Shares of AstraZeneca fell after reports it could merge with Bristol Myers Squibb in a deal worth nearly $400 billion — a combination that would rank among the largest in pharmaceutical history, if it happens.

The Financial Times first reported the potential tie-up on August 3, 2026, saying any deal would involve both cash and shares, though the structure was unclear. Neither company confirmed: AstraZeneca declined to comment and Bristol Myers did not immediately respond.

AstraZeneca is valued at roughly $245 billion and Bristol Myers at about $133 billion.

Investors were sceptical

Rather than cheering, the market recoiled. AstraZeneca’s stock fell — reports put the drop at anywhere from around 2.7% to more than 8% during the session — while Bristol Myers shares slipped too.

Both falling is unusual and informative. In a typical acquisition rumour the target rises, since it would be bought at a premium, and the acquirer falls if investors doubt the price or the logic.

Both declining suggests the market saw no clear winner — a combination creating problems for each party rather than transferring value from one to the other.

Why analysts are dubious

“If there is one company that doesn’t need financial engineering, it’s AstraZeneca,” Jefferies analyst Michael Leuchten wrote.

AstraZeneca has projected topping $80 billion in sales by 2030 with more than 25 billion-dollar products, while Bristol Myers faces looming patent cliffs on top sellers like Opdivo and Eliquis.

That asymmetry is the core objection. Large pharmaceutical mergers are usually undertaken by companies facing a revenue gap, buying growth they cannot generate internally. AstraZeneca is in the opposite position, and acquiring a company with expiring products would dilute its growth rate rather than support it.

What the patent cliffs mean

The two products named account for a substantial share of Bristol Myers’s revenue, and both face exclusivity loss within a few years.

Eliquis, an anticoagulant, is among the highest-revenue medicines in the world and will meet generic competition that typically removes most of a product’s sales within a year or two. Opdivo, a checkpoint inhibitor, faces biosimilar competition, which erodes more slowly but substantially.

A buyer therefore acquires current revenue with a visible expiry date, and must value the company on what replaces it — which is a judgement about a pipeline rather than about the products generating today’s cash.

The regulatory obstacle

Analysts also flagged heavy antitrust scrutiny, particularly in the UK, where AstraZeneca is a cornerstone of biopharma research.

Competition review of pharmaceutical mergers focuses on overlapping products and pipelines, and both companies are substantial in oncology and immunology — likely requiring divestitures.

The UK dimension is partly political rather than competitive. AstraZeneca is one of the country’s largest companies and a central pillar of its life-sciences sector, and a combination that shifted control or research investment would attract scrutiny beyond the ordinary antitrust question.

Why megamergers keep being contemplated

Deals of this scale recur despite a mixed record, because the arithmetic is attractive on paper.

Combining two large companies permits elimination of duplicated commercial, administrative and manufacturing functions, and those savings are calculable in advance in a way pipeline value is not.

The historical difficulty is that integration consumes management attention for years, research productivity frequently falls as combined organisations restructure, and promised synergies materialise less completely than modelled. Several previous pharmaceutical megamergers are regarded as having destroyed value rather than created it.

How to read an unconfirmed report

A story of this kind moves hundreds of billions in market value on information neither party has confirmed, which is worth examining as a phenomenon in itself.

Reports of this scale generally originate with someone close to a process, and the motivation for leaking varies. A target may leak to attract other bidders or to pressure the price upward. A bidder may leak to test market reaction before committing. Advisers may leak to advance a transaction that is stalling.

The market response then becomes information the parties use. A sharply negative reaction of the kind seen here — the presumed acquirer falling several percent — is a direct message from shareholders that they would not support the deal, and boards do respond to that.

Which means the story can change what it reports. Talks that were exploratory may not proceed after a reaction like this, and it becomes impossible afterwards to know whether a deal was ever close. “Declined to comment” is the standard formulation whether talks are advanced or never occurred, so the absence of denial confirms nothing either way.

For now it remains an unconfirmed rumour — but one big enough to move markets. Business news, not investment advice.