Something shifted in China-to-West licensing between 2024 and 2025, and the numbers are stark enough to be worth taking seriously. In H1 2025, US and European companies signed 14 licensing agreements for Chinese assets worth up to $18.3 billion. In H1 2024, that number was two deals. Every large-cap pharmaceutical company has now sourced from China. The question that matters for everyone below that tier is what this actually means for how deals are structured, priced, and closed.
The Bargain Era Is Over
The first thing the data shows clearly is that the window of opportunistic pricing has closed. Average upfront payments for China-origin assets have risen three to five times since 2021. Phase II oncology assets are now commanding $200 to $500 million upfront from Big Pharma buyers. The Merck licence for Hansoh’s oral GLP-1 closed at $112 million upfront against $1.9 billion in milestones. Merck’s deal for LaNova’s PD-1/VEGF bispecific came in at $588 million upfront.
These are not mid-market deals. But they set the reference point that Chinese originators now use when they think about what their asset is worth. The companies that are still approaching China sourcing with 2021 pricing expectations will find that the conversations have changed.
Five Pathways, Two That Are Realistic for Mid-Market
The deals that make headlines use structures that are not necessarily available to mid-sized EU generics or specialty pharma companies. It is worth being clear about which is which.
Out-licensing remains the dominant structure and the one most relevant to SEQOVA’s partners. The Chinese originator retains China rights; the Western partner takes ex-China development and commercialisation rights in exchange for upfront, milestones, and royalties. This structure works because it aligns incentives cleanly: the originator keeps its home market economics, the Western partner takes on global development without acquiring a company.
Option-to-license is increasingly used for preclinical or early Phase I assets where the Western partner wants to see clinical proof-of-concept before committing fully. GSK’s deal with DualityBio for a preclinical ADC opened with a $30 million option payment against up to $975 million post-exercise. For smaller companies evaluating early-stage Chinese assets, this structure preserves the right to walk away before the heavy capital is committed.
The other structures — co-development and profit-share, NewCo platform deals, and outright acquisition — are largely the territory of Big Pharma and specialist VC. BMS’s 50:50 global profit-share with BioNTech came with a $1.5 billion upfront and $7.6 billion in milestones. AstraZeneca’s acquisition of Gracell sits at the upper bound of what acquisition looks like at around $1.2 billion, structured as a “license first, acquire later” move. These are not structures a 200-person EU specialty pharma company is running.
Knowing which pathway is realistic is not a limitation. It is what makes a mandate specific enough to act on.
What the Regulatory Data Actually Says
Four bottlenecks consistently appear in deals that stall or fail, and only one of them gets the attention it deserves.
The CMC and GMP gap is discussed in most due diligence frameworks. China’s FDA inspection observation rate sits at 69%. The gap between what NMPA requires and what EMA or FDA expects in process validation documentation remains a genuine source of deal delay, not just theoretical risk.
Less discussed is the clinical data acceptance question. The sintilimab ODAC case in 2022 — a 14:1 vote against approval — established that China-only trial data needs applicability to Western populations, appropriate comparators, and overall survival endpoints to support a US filing. Assets built on China-only Phase III data with no bridging plan are not as de-risked as their NMPA approval status suggests.
This points to a broader bridging reality. NMPA approval establishes that a product works and is manufactured to Chinese standards. It does not answer what FDA or EMA will ask. On the CMC side, process validation documentation, analytical methods, and batch record formats typically need supplementation or reformatting to meet ICH Q8/Q9/Q10 standards. On the clinical side, a bridging strategy needs to be planned before the deal closes, not after. Data integrity is a third layer: FDA has applied heightened scrutiny to audit trails and documentation standards in China-generated data since 2022, and gaps here can trigger inspection delays regardless of the underlying science.
For US-bound deals specifically, the BIOSECURE Act signed into law in December 2025 adds a CDMO and supply chain dimension that did not exist two years ago. The Act does not prohibit licensing of Chinese-origin assets, but it restricts federal procurement chains and affects CDMO relationships with certain named entities. Any deal that depends on a Chinese CDMO for US commercial supply needs to map this before the term sheet is signed.
IP freedom-to-operate is the fourth area. Composition-of-matter, antibody sequence, and linker-payload patents all require diligence in Western jurisdictions regardless of what the NMPA filing shows. China’s data exclusivity period of six years is shorter than the US and EU equivalents, which affects how Chinese partners think about the value of a licensing deal versus waiting for their own Western registration.
What This Means for Companies Not at the Top of the Market
The data is dominated by billion-dollar deals between large-cap buyers and well-capitalised Chinese biotechs. But the pipeline below that level is larger than the headlines suggest, and the deal structures available there are different.
For mid-sized EU and US companies, the realistic landscape is out-licensing and option-to-license deals for assets at NMPA-registered or late-stage clinical status, in categories where EU and US commercial infrastructure already exists. The economics are more modest, the assets are less competed, and the qualification work required before a term sheet is issued is more intensive.
That last point is where most mid-market deals either move or stall. A Chinese manufacturer holding an NMPA-approved injectable generic of a shortage-listed cancer drug is not automatically a deal-ready asset. The GMP gap analysis, the API supplier CEP status, the QP audit readiness, the bridging data plan — these need to be mapped before serious engagement begins, not during technology transfer.
The companies that do this work early, on assets that have not yet been picked up by the large-cap buyers, are operating in a part of the market where first-mover advantage is still real.