AstraZeneca, Bristol Myers Squibb Weigh a $400B Merger

A four hundred billion dollar drugmaker nearly exists. AstraZeneca and Bristol Myers Squibb have been in the room for months — and today the Financial Times put it on the record.

And it lands at the end of a week where the evidence took apart three separate programs — a schizophrenia trial that hit its endpoint and lost two-thirds of a market cap anyway, a Duchenne cell therapy voted down, and a second Phase 3 flop in endometrial cancer. Let's take the week apart.

Welcome to The Pharma Closeout Week in Review for Sunday, August second. I'm Alex Mercer.

And I'm Maya Patel. It was a big week. Let's get into it.

Start with the number, because the number is the story. The Financial Times reported today that AstraZeneca is in talks to merge with Bristol Myers Squibb. Combined value: roughly four hundred billion dollars. AstraZeneca sits at about two hundred sixty-three billion. Bristol Myers, around one hundred thirty-three billion. Talks have been ongoing for some time now. Could produce a deal soon — could also delay, could also collapse.

Neither company responded to CNBC's request for comment. That's its own data point. This is a report, not an announcement.

Right. And the reason it lands with this much force isn't the arithmetic — it's what each side is trying to solve. AstraZeneca just posted a six percent revenue increase, with oncology at nearly half of product revenue. It elevated its New York listing this year. The CFO has been publicly confident about hitting the eighty billion dollar revenue target. That is not a company merging from weakness.

Bristol Myers is a different animal entirely. It raised its 2026 outlook — and in the same breath delivered another delay on the Cobenfy Alzheimer's data and a pushed-back milvexian readout. Those were the two programs meant to define what Bristol Myers looks like in 2030. When both slip in one quarter, the conversation inside that building changes shape. And a company leaning into AI partnerships for drug discovery, including a collaboration with Anthropic, is telling you it already knows the internal engine needs help.

So here's the turn. Per the Benzinga report, this would push AstraZeneca into the upper tier of the world's largest drugmakers by market capitalization. And it would dwarf its own precedent — the Alexion acquisition in 2021, its largest deal to date. Structure undetermined. Likely cash and shares.

I'd push back on your framing, though. You're calling this a growth story stapled to a repair story. That's not the tension. Both companies have significant cancer divisions — that's the flagged antitrust exposure, and it's also the strategic problem. Combine two large oncology portfolios and you don't get twice the oncology. You get one portfolio and a divestiture list. The question isn't what they buy. It's what they're forced to sell.

I'd take the correction, because it sharpens where this lands. The logic isn't additive. It's about which assets survive overlap review and which get carved out. Which means the read-through for the rest of oncology is that a very large auction may be coming.

And that's the synthesis neither of us walked in with. The most consequential thing about a four hundred billion dollar merger, for everyone who isn't in it, might not be the combined company at all. It's what falls out of it. Any mid-cap hunting for a marketed oncology product just acquired a potential seller.

Let's bring in Marcus Webb on this one — structure is the whole game here.

The competitive read: if you run commercial strategy anywhere in oncology, this week's actionable item isn't the merger. It's building the divestiture watchlist now — overlap review on two portfolios this size takes months, and the assets that come loose will move fast.

And the open thread is regulatory on two fronts. Antitrust scrutiny is expected; that's in the reporting. The second is British politics, where AstraZeneca tilting further toward the US market is already a live question. Neither one kills a deal. Both extend a timeline. So what to watch over the next several weeks isn't whether they agree — it's whether anyone puts a name on the record.

What tied this week together wasn't dealmaking, though. It was evidence getting graded harder than anyone expected.

Three cases in five days. July twenty-seventh: MapLight reports Phase 2 ZEPHYR hit its primary endpoint in schizophrenia. Statistically significant PANSS improvement at Week 5. The effect size was modest but notable. A meaningful margin over placebo. The result cleared the bar for statistical significance. That's a win by the protocol's own definition. Stock lost roughly two-thirds of its value that week.

And CEO Christopher Kroeger went on the record with BioPharma Dive defending the drug on the totality of the data.

He's not wrong on the totality — and this is the piece I'd flag. The prespecified completer analysis, no modelled assumptions for missing data, came in at effect size zero point five zero. Six-point difference. The result was highly statistically significant. The entire gap between those two numbers is how you treat dropouts, and all-cause discontinuation across both active arms was nineteen point nine percent. That's the whole argument in one line. Investors priced the mITT. The company is arguing the completers.

And the market took the harsher read because Bristol Myers already has Cobenfy approved in the class. There's a bar now.

There's a bar. And that's the through-line. Two days later — July twenty-ninth — FDA's Cellular, Tissue and Gene Therapies Advisory Committee voted decisively that Capricor's deramiocel didn't show substantial evidence of effectiveness in Duchenne-related cardiomyopathy. Panelists kept using one word about the data. Fragile.

And the detail that actually decided that room?

The full Phase 3 results published in The Lancet as the meeting was starting. And the paper revealed the treatment had missed statistical significance on the heart benefits Capricor had been touting. The company issued a clarification in a regulatory filing.

The publication contradicted the pitch in real time. I've not seen that sequence before.

FDA reviewers said the statistical changes Capricor employed lacked a scientific basis. Capricor's answer was that the agency leaned on an older, irrelevant analysis. That dispute is unresolved, and the decision on the latest submission is expected by August twenty-second.

Then Karyopharm closes the week out. Late Thursday, July thirtieth — Xpovio misses its primary endpoint in Phase 3 endometrial cancer. No significant difference in progression-free survival in TP53 wild-type patients. Shares down around sixty-five percent. Second Phase 3 flop for that drug in that indication.

The second attempt is the part that matters. The first failure handed them a biomarker hypothesis — TP53 wild-type. They ran it back on the enriched population, and it still didn't separate. When a prespecified biomarker strategy fails on its second pass, you've stopped looking at a trial design problem. You're looking at the molecule.

So: three programs, three modalities, and in every case the market or the agency read the evidence more conservatively than the sponsor did.

Which sets the environment for anyone sitting on a borderline dataset heading into an end-of-Phase 2 meeting or an advisory committee. The flexibility everyone assumed was arriving with new FDA leadership did not show up this week. What showed up was a harder read on statistical analysis plans. Plan into that, not into the narrative.

Under the radar this week: the same advisory committee that voted Capricor down on Wednesday sat down to a completely different application on Thursday. Same panel. Consecutive days. Opposite directions.

Which is genuinely useful intelligence about that committee. Those votes weren't a mood. They were an evidence read, application by application. A panel that says no on Wednesday and yes on Thursday is a panel doing its job — not one signaling a posture.

And that corrects a narrative a lot of people have been carrying. The running assumption has been that newer agency leadership means a broadly softer review environment. Consecutive days in the same room showed something narrower: the standard is being applied case by case, and the deciding variable both times was the integrity of the statistical analysis plan. Not the severity of the disease.

So if your regulatory strategy is built on institutional flexibility, that word needs a much tighter definition than it had a month ago.

Looking ahead — and the calendar is front-loaded. Eli Lilly reports second-quarter results August fifth. Consensus is looking for solid revenue and per-share earnings roughly in line with expectations. Lilly also declared its regular cash dividend, with the ex-date set for mid-August.

The line I'd listen for on that call isn't the headline. It's Foundayo. Truist has described the U.S. launch as off to a slower than expected start, while Novo's oral Wegovy reached a major patient milestone in May after a strong initial run on the market — what CEO Maziar Mike Doustdar called a record-breaking start. If Lilly's commentary starts reframing Foundayo as an international story rather than a U.S. one, that's a strategic repositioning dressed up as a quarterly update.

Also August fifth: Moderna's flu vaccine, mRNA-1010 — MFLUSIVA — carries an FDA action date. Then later in the month, Capricor's deramiocel decision is expected by August twenty-second, following that panel vote.

And on the promise we left open Friday — we said we'd watch whether Novo said anything further on ZEUS before the week closed. Nothing has landed as of this recording. That one stays open.

So by Wednesday you'll know whether Lilly's oral franchise is a slow build or a structural problem, and whether the AstraZeneca report firms into something either board will confirm. That's a lot of clarity arriving inside seventy-two hours.

That wraps our Week in Review — a week whose biggest headline was a four hundred billion dollar merger nobody has confirmed, and whose real story was three programs getting graded harder than their sponsors expected. AstraZeneca and Bristol Myers. MapLight. Capricor. Karyopharm. Watch Wednesday — Lilly's numbers and Moderna's flu decision land the same day. Follow wherever you listen and you'll start every weekday with this.

Have a good week. I'm going to read the Capricor Lancet paper properly before the twenty-second — that one deserves more than a headline. Let's go!