Merck's 2028 Keytruda Cliff: The Oral PCSK9 Bet

Merck just told you exactly how it plans to survive 2028 — and the answer isn't a molecule that came out of its own labs.

Pfizer stacked another two and a half billion onto the cost program while ruling out a mega-deal, and BioNTech handed the chief executive job to a Sobi operator while its founder walks toward a startup. Busy Tuesday. Let's get into it.

Welcome to The Pharma Closeout for Tuesday, August 4th, 2026. I'm Alex Mercer.

And I'm Maya Patel.

Merck lost one point three four billion dollars in the second quarter — and beat Wall Street on both lines anyway. Revenue sixteen point six one billion against a sixteen point three six consensus. An adjusted loss that came in narrower than the street had modeled. Revenue outlook up, profit outlook down, same press release. And every moving part of that quarter is bolted to one date: 2028, when Keytruda loses exclusivity.

The number I'd pull sits inside Keytruda itself. Eight point three seven billion for the quarter, up five percent, ahead of the eight point two seven the street wanted. But four hundred sixty-three million of that came from the subcutaneous version. That's the conversion figure. It's the only slice of that franchise that carries cleanly past 2028 — the rest is exactly what the biosimilars are aimed at.

Four hundred sixty-three million out of eight point three seven billion.

That's the whole race, in one ratio.

Which is why the pipeline commentary matters more than the beat. Merck now has an oral inhibitor in the PCSK9 class — the first pill in that class, approved in July, per the company's second-quarter report. And it's planning a trial pairing sacituzumab tirumotecan with Keytruda. Underneath those two, the newer launches are doing real work: Winrevair, five hundred eighty-eight million, up seventy-five percent. Capvaxive, one hundred eighty-four million, up forty-two. Animal health, one point seven eight billion, above estimates.

Here's my issue with calling any of that replenishment. Look at what the profit cut is actually made of. Adjusted earnings guidance narrowed to a tighter range, down from a wider prior guidance range. That swing is a five point seven billion dollar charge tied to Terns, and a nine billion dollar charge tied to Cidara back in January. Merck isn't refilling from its own labs. It's buying the replacement and expensing it in the same breath.

I'd take the correction, because it sharpens where this lands. The revenue guide moved higher, reflecting management's growing confidence. The profit guide went down. The distance between those two moves is the invoice for the buying spree. So this isn't a pipeline story. It's portfolio repair, funded off the balance sheet. And the competitive read is clean: if you're running commercial strategy in cardiometabolic, an oral entrant in the PCSK9 class now has a top-tier launch engine behind it — every access assumption you built around injectables in that space just got a shorter shelf life.

Then the sharper way to read the combination trial is as defense, not growth. A biosimilar can only reference the originator's label. Indications generated after entry — combination settings, new populations — are label real estate the copies can't walk into without running their own studies. So sac-TMT plus Keytruda isn't purely an oncology bet. It's a fence. And the open question is timing, because a combination label that shows up in 2029 protects nothing.

That's the whole game.

Pfizer reported into the same morning, and the shape of its answer could not be more different. Adjusted earnings came in ahead of expectations, a solid beat. Revenue fifteen point oh three billion against fourteen point four one, up three percent. Low end of full-year revenue guidance up to sixty point five billion. Covid products cut to four billion from around five. Eliquis went straight through its number — two point four three billion, up nineteen percent, against a two point oh eight estimate. Recently launched and acquired products delivered three point two billion in the quarter. And then the line for the year: an additional two point five billion in cost savings across two separate programs, running over the next several years. Stock up modestly on the news.

What interests me is the pairing. Bourla is ruling out mega-M&A while the savings target keeps climbing. That's a company telling you growth has to come from inside the portfolio it already owns. Fine — but that posture only works if the assets you own actually work. Ask the four point three billion dollar impairment how that's going.

Let's bring in Marcus Webb on the structure here.

Noted. And Marcus, that's the same tension we just described at Merck, run in reverse.

Which brings us to the third answer — leadership. BioNTech named Guido Oelkers chief executive, with the transition set for early next year, succeeding Ugur Sahin. Oelkers comes from Swedish Orphan Biovitrum, where he ran the company. Sahin is heading toward a fledgling mRNA startup, Özlem Türeci steps out of management to lead an independent company alongside him, and there's a chief medical officer search underway.

BioNTech's board didn't hire a discovery scientist. They hired a commercial operator with an oncology and immunology track record — and that CEO choice is the pivot. Away from founder-led mRNA platform strategy, toward a portfolio company that has to make late-stage oncology pay off within the next several years. The leadership change isn't a consequence of the strategy shift. It is the strategy shift. So for anyone mapping immuno-oncology, BioNTech just moved from a platform question to an execution question, and those two get priced very differently.

And Sahin walking to a startup is the part the market will chew on longest.

On the regulatory side, the item worth sitting with is buried in Pfizer's impairment. That four point three billion dollar noncash charge came partly from disappointing late-stage results for sigvotatug vedotin in previously treated non-small cell lung cancer — and partly from removing revenue projections for Oxbryta after recent discussions with the FDA. Pfizer pulled that product from the market in 2024. Taking the forecast to zero after agency conversations is the company saying, in accounting language, that there is no route back.

That closes a chapter in sickle cell a lot of people assumed was still open.

It does — and this is where I'll add the layer, because it isn't obvious. A market withdrawal isn't automatically terminal. Sponsors have come back after a safety review with a narrowed population and tighter monitoring built into the label. Writing the projection to zero says that conversation happened and did not produce a path. Every other sponsor working that disease area should read it as a signal about where the agency's tolerance sits on that specific safety question.

And the flip side of the ledger — Merck's oral entrant in the PCSK9 class cleared in July.

Which leaves the question that actually decides the business, and we don't have it on air yet: what does the label say. Broad LDL lowering versus a statin-intolerant or established-risk niche are two completely different companies built off the same molecule. Until that's pinned down, anyone forecasting off "first pill in the class" is forecasting a headline, not a market — and payers will settle it long before the sales force does.

Looking ahead, Pfizer has crucial data still to come this year, including a combination regimen pairing its GLP-1 injection with an amylin asset. That readout decides whether ten billion dollars for Metsera reads as foresight or as full price. On the Merck side, generic entry for Januvia and Janumet lands later this year, and the sac-TMT plus Keytruda trial is the pipeline item to track from here.

For BioNTech, February is the calendar date — but the tell arrives earlier. Watch whether oncology prioritization shows up in program decisions before Oelkers is even in the chair, and watch who takes the chief medical officer job. That hire tells you how aggressive this pivot really is.

Put it together and the back half of this year is a scoreboard for three different answers to one problem. Merck buys what it needs. Pfizer optimizes what it owns. BioNTech changes who's deciding. By January, we'll know which one bought time and which one just booked a charge.

And that is your Pharma Closeout for Tuesday, August 4th — Merck's oral PCSK9 pill against the 2028 Keytruda clock, Pfizer stacking another two and a half billion onto cost while Eliquis carries the quarter, and BioNTech handing the keys to Guido Oelkers. Three companies, one deadline, three completely different bets. Follow wherever you listen and tomorrow's briefing lands on its own.

Mine is the label. Whenever the prescribing information for that oral cholesterol pill posts, that's the document I'm reading line by line — it decides how much of the 2028 hole Merck actually fills. Have a good evening.