The Year’s Two Largest Conjugate Deals Were Not ADCs, and Not Cancer Drugs

Buyers have stopped paying for conjugate drugs and started paying for conjugate chemistry. Across 301 disclosed conjugate transactions since January 2021, this year breaks the pattern of every prior one: flat volume, roughly double the money, and seven in ten deals signed before a molecule exists. The antibody-drug conjugate is now one output of a toolkit that has already been sold into influenza and muscular dystrophy.

Flat volume at double the money is not a cooling market, it is a concentrating one

The headline count says the conjugate boom has levelled off. Thirty-eight deals were announced in the 1 January to 21 July window this year, against forty in the identical window of 2025. Read alone, that is a plateau, and most sector commentary has read it alone.

The money says otherwise. Disclosed upfront consideration over the same window went from $3.36bn spread across eleven transactions to $6.34bn across seven. Mean disclosed upfront moved from roughly $305m to roughly $906m. Fewer buyers are writing cheques, and the cheques are three times the size.

Two transactions carry most of this year’s total. Gilead closed its acquisition of Tubulis on 21 May 2026 at $3.15bn upfront with up to $1.85bn contingent, keeping the Munich site as a dedicated conjugate research organisation rather than folding it into an existing unit. Three days after that deal was announced, Pfizer and Innovent signed a twelve-programme collaboration on 28 May 2026 at $650m upfront and up to $9.85bn in milestones, covering eight Innovent-originated early-stage programmes and four Pfizer discovery programmes. Neither buyer purchased a marketed product. Both purchased the capacity to generate them.

The payload class is the asset now, and it is bought before it reaches a patient

Novartis agreed on 6 July 2026 to acquire Myricx Bio for $1.1bn upfront and up to $400m in milestones. What it bought was a payload chemistry: N-myristoyltransferase inhibitors, positioned explicitly against the limits of the topoisomerase-1 inhibitor class that sits inside almost every competing conjugate. The two lead assets are preclinical. The payload platform is the thesis.

Johnson and Johnson completed its purchase of Firefly Bio on 29 July 2026 for $1bn in cash, acquiring the Firelink degrader antibody conjugate platform aimed at KRAS-driven tumours. The accounting is the tell. The company is treating it as an asset acquisition, booking an in-process research and development charge of approximately $1bn in the third quarter and guiding to roughly $0.46 of dilution to 2026 adjusted earnings per share. A billion dollars written off on day one, for chemistry with no clinical package.

Underneath the headline transactions, a structural change is running. The file contains nine dual-payload conjugate deals, and every one of them was signed after January 2025. Two mechanisms on one antibody was a research curiosity eighteen months ago. It is now a routine deal object.

Conjugate chemistry left oncology before the sector narrative caught up

Of the 301 deals in the file, 248 are oncology. On count, this remains a cancer story, which is why every industry graphic frames it as an ADC arms race. On capital, that framing is already wrong.

Novartis completed its acquisition of Avidity Biosciences on 27 February 2026 at $72.00 per share, valuing the company at approximately $12bn. The asset is an antibody-oligonucleotide conjugate platform delivering RNA to muscle, with late-stage programmes in myotonic dystrophy, Duchenne and FSHD. Merck completed its $9.2bn acquisition of Cidara Therapeutics on 7 January 2026 for CD388, a drug-Fc conjugate carrying multiple copies of a small-molecule neuraminidase inhibitor on an antibody fragment, in Phase III for influenza prevention. Merck’s own description is unusually precise on the point: these are not vaccines and not monoclonal antibodies.

Those two total $21.2bn. Every conjugate acquisition in oncology over the same twelve months totals $5.68bn upfront combined. The largest cheques in conjugate chemistry are being written outside cancer, for constructs where the antibody is a delivery chassis rather than the therapy.

Payers are still litigating the last conjugate while the next forty are assembled

The most commercially successful conjugate in the world is still not routinely funded in England for its flagship indication. Trastuzumab deruxtecan was rejected for HER2-low metastatic breast cancer in 2024, approved in Scotland, and remains unavailable to English patients more than two years later. NICE moved its cost-effectiveness threshold in 2026 and the position has not changed. An Early Day Motion tabled in the 2026-27 session, signed by 32 members, urges Daiichi Sankyo, AstraZeneca, NHS England and NICE to reach a deal.

That is the environment receiving the output of a platform-stage land grab. Payload convergence gives payers one lever: where several conjugates carry near-identical topoisomerase-1 chemistry into overlapping lines, class competition becomes possible. Novel payload classes such as NMTi are bought precisely to escape it.

The dual-payload construct removes the lever entirely. A combination regimen is negotiable because a payer can price each component and discount the weaker one. A single molecule carrying two mechanisms arrives as one product code, one dossier, one price. Nine such deals have been signed since January 2025 and none has yet reached a reimbursement decision. When the first does, the assessment methods currently used for combination therapy will have no obvious place to attach.