AbbVie's Etentamig Clears Phase 3 Myeloma Trial on Both Endpoints & GSK Pays HUTCHMED $110M Upfront for KRAS-EGFR Drug

AbbVie beat standard therapy on both primary endpoints in myeloma today. Then it built the announcement around where the drug gets given.

And GSK wrote a hundred and ten million dollar cheque to HUTCHMED for a drug not yet dosed outside China. Plus the first treatment ever approved for Alexander disease, a Keytruda biosimilar deal, and a GLP-1 heading to Europe. Let's get into it.

Welcome to The Pharma Closeout for Thursday, September 3rd, 2026. I'm Alex Mercer.

And I'm Maya Patel.

AbbVie and etentamig. It's a BCMA-by-CD3 bispecific in multiple myeloma, and Phase 3 CERVINO cleared both primary endpoints. Seventy-four percent response rate against forty-five point seven percent for standard therapy. And a sixty percent cut in the risk of progression or death.

Three hundred ninety-three patients, triple-class exposed, a median of three prior lines. That's the population where response rates usually collapse. Seventy-four percent there is a real number.

The comparator was no straw man.

Investigator's choice — carfilzomib, elotuzumab or selinexor, each with dexamethasone. Then the hazard ratio is the honest headline. Point four zero. This drug didn't just produce more responses. It held them longer, against regimens people actually use.

Clean win, then.

Clean, and still not the argument AbbVie is leading with. Read past the headline numbers, and every sentence in that release is about how the drug is given.

Fair — the release isn't selling the response rate. One step-up dose, then monthly dosing from the start. Cytokine release syndrome in twenty-eight percent of patients, almost all of it mild, and none of it severe. AbbVie says those results support treatment in outpatient and community settings. That safety section is written for a community clinic, not a hospital ward.

Then the asset AbbVie is filing isn't the response rate. It's one sentence about setting. Get that sentence into a label and you have something the other bispecifics don't have.

That's the sentence regulators will push on hardest. And AbbVie is arriving late. The other bispecifics in this disease already have step-up protocols and hospital monitoring built into practice.

You can arrive late if you're easier to give. The commercial question for the next eighteen months isn't whether the drug works. It's whether the first dose can happen where the patient already is.

Full data lands in a plenary at the myeloma society meeting in Glasgow on September twenty-fifth. AbbVie says it will take the results to regulators. Maya, what are you watching for in that plenary?

The infection numbers, not the response rate. And the comparator arm. Forty-five point seven percent tells you how sick these patients are. That's the bar this drug cleared. It's also the bar every bispecific after it gets measured against.

To the deal side. Cipla's InvaGen arm signed an exclusive US licence with Qilu for QL2107, a pembrolizumab biosimilar. Qilu does development, registration and supply. Cipla USA sells it, subject to approval.

Cipla bought the channel, not the molecule.

And their North America chief said so out loud — he called it a biosimilar portfolio play. The bigger one today is Gan and Lee. They've licensed bofanglutide to Menarini for thirty-nine European countries. Sixty-two million euros upfront, up to seven hundred and twenty-six million in total before royalties. And the hook is dosing. Once every two weeks, which is half the injections of a weekly GLP-1.

An interval only differentiates if the curve holds at that interval. What Menarini bought is a Chinese Phase 3 and a US Phase 2. Both met their primary endpoints. That's permission to run a global program.

I'd still take the convenience argument.

Take it, and price it honestly. Gan and Lee say they intend to start that global Phase 3 to support European registration. Until it reads out, Menarini has bought a convenience argument and not a label claim.

Two deals, one shape — and the day's biggest licence has it too. The originator keeps the development risk. The partner takes the territory. The money waits behind milestones. If you're modelling any of these, model the milestone schedule. That's the only place a buyer's real conviction gets disclosed.

The FDA approved Ionis's Zanvastro, the first treatment for Alexander disease. It landed more than two weeks ahead of the September twenty-second decision date. It's an antisense drug that lowers GFAP, the protein that drives the disease.

We flagged that date on Tuesday's calendar.

And an early approval is its own signal. The pivotal trial had forty-nine patients aged five and up. Treated patients held their walking speed steady over sixty-one weeks. Untreated patients declined. And in a substudy of four children under two, motor skills improved rather than stabilised.

Improved. Not just held. Forty-nine patients, in a disease that affects one in three million people. Nobody builds a revenue model on that. This is halo math. The value isn't the sales line. It's what an antisense drug just proved it can do in a leukodystrophy.

The halo doesn't decide this one. The label covers all ages, and that's the asset Ionis actually won. The open question is payers. Do they treat a flat line in a degenerative disease as the win the FDA just called it?

Quick note before our second story — if The Pharma Closeout is how you close out your pharma day, follow the show on Spotify or Apple Podcasts. Now — GSK and HUTCHMED. A hundred and ten million dollars upfront for HMPL-A830, an antibody-targeted conjugate hitting EGFR and KRAS at once. Worldwide rights outside China, Hong Kong, Macau and Taiwan. HUTCHMED jumped sixteen percent in London. GSK barely moved.

Sixteen percent, on one licence, for a molecule never tested outside China. HUTCHMED runs the global Phase 1 itself, and it starts in the second half of this year. GSK committed before the first ex-China patient is dosed.

And the structure says so. Up to one point two nine five billion in total, and under a tenth of that is cash today. The other one point one eight five billion is contingent. GSK isn't buying an asset here. It's buying an option on a mechanism, at option prices.

Then be precise about the option. One antibody, one payload, two mechanisms. I run the second-asset test on deals framed this way, and there is no second asset. There's one molecule and a guess about which tumours carry both targets. Nobody has shown us that population yet. That's the enrolment question the Phase 1 has to answer before a single milestone becomes real.

That's the shape to know if you're doing oncology business development this quarter. Small cheque. Originator keeps the early clinical work. Value parked behind data that doesn't exist yet. HUTCHMED just got repriced sixteen percent on a hypothesis, and the first dose settles it.

On AbbVie, don't track the response rate. Track whether regulators accept the outpatient claim. The safety data is the entire argument behind it. That's the difference between an approval and a market.

On HMPL-A830, that global Phase 1 starts in the back half of this year. It's the first read outside China on the mechanism GSK just paid for. And on Zanvastro, watch the price. A one-in-three-million disease with an all-ages label is where rare-disease pricing gets tested in public.

And watch whether today's pattern holds through the fourth quarter. Originator keeps the trial. Partner takes the territory. If it does, the buy-side conversation about Chinese-originated assets changes. It stops being about price. It starts being about who's running the study.

The one I'm sitting with tonight is AbbVie. A seventy-four percent response rate, and the company chose to sell the setting instead. When a class gets this crowded, whoever gets out of the hospital first wins. Have a good evening.

And that is your Pharma Closeout for Thursday, September 3rd. Two of today's biggest stories weren't about whether the molecule works. AbbVie bought a setting. GSK bought a first dose it hasn't seen yet. 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.