watsek herself cautioned it's premature to size the threat. So the implication for anyone with an exon program in the clinic: the efficacy bar moved, and the only lane left is tolerability. If your Phase 3 is built to beat chemo on response, you're now answering a question nobody's asking.

Which brings me to the transaction I've been waiting to talk about all morning, because it's the same lesson from the other end of the telescope. Skye Bioscience and Redx Pharma announced a transaction agreement this morning with roughly a hundred and twenty-five million dollars committed. The combined company becomes Fibrx Therapeutics, Nasdaq-listed, run by Redx's board and management out of Alderley Park. But the ownership split is the story — pre-transaction Skye holders end up with about five point four percent.

Five percent isn't a merger. That's a listing being sold.

It is a listing transaction, and I'd defend it as the right one. Skye ended eye disease R&D in twenty twenty-four after a glaucoma failure. Last October nimacimab failed to add meaningful weight loss on top of semaglutide at twenty-six weeks and the stock dropped sixty percent. They ran a strategic review in Q2. This is what a board does when the pipeline is gone and the shell still has value — and Skye holders keep a contingent value right on ninety percent of any future nimacimab monetization.

The asset is the more interesting half of it. RXC008 is a GI-restricted pan-ROCK inhibitor for fibrostenotic Crohn's, heading into Phase 2. GI-restricted is the entire design thesis — you inhibit ROCK in the bowel wall without systemic exposure, and Phase 1 showed no clinically relevant systemic breakthrough and no hypotension. That's the safety problem that has haunted ROCK inhibition, and they appear to have engineered around it.

Fast Track in January, open IND, cash into twenty twenty-nine.

Cash into twenty twenty-nine to deliver a single topline in the second half of twenty twenty-eight. In an indication where the current alternative is bowel resection surgery — so the need is real, and the science is legitimate. But anyone buying today is buying twenty twenty-eight, and there is almost nothing in between to reprice on.

Which is exactly why the structure matters more than the science here. Fibrx got a hundred and twenty-five million and a Nasdaq ticker without touching an IPO window. If you're a small-cap with a dead pipeline and a live listing, you now have a demonstrated route that doesn't involve winding down. Expect more Skye-shaped shells to go looking for Redx-shaped assets before year-end.

Looking ahead. Capricor's action date is August twenty-second, though based on what Marbán described, expect an extension rather than a decision. On Bristol Myers, the progression-free survival readout from Excaliber-RRMM is expected this year — and that's the confirmatory evidence the whole accelerated approval is standing on.

And a loop I owe you. Definium's second Phase 3 in generalized anxiety — the Panorama study — reads out next month. Nothing new landed this week; it's still ahead of us. I'll have the design open when it does, because that low-dose control arm is what makes the trial worth reading closely.

And mezigdomide, the second CELMoD, has a decision date in May twenty twenty-seven. That's the one that tells you whether today was a class arriving or just a drug arriving.

The thing I'll be sitting with this weekend is that we spent the whole show talking about an approval whose foundation hasn't been published yet. Everything today hangs on a survival curve nobody has seen. Go enjoy your weekend.

And that is your Pharma Closeout for Friday, August fourteenth — the first CELMoD approval and the endpoint that came with it, Capricor clawing a Duchenne filing back off the mat, Taiho and Cullinan stopping a Phase 3 early, and Skye and Redx becoming Fibrx. Genuinely one of the more consequential Fridays we've had this year. Follow the show on Spotify, Apple Podcasts, or wherever you listen — the Week in Review lands Sunday.