Hansoh Pharmaceutical Group 翰森製藥: China's Licensing Machine

I. Introduction & Episode Roadmap

Start with a receipt, not a thesis.

On October 20, 2023, GSK agreed to pay a Chinese drugmaker most Western investors had never heard of $85 million in cash for the worldwide rights — outside Greater China — to an experimental cancer medicine called HS-20089, with up to $1.485 billion more contingent on it working.¹ Two months later, on December 20, GSK came back for a second molecule from the same company, HS-20093, this time putting $185 million on the table and up to $1.525 billion in milestones behind it.² In December 2024, Merck & Co. wired $112 million upfront for a pill that had never been in a human being.³ In June 2025, Regeneron signed for a metabolic asset in a deal worth up to $2.01 billion. In October 2025, Roche did the same for a colorectal cancer candidate at up to $1.53 billion.⁴

Add it up and you get north of $8 billion of potential deal value handed to one company in roughly twenty-four months by four of the most sophisticated buyers of pharmaceutical science on Earth. The company is 翰森製藥集團 Hansoh Pharmaceutical Group Company Limited (3692.HK), headquartered in Lianyungang, a mid-sized port city in Jiangsu province that almost no one outside China could find on a map.

The question this article exists to answer is not whether those deals happened. They did, and the paperwork is public. The question is what they mean. There are two readings, and they lead to very different places.

Reading one: Hansoh cracked a genuinely hard problem. It built an in-house discovery engine capable of producing molecules that Big Pharma's own scientists judged better than what they had internally — at Chinese cost, on Chinese timelines — and it is now monetizing that engine globally without spending a dollar on a Western sales force.

Reading two: Hansoh is a well-run participant in a sector-wide arbitrage. Western pharma is staring down a patent cliff, its internal R&D productivity is poor, and Chinese assets are cheap. China's share of global biopharma out-licensing deals rose from 5% in 2020 to 21% in 2024 and 32% in the first half of 2025, and Chinese assets have historically commanded upfront payments 60–70% below comparable Western ones.⁵ In that reading, Hansoh's deal flow is less a verdict on Hansoh than on the exchange rate between Chinese and Western science — and exchange rates move.

The numbers as of today: for the year ended December 31, 2025, Hansoh reported revenue of approximately RMB15,028 million, up 22.6%, and profit of approximately RMB5,555 million, up 27.1%, on basic earnings per share of RMB0.93.⁶ Innovative medicines and collaborative products contributed RMB12,354 million, or 82.2% of the top line — a figure that was a rounding error a decade ago. R&D expenditure reached RMB3,358 million, 22.3% of revenue.⁶ The company has been listed on the Hong Kong Stock Exchange since June 2019 and carried a market capitalization of about HK$209 billion — roughly US$27 billion — as of August 10, 2026, at a trailing P/E in the low thirties.⁷ Zhong Huijuan and her family control roughly two-thirds of the equity.⁸

The road from there to here runs through a chemistry teacher who left a classroom in 1995, a generics business that Beijing systematically dismantled, a lung cancer pill that became the first China-developed EGFR inhibitor ever launched in the West, and a business development team that learned to sell molecules the way Foxconn sells assembly.

A note on how to read what follows. This is a company where the most important analytical work is not gathering facts — Hansoh's Hong Kong filings are detailed and reasonably candid — but sorting them. Almost every headline number in this business has two components: one that recurs because patients take medicine, and one that arrived because a contract was signed. Management reports both, correctly, inside a single growth rate. The reader's job is to pull them apart, and most of the analytical tension in the sections ahead comes from doing exactly that.

The structure runs chronologically until it doesn't. Origins and the IPO are handled briskly, because the interesting decade is the current one. The volume-based procurement shock gets real space because it explains why the company exists in its present form. The out-licensing engine gets the most space of all, because it is where the value, the risk, and the unresolved questions now live.

↑ Back to Top

II. Origins: A Chemistry Teacher Builds a Generics Empire (1995–2010s)

Lianyungang in the mid-1990s was not a place where global pharmaceutical companies came from. It was a coastal port in northern Jiangsu, a second-tier city in a province whose economic center of gravity sat two hundred kilometers south in Nanjing and Suzhou. It had salt flats, a harbor, and a state drug administration bureau.

Zhong Huijuan 钟慧娟 worked in and around that world. She took a bachelor's degree in chemistry from Jiangsu Normal University in 1982, taught chemistry at a local middle school, and later worked at the drug administration bureau in Lianyungang before founding Hansoh Pharmaceutical Group in 1995 with backing from the investor Cen Junda.⁸ She was in her mid-thirties. She was leaving a stable, respectable job in a country where stable, respectable jobs were the entire point.

The detail that matters most about her background is not the entrepreneurial romance of it. It is the chemistry. Zhong understood, at a technical level, what a molecule was and what it took to make one reliably at scale — which is a materially different starting position from the trading and distribution backgrounds that produced many of China's early pharmaceutical fortunes. Her husband, 孙飘扬 Sun Piaoyang, was building 江苏恒瑞医药 Jiangsu Hengrui Pharmaceuticals in the neighboring city of Lianyungang as well; by November 2025 Forbes valued his fortune at $13.6 billion, separately from hers.⁸ Two of China's largest pharmaceutical companies were founded, and are still controlled, by one married couple. That is a governance fact worth returning to.

Hansoh's first two decades were a generics business, and a very good one. The company built a position in central nervous system drugs — olanzapine, paliperidone, agomelatine, the workhorse antipsychotics and antidepressants of Chinese hospital psychiatry — and in oncology chemotherapy staples like pemetrexed, gemcitabine and imatinib.⁹ These were not glamorous molecules. They were off-patent compounds where the competitive game was manufacturing quality, regulatory compliance, and, above all, a hospital sales organization that could get a product onto formulary in thousands of institutions across a country with the population of Europe and the Americas combined.

That sales organization is the underappreciated asset of the early era. Chinese pharmaceutical commercialization is a ground war: provincial tenders, hospital-by-hospital listing decisions, physician education, a distribution chain of enormous complexity. Building it took twenty years. It cannot be replicated by a venture-funded biotech in three.

The CNS franchise deserves a specific note, because it shaped the company's temperament. Psychiatric medicine in China through the 2000s was a hard, low-glamour, high-friction business. Antipsychotics are prescribed by a relatively small, specialized physician base in institutions that are chronically underfunded; the patients are stigmatized; the products require long-term adherence and careful titration. Winning share there meant building genuine clinical relationships over years rather than buying them with a marketing budget. Hansoh became China's largest supplier of psychotropic medicines by doing exactly that. It is not a coincidence that the same organization later proved capable of getting a novel oncology drug onto formulary at speed.

There is also a structural point buried in the geography. Lianyungang, together with the surrounding corridor of northern Jiangsu, became one of the densest pharmaceutical manufacturing clusters in China — a concentration of API plants, formulation facilities, quality-control laboratories, and, crucially, trained technical staff. Two of China's largest drugmakers emerging from the same small city is less improbable than it sounds once you account for the cluster. It is the pharmaceutical equivalent of Shenzhen's electronics supply chain: not a single company's advantage, but an ecosystem that lowered the cost of every subsequent decision Hansoh made.

The second inheritance from the generics era is the one management talks about most, and it deserves scrutiny rather than applause. Hansoh reinvested an unusually high share of revenue into R&D for a Chinese generics manufacturer, and kept escalating it: R&D expenses rose to RMB2.10 billion in 2023, roughly 21% of revenue, and reached RMB3,358 million, or 22.3% of revenue, in 2025.⁹⁶ By comparison, a typical global generics manufacturer spends in the mid-single digits. The claim that this discipline "predates the crisis" is testable, and it survives the test: the spending ramp is visible in filings from years before the policy shock that made it necessary.

What the generics era did not produce was a novel drug. Through the 2000s and most of the 2010s, Hansoh's R&D was largely process chemistry, formulation, and incremental improvement — real capability, but not discovery. The transition from "we can make anyone's molecule better" to "we can find a molecule no one else has" is the hardest jump in the industry, and Hansoh had not yet made it when it decided to raise public capital.

It is worth being precise about why that jump is so hard, because it explains the pace of everything that follows. Generic manufacturing is a known-answer problem: the molecule works, the dose is established, the regulatory path is defined, and the only question is whether you can make it purely and cheaply. Discovery is an unknown-answer problem with a base rate of failure above 90% and a feedback loop measured in years. The organizational muscles required are almost opposites — one rewards process discipline and cost control, the other rewards tolerance for expensive failure. Most companies that try to convert from the first to the second fail, not because they lack money but because the internal culture cannot absorb a decade of research spending with nothing to show.

Hansoh's answer was to buy time with cash flow and to build the discovery organization physically apart from the legacy business — Shanghai for biologics and platform chemistry, Maryland for a Western scientific presence, with Lianyungang and Changzhou anchoring process development and manufacturing. Separating discovery from operations is a well-worn corporate device and it fails more often than it works. Here it worked, though the evidence for that arrived only much later.

The company that walked into the Hong Kong Stock Exchange in 2019 was, therefore, a paradox: enormously cash-generative, commercially formidable, technically competent — and sitting on a product base that a single policy decision in Beijing could render nearly worthless.

↑ Back to Top

III. The IPO and the Pivot Nobody Was Forced Into (2019)

June 14, 2019 was a strange day to list a Chinese pharmaceutical company. The trade war was escalating. Hong Kong's streets were filling with protesters. And the domestic pharmaceutical industry was six months into a policy experiment that most investors had not yet priced.

Hansoh priced anyway, and priced well: roughly 551 million new shares at HK$14.26, the top of an indicated HK$13.06–14.26 range, raising approximately HK$7.9 billion, or about US$1 billion.¹⁰⁸ It was among Hong Kong's largest offerings of the year. On listing, Zhong Huijuan became — by the standard league-table arithmetic — the richest self-made woman in the world.

Here is the part that deserves interrogation. Hansoh did not need the money in any conventional sense. The generics business was throwing off cash. The company carried minimal debt. There was no acquisition on the table. Management's stated purpose was to fund the innovative-medicine transition, and the honest analytical framing is that this was a pre-emptive raise: capital taken on before the balance sheet demanded it, to fund a transformation before the market forced it.

Was that foresight or luck? The fair answer is: partly foresight, mostly industry-wide reading of an obvious signal. Hansoh was not alone. Hengrui had been shifting toward innovation for years. Innovent, BeiGene and a cohort of Hong Kong-listed biotechs had raised capital under the exchange's new Chapter 18A regime for pre-revenue biotech. The entire Chinese pharmaceutical sector understood by 2018 that Beijing intended to squeeze generic-drug margins and redirect the savings toward innovation. That was stated policy, not a secret.

What distinguished Hansoh was not the insight but the starting position. It entered the transition with a decade of accumulated R&D spend, a manufacturing base, a national sales force, and — critically — an existing profit stream to fund the journey. The pure-play biotechs had science and burn rates. Hansoh had science, cash flow, and distribution. That asymmetry is the closest thing to a structural advantage in this story, and it is worth holding onto as the rest unfolds.

The Hong Kong context matters too. In 2018 the exchange had introduced Chapter 18A, permitting pre-revenue biotechnology companies to list — a rule change that transformed Hong Kong into the primary funding venue for Chinese life sciences almost overnight and drew a wave of loss-making biotechs to the market. Hansoh listed alongside that cohort but was categorically different from it: profitable, cash-generative, and selling an established product portfolio. Investors buying the 18A names were buying pure clinical optionality. Investors buying Hansoh were buying a real business with an option attached, which is a materially different risk profile and one reason the stock has behaved differently from the sector's more speculative names through subsequent drawdowns.

There is one detail about the offering worth flagging for anyone reading older coverage. The gross raise was large, but the free float was small — the company sold a minority slice of a business it had no intention of ceding control over, and the practical amount of stock available to trade was a fraction of the headline. Low float is a double-edged instrument: it supports the share price on the way up and makes it violent on the way down, and it is one reason Hansoh's stock has historically moved more than its fundamentals in both directions.

The IPO also locked in something else: control. The offering was a minority float. Zhong and her family retained roughly two-thirds of the shares, held principally through a family trust structure that stood at 65.7% of the register as of mid-2024.⁹⁸ Public shareholders bought into a company where every material capital allocation decision would be made by one family, with no realistic mechanism to overrule it. In 2019, with the pivot unproven, that was a leap of faith. Whether it has been rewarded is a question for a later section.

Within months of the listing, the policy squeeze that everyone had seen coming arrived with a force that almost no one had modeled correctly.

↑ Back to Top

IV. The VBP Shock: When Beijing Deleted the Generics Business (2019–2023)

To understand what happened next, you need to understand a piece of Chinese healthcare policy with an unlovely name: 集中带量采购, National Volume-Based Procurement, universally shortened to VBP.

The mechanics are elegant and brutal. Historically, Chinese hospitals bought drugs through fragmented provincial tenders where manufacturers competed on relationships and marketing spend rather than price — a system that produced enormous sales forces, enormous promotional budgets, and generic drug prices that were, by international standards, absurdly high. VBP inverted it. The state pools the demand of the entire public hospital system for a given molecule, runs a national reverse auction, and awards guaranteed volume — typically the majority of national demand — to the lowest bidders. Win, and you get scale you could never buy. Lose, and you are effectively excluded from the hospital channel.

The price consequences are not incremental. Winning bids routinely land 80–95% below pre-tender prices. In Hansoh's own disclosure, the lowest VBP-cleared price for certain of its legacy molecules fell to figures like RMB11.7, RMB8.7 and RMB3.6 per month of therapy.⁹ That is not margin compression. That is the deliberate conversion of a branded-generic business into a commodity utility.

The policy logic, from Beijing's side, is coherent and worth stating fairly. China spends a rising share of a strained public insurance fund on medicines whose patents expired decades ago, sold at prices inflated by a promotional system that added no clinical value. VBP redirects that spending. The savings fund reimbursement for genuinely novel medicines through the National Reimbursement Drug List, and the policy simultaneously destroys the economic rationale for the enormous generic sales forces that had grown up around the old tender system. It is industrial policy executed through procurement, and by its own objectives it worked.

The effect on Hansoh's revenue mix is the single most important number in the company's modern history. Generic drugs fell from 82.0% of revenue in 2020 to 32.1% in 2023.⁹ In three years, four-fifths of the business became one-third of it.

It is worth sitting with what that means operationally. This was not a demand shock — patients still needed olanzapine and pemetrexed. It was a price shock imposed by a monopsony buyer with the legal authority to set terms. There was no negotiation, no substitution, no pricing power to exercise, no brand equity to defend. The Chinese state simply decided that Hansoh's largest business should earn less, and it did.

Was this mismanagement? No. It hit the entire industry simultaneously, and academic work on VBP has documented its market-concentration effects across the Chinese pharmaceutical sector.¹¹ Hengrui went through the same purge. So did every domestic manufacturer with a meaningful off-patent portfolio. Judging Hansoh's management on the fact of the decline would be like judging an oil producer for the price of Brent.

What management can be judged on is the response, and here the record is genuinely strong — with an important caveat. Hansoh did not attempt to defend the generics business through price wars or lobbying. It let the legacy portfolio reprice, kept the manufacturing and commercial infrastructure, and redirected essentially all incremental investment into innovative medicines. Innovative drug sales reached RMB6.87 billion in 2023, up 37.1%, taking them to 68% of revenue from just 18% in 2020.⁹ The crossover happened inside four years.

The caveat is timing, and it cuts against the heroic reading. The R&D that produced Hansoh's innovative portfolio was committed years before VBP bit. Aumolertinib was approved in March 2020 — meaning it had been in development since the mid-2010s.⁶ The company did not pivot in response to the shock; it happened to already be halfway across the river when the bridge burned. That is a meaningfully weaker claim than "management saw it coming and acted," and it should temper any narrative of strategic genius. What Hansoh demonstrated was not prescience but the absence of a fatal error: it did not double down on a dying business, and it had a second act ready.

For investors, the durable lesson is about the shape of Chinese pharma risk. The buyer in this market is the state, and the state's objectives are patient access and fiscal restraint, not manufacturer profitability. Any thesis on a Chinese drugmaker that does not price policy as a first-order variable is incomplete.

There is a second-order consequence worth flagging, because it explains the out-licensing chapter that follows. VBP did not merely reduce Chinese generic prices; it compressed the entire domestic pharmaceutical profit pool and, in doing so, made overseas revenue structurally more attractive to every Chinese manufacturer simultaneously. A Chinese company that develops a novel drug faces NRDL negotiation at home — meaning volume at a negotiated discount — versus Western markets where the same molecule can command multiples of the Chinese price. The gap between what a drug earns per patient in Shanghai and what it earns per patient in Boston is the single largest economic force acting on Chinese pharmaceutical strategy today. Hansoh's decision to sell ex-China rights to almost everything is not primarily a story about scientific validation. It is a story about where the money is.

Which raises the obvious question about the new business: if Beijing deleted the economics of generics, what stops it doing the same to innovative drugs? The honest answer is: nothing structural, only a difference in stated policy intent. NRDL negotiation is gentler than VBP by design, because the state wants domestic innovation to be profitable enough to continue. But it is the same buyer with the same fiscal constraint, and the discounts extracted at each renewal are real.

↑ Back to Top

V. Building the New Core: Innovative Drugs and Almonertinib

In a lung cancer clinic in Shanghai, a patient with non-small cell lung cancer gets a biopsy sequenced. The report comes back with a mutation in a gene called EGFR — the epidermal growth factor receptor. In Chinese lung cancer patients, this happens roughly 40–50% of the time, far more often than in Western populations. That single epidemiological fact is the foundation of Hansoh's new business.

Here is the biology in plain terms. EGFR is a protein that sits on the surface of a cell like a doorbell. When the right signal presses it, the cell divides. In these tumors, a mutation jams the doorbell in the "pressed" position, so the cell divides continuously. A drug called an EGFR tyrosine kinase inhibitor — a TKI — is a small molecule that slips inside the cell and blocks the wiring behind the doorbell.

First-generation TKIs worked, for about a year. Then tumors evolved a new mutation, T790M, that changed the shape of the lock. Third-generation TKIs were designed to fit the changed lock and the original one, while largely ignoring the normal EGFR on healthy cells — which is what limits the rash and diarrhea that made earlier drugs miserable to take.

Hansoh's entry is Ameile 阿美樂 (aumolertinib mesylate, also written almonertinib, internal code HS-10296), the first original third-generation EGFR-TKI developed in China.⁶ It was approved by the NMPA in March 2020 for T790M-positive patients who had progressed on prior therapy, and in December 2021 as a first-line treatment for patients with the two most common activating EGFR mutations.⁶

What has happened since is a case study in how a single molecule becomes a franchise. Hansoh has spent five years methodically expanding the label into earlier and larger patient populations. In March 2025, a third indication was approved for unresectable stage III disease that had not progressed after chemoradiotherapy. In May 2025, a fourth covered adjuvant treatment after surgical resection — the earliest, largest and most durable segment of the market. In January 2026, a fifth added combination use with chemotherapy in the first line.⁶

The supporting data are strong on their face. In the ARTS study of adjuvant treatment after complete resection, disease-free survival improved with a hazard ratio of 0.17 and an investigator-assessed two-year DFS rate of 90.2%. In AENEAS2, adding chemotherapy to aumolertinib in the first line extended median progression-free survival to 28.9 months with a hazard ratio of 0.47 versus monotherapy and an objective response rate of 93.2%.⁶ Both were presented at the 2025 AACR annual meeting. A hazard ratio of 0.17 in an adjuvant setting is a large effect; the appropriate skepticism is that adjuvant DFS benefits in EGFR-mutant lung cancer have historically translated into overall survival benefits slowly and incompletely, and these are company-reported readouts of company-sponsored trials in a single-country population.

There is a competitive fact that neither the company nor most bullish commentary emphasizes enough: aumolertinib is not the only third-generation EGFR-TKI on the Chinese market, and it was not the first. AstraZeneca's osimertinib — Tagrisso — defined the class globally and arrived in China ahead of it, and several other domestic third-generation inhibitors have since been approved. Aumolertinib's clinical positioning rests on a comparable efficacy profile with a differentiated tolerability story, and its commercial positioning rests on price, reimbursement and a Chinese sales organization operating on home ground.⁹ That is a real but replicable set of advantages. It also means the drug's long-run Chinese economics are governed by NRDL renewal cycles and eventual generic entry rather than by any durable monopoly.

The patent-cliff question therefore sits underneath the whole franchise. A small molecule enjoys roughly two decades of patent life from filing, of which the commercially useful portion after approval is typically eight to twelve years. Aumolertinib was approved in China in 2020. Every indication expansion extends the revenue runway and deepens the clinical moat, but none of it changes the underlying compound patent clock. Investors modeling Hansoh's oncology base should treat aumolertinib as a decade-long annuity with a defined end, not a perpetuity — and should note that the company's entire strategic direction, toward a broad pipeline and offshore licensing, is a rational response to exactly that arithmetic.

The genuinely novel development is geographic. In June 2025, the UK's MHRA approved aumolertinib under the trade name Aumseqa — the first China-developed EGFR-TKI ever approved in a major Western market.¹²⁶ In February 2026, the European Commission followed, granting EU approval for both the first-line and T790M indications after a positive CHMP opinion.⁶ In December 2025, Hansoh licensed the drug to Glenmark Specialty for the Middle East and Africa, South and Southeast Asia, Australia, New Zealand, Russia/CIS and selected Caribbean markets, in a deal that could exceed US$1 billion in aggregate.⁶

Read carefully, that sequence tells you something important about strategy. Hansoh secured Western regulatory approval — the hard, expensive, credibility-conferring part — and then licensed commercialization to a partner in the mid-tier geographies rather than building its own field force. It is keeping the science and renting the distribution.

Behind aumolertinib sits a portfolio rather than a single asset, which materially changes the risk profile. Seven innovative medicines now generate product sales in China.⁶ Hansoh Xinfu 豪森昕福 (flumatinib) is China's first original second-generation TKI for chronic myeloid leukemia. XINYUE 昕越 (inebilizumab) — in-licensed from Viela Bio in May 2019, an asset that traveled through Horizon Therapeutics to Amgen — has now been approved in China for three indications: neuromyelitis optica spectrum disorder in March 2022, IgG4-related disease in August 2025, and generalized myasthenia gravis in March 2026.⁶ Fulaimei 孚來美 (PEG-loxenatide) was the world's first PEGylated weekly GLP-1 receptor agonist. Saint Luolai 聖羅萊 (pegmolesatide) treats renal anemia. Hengmu 恒沐 (tenofovir amibufenamide) treats hepatitis B.

The therapeutic mix as of FY2025 shows how concentrated the business has become: oncology generated approximately RMB9,974 million, or 66.4% of revenue; metabolic and other diseases RMB2,158 million (14.3%); anti-infectives RMB1,586 million (10.6%); and CNS RMB1,310 million (8.7%).⁶ Two-thirds of Hansoh is now an oncology company.

That concentration puts it directly against the most crowded, best-capitalized competitive set in Chinese pharma. Hengrui remains the scale leader: FY2025 revenue of RMB31.63 billion, up 13%, with innovative drug sales of RMB16.34 billion and R&D spending of RMB8.72 billion — 27.6% of revenue, more than double Hansoh's absolute R&D budget.¹³ Innovent Biologics grew faster off a smaller base, reaching RMB13.0 billion in 2025, up 38.4%, and turned its first full-year profit of RMB814 million.¹⁴ BeOne Medicines — the former BeiGene — operates on an entirely different plane commercially, with FY2025 revenue of $5.3 billion, up 40%, driven by $3.9 billion of global Brukinsa sales.¹⁵ Junshi Biosciences and Akeso round out the domestic field.

Where does Hansoh actually sit? On revenue, roughly half of Hengrui and a fifth of BeOne. On profitability, better than both relative to size: RMB5,555 million of profit on RMB15,028 million of revenue is a 37% net margin, against Hengrui's roughly 24% and Innovent's mid-single digits.⁶¹³¹⁴ On R&D intensity, mid-pack. On global commercial infrastructure, decisively behind BeOne, which sells Brukinsa itself in the United States and Europe.

One further comparison sharpens the picture. Hengrui's FY2025 licensing income was RMB3.39 billion, up 25.6% — larger in absolute terms than Hansoh's RMB2.12 billion of collaboration revenue, and a similar share of a much bigger business.¹³⁶ In other words, the out-licensing model that looks distinctive when you examine Hansoh alone is, at the industry level, the default strategy of every serious Chinese innovator. Hansoh is executing a common playbook well, not running a proprietary one.

That last gap is the honest structural limitation. BeOne chose to build a Western commercial organization and absorbed years of losses to do it; it now captures the full economics of a $3.9 billion global product. Hansoh chose to license. The licensing model is capital-efficient and lower-risk, but it caps the upside: a royalty on a partner's sales is a fraction of the profit on your own. Hansoh's model wins on return on invested capital and loses on terminal value per molecule. Whether that trade is correct depends entirely on whether the pipeline is deep enough that optionality matters more than depth on any single asset.

And the pipeline is where the last three years of this story actually live.

↑ Back to Top

VI. The Out-Licensing Machine: Turning Chinese ADCs Into Global Cash (2023–2025)

Picture the meeting. A GSK oncology team flies to Shanghai in 2023 to look at early clinical data on a molecule targeting a protein called B7-H4. The presenting company has never launched a drug outside China. Its executives are largely unknown in London and Philadelphia. And within months, GSK signs — twice.

To understand why, you need to understand what an ADC is, because ADCs are the reason this section exists.

An antibody-drug conjugate is a guided missile. Chemotherapy is a carpet bomb: a cytotoxic poison that kills fast-dividing cells everywhere, which is why it destroys hair follicles, gut lining and bone marrow along with tumors. An ADC takes that same poison, attaches it via a chemical tether — the linker — to an antibody engineered to bind a protein that appears far more densely on cancer cells than on healthy ones. The antibody finds the tumor, the cell swallows the package, the linker releases the payload inside, and the poison does its work locally at a concentration that would be lethal systemically.

Three components, three places to fail. Pick the wrong target and you poison healthy tissue. Build the wrong linker and the payload sheds in the bloodstream. Choose the wrong payload or the wrong drug-to-antibody ratio and you get either no efficacy or intolerable toxicity. ADCs are, in other words, an engineering discipline layered on top of a biology discipline — enormous numbers of iterations, each one a molecule that has to be made, purified, and tested.

That is precisely the kind of problem where a large, well-trained, comparatively inexpensive Chinese research organization has a structural cost advantage. Hansoh runs more than 2,300 research staff across four R&D centers — Shanghai, Lianyungang and Changzhou in China, plus one in Maryland — and had more than 70 clinical trials running across more than 40 innovative candidates during 2025.⁶ It is not that Chinese scientists are better at ADC chemistry. It is that iteration is cheaper, and ADC development is iteration.

The deal sequence that followed reads as a repricing of that capability in real time.

GSK, October 2023 — HS-20089. A B7-H4-targeted ADC with a topoisomerase inhibitor payload, aimed at gynecologic cancers. $85 million upfront, up to $1.485 billion in milestones, tiered royalties, GSK taking worldwide rights outside Greater China. GSK's oncology head said publicly that the asset had "best-in-class potential" in ovarian and endometrial cancer.¹

GSK, December 2023 — HS-20093. A B7-H3-targeted ADC. $185 million upfront, up to $1.525 billion in milestones.² A repeat purchase from the same buyer within sixty days is the most informative signal in the sequence: GSK had now had two months of diligence inside Hansoh's chemistry, manufacturing and data, and came back for more.

Merck & Co., December 2024 — HS-10535. An oral small-molecule GLP-1 receptor agonist, still preclinical. $112 million upfront, up to $1.9 billion in milestones.³¹⁶ Merck paid nine figures for a molecule that had never been dosed in a human. That is a bet on the platform, not the asset.

It is also a bet with a specific commercial logic behind it. The obesity and diabetes market is being reshaped by injectable GLP-1 medicines, and the widely held view across the industry is that whoever delivers an effective oral version at scale captures an enormous incremental population — patients unwilling to inject, and health systems unwilling to pay for cold-chain distribution. Merck arrived at that race late and without a leading internal asset. Buying a Chinese preclinical oral candidate for $112 million is, from Merck's side, a cheap lottery ticket in the biggest lottery in pharmaceuticals. From Hansoh's side, it is $112 million of essentially costless revenue for a molecule with a high probability of never reaching a patient. Both parties can be making a good decision simultaneously; that is what an efficient options market looks like.

Regeneron, June 2025 — HS-20094. A dual GLP-1/GIP receptor agonist, licensed worldwide excluding Chinese Mainland, Hong Kong and Macau, for total consideration of up to US$2.01 billion.⁶

Roche, October 2025 — HS-20110. A CDH17-targeting ADC for colorectal and other solid tumors, worldwide excluding Greater China, up to US$1.53 billion.⁶¹⁷

Avere Therapeutics, July 2026 — HS-20118. The most structurally interesting of the set, and the newest. Hansoh licensed ex-Greater China rights to an oral cyclic peptide IL-23 receptor antagonist for plaque psoriasis, taking $120 million upfront and up to $2.18 billion in milestones — but also committed $320 million to a concurrent private placement as Avere merged into Nasdaq-listed NextCure, leaving Hansoh with an expected 30–40% of the combined company.¹⁸

Hansoh has also been a buyer. In March 2024 it in-licensed a bispecific antibody targeting EGFR and c-MET from Biotheus for up to RMB5 billion — roughly $695 million — in upfronts and milestones, and turned it into HS-20122, a bispecific ADC that entered the clinic in April 2025.¹⁹⁶ In April 2024 it in-licensed the IL-23p19 antibody HS-20137 from Qyuns for China, and in December 2025 it in-licensed a calcium-sensing receptor modulator, HS-10568, from Hengrui — a transaction between two companies controlled by the same married couple, which is precisely the kind of arrangement minority shareholders should want disclosed and priced at arm's length.⁶

Now the hard question: were these good deals for Hansoh?

The benchmark case is unavoidable. In October 2023 — the same month GSK signed the first Hansoh ADC — Merck agreed to pay Daiichi Sankyo $4 billion upfront plus $1.5 billion in continuation payments over 24 months, with total potential consideration of up to $22 billion, for three DXd ADCs. One of them, ifinatamab deruxtecan, was a Phase 2 B7-H3-targeted ADC.²⁰ Same target class, same payload chemistry class, same broad stage of development. Daiichi received billions at signing across three assets. Hansoh received $185 million for its B7-H3 asset.

You can construct defenses. Daiichi had a proven ADC platform validated by Enhertu, one of the most commercially successful oncology launches of the decade. Hansoh had early clinical data and no track record. The Daiichi package included three assets and deep co-development economics. All true. But the gap is roughly an order of magnitude, and it is consistent with the sector-wide observation that Chinese-originated assets clear at 60–70% lower upfronts than global equivalents.⁵ The most probable explanation is not that Hansoh's science was ten times worse. It is that in 2023 the market had not yet learned to price Chinese assets, and Hansoh — a first mover with no negotiating precedent and no comparable transactions to point to — took the discount.

The counter-evidence is that the discount has narrowed. Upfronts moved from $85 million to $185 million to $112 million for a preclinical asset to $120 million from a small partner, while headline values climbed from $1.5 billion to over $2 billion. And Hansoh's later deals increasingly capture equity or structural upside rather than pure cash — the Avere/NextCure transaction gives it a controlling-scale stake in a US-listed vehicle carrying its own molecule. That is a company learning to negotiate.

The Avere structure deserves a moment on its own, because it may be the most important strategic signal in the whole sequence. In a conventional out-licensing deal, Hansoh sells the asset's ex-China economics for cash and a royalty and has no further influence over how the molecule is developed. In the Avere transaction it took cash and wrote a $320 million cheque to own an expected 30–40% of a Nasdaq-listed company whose lead programme is Hansoh's own molecule.¹⁸ If HS-20118 succeeds, Hansoh captures milestones, royalties and a third of the equity value created. If it fails, Hansoh has lost $320 million rather than nothing.

That is a deliberate increase in risk in exchange for a claim on terminal value, and it is exactly what a company would do if it believed it had been selling too cheaply. It is also, notably, a way of establishing a US-listed vehicle and a Western development presence without building either from scratch — a partial answer to the structural limitation identified earlier. Whether it works is unknowable today; the fact that management chose it tells you how they read their own past deal terms.

It is also a company that remains a mid-tier seller in absolute terms. In October 2025, Takeda paid Innovent $1.2 billion upfront — including a $100 million equity investment — for two late-stage assets and an option on a third, in a package worth up to $11.4 billion.²¹ In May 2026, Bristol Myers Squibb and Hengrui announced strategic agreements across 13 programs with $600 million upfront and total value reported at $15.2 billion.²² Hansoh's deals are real, repeatable and profitable. They are not the largest in the market.

Which brings us to the accounting, and this is where an investor should slow down.

Hansoh reports revenue in two lines. In FY2025, pharmaceutical product sales were RMB12,913 million and collaboration revenue was RMB2,116 million — recognized, per the accounts, at a point in time.⁶ Collaboration revenue was RMB1,573 million in FY2024. So roughly 14% of 2025 revenue came from licensing, up from 12.8%, and the disclosure explicitly notes that the US$112 million Merck upfront was received during 2025 and included in collaboration revenue.⁶

Three implications follow, and management has not addressed them in as much detail as the accounting warrants.

First, the incremental margin on licensing revenue is close to 100%. An upfront payment carries almost no associated cost of sales. That means Hansoh's reported 37% net margin flatters the underlying product business: strip out RMB2.1 billion of near-costless collaboration revenue and the margin on the operating business is materially lower. The company's own cost of sales was RMB1,498 million, 10.0% of total revenue — an optically extraordinary gross margin that is partly an artifact of licensing income sitting in the numerator with no cost attached.⁶

Second, this revenue is lumpy by construction. It arrives when a deal is signed or a milestone is hit, not when patients take pills. The half-year split makes the point vividly: first-half 2025 revenue was RMB7,434 million, up 14.3%, with profit of RMB3,135 million, up 15.0%.²³ The full year came in at +22.6% revenue and +27.1% profit — meaning the second half grew far faster than the first, in the same period the Roche and Glenmark agreements were signed. Growth that accelerates because deals landed in December is not the same quality of growth as growth from prescription volume.

Third, and most importantly for anyone modeling forward: recurring product revenue and non-recurring deal revenue are being reported inside a single growth rate. Hansoh discloses the split, which is to its credit. But the headline the market reacts to is the consolidated number, and the consolidated number embeds an assumption that the deal engine keeps running at the same cadence. In a year with no major signing, the same underlying business would print a visibly worse result.

None of this makes the licensing income low-quality in an accounting sense — these are real cash payments from investment-grade counterparties for real assets, and the milestones convert to cash if the science works. It makes it differently quality. The right way to hold it is as a series of options, not an annuity. And options expire.

A related disclosure nuance is worth understanding, because it is easy to conflate two different things. Hansoh's headline "82.2% of revenue" metric refers to innovative medicines and collaborative products — a category that includes product sales of drugs Hansoh in-licensed from others, such as XINYUE, alongside its own discoveries and alongside collaboration income. The separate line item "collaboration revenue" in the revenue note is the licensing money specifically. The two are frequently blurred in commentary. An investor tracking the transition should watch the revenue note, not the highlights page.

There is also a working-capital tell in the 2025 accounts. The company's current ratio fell from approximately 10.2 to approximately 8.1 during the year, and management attributed the decline to increased contract liabilities.⁶ Contract liabilities are, in substance, cash received for performance obligations not yet delivered — consistent with upfront payments arriving ahead of the research and development work Hansoh owes its partners. That is not a red flag; it is exactly what a growing licensing book should look like on a balance sheet. But it does mean some of the cash on hand is spoken for, and it reinforces the point that licensing income is a forward commitment as much as a windfall.

The environment that generated them, meanwhile, keeps getting hotter. Chinese out-licensing hit a record $137.7 billion across 186 cross-border deals in 2025, against $13.9 billion in 2021, and by mid-February 2026 had already logged 38 transactions worth roughly $49 billion at an average size of about $1.3 billion — a 76% increase on the 2025 average.²⁴ For Hansoh as a seller, that is a tailwind on price and a headwind on scarcity. More Chinese ADC and metabolic assets competing for the same finite number of Big Pharma business development budgets is not obviously good for the terms of the marginal deal.

↑ Back to Top

VII. Management: Founder-Control, Incentives, and Capital Allocation

Hansoh's annual results announcement contains a paragraph that most readers skip and every governance analyst should read twice.

Under the Hong Kong Corporate Governance Code, provision C.2.1 requires that the roles of chairman and chief executive be held by different people. Hansoh does not comply. Zhong Huijuan is both chairlady and CEO, and the board's stated justification is that "due to the nature and the extent of the Group's operations and Ms. Zhong's in-depth knowledge and experience in the PRC pharmaceutical industry, the balance of power and authority under the present arrangement is not impaired."⁶ The board says it will consider splitting the roles "at a time when it is appropriate."

That is the only code provision Hansoh breached in 2025, and the disclosure is candid. But it sits on top of an ownership structure that already concentrates authority: the family holds roughly two-thirds of the shares, principally through a trust vehicle that stood at 65.7% of the register in mid-2024.⁹⁸ The board comprises Zhong as chairlady and executive director, her daughter Sun Yuan and Dr. Lyu Aifeng as executive directors, one non-executive director and independent non-executive directors.⁶ Sun Yuan's remit, per Forbes, covers research and development.⁸

The alignment argument is genuine and should not be dismissed. A controlling family with two-thirds of the equity and a multi-decade horizon can fund R&D through a policy shock that would have gotten a professional CEO fired. Hansoh's willingness to let its largest business reprice by four-fifths while raising R&D spending is exactly the kind of decision that quarterly-reporting, diffusely-held companies struggle to make. Zhong has run this company for thirty-one years. That continuity is an asset.

The cost is that there is no external check. And 2025 produced a capital allocation decision that a skeptical investor is entitled to challenge.

On August 20, 2025, Hansoh agreed to place 108,000,000 new shares at HK$36.30 each through Citigroup and Morgan Stanley, against a closing price of HK$38.82 the prior day — a discount of roughly 6.5%. Net proceeds were approximately HK$3,896.54 million, earmarked 65% for R&D and in-licensing, 25% for facilities including the Shanghai global R&D headquarters, and 10% for working capital.⁶²⁵

Now set that against the balance sheet. At December 31, 2025, Hansoh held cash and bank balances of RMB31,549 million, up from RMB22,622 million a year earlier. Operating activities generated a net cash inflow of RMB6,738 million during the year. Capital expenditure was RMB458 million. Net current assets stood at RMB31,160 million and the gearing ratio — total liabilities to total assets — was 11.4%.⁶

The company raised roughly HK$3.9 billion while holding cash equivalent to more than eight times that amount, generating operating cash flow of nearly twice the raise annually, and spending capex of about one-tenth of the raise. And by December 31, 2025 — four months after the placement — only approximately HK$436 million of the net proceeds had been deployed, leaving about HK$3,461 million unutilized.⁶

There is a defensible reading: pharmaceutical R&D is lumpy, in-licensing opportunities appear without warning, a war chest has option value, and broadening the shareholder base — which the directors cited explicitly — has real merit for index inclusion and liquidity. There is also a less flattering reading, and it is the one an activist would press: the company issued equity it demonstrably did not need, at a moment when its shares had run hard, diluting existing holders to add cash to a pile it was already unable to spend. Selling stock you believe is cheap is value-destructive. Selling stock you believe is expensive is rational — and if that was the judgment, it is a judgment about valuation that management has not articulated publicly.

The dividend policy sharpens the tension rather than resolving it. Hansoh declared an interim dividend of HK$23.16 cents per share for the first half of 2025 and recommended a final dividend of HK$20.00 cents, taking the full-year distribution to HK$43.16 cents.²³⁶ Paying out cash with one hand while issuing shares with the other is internally coherent only if the equity raise funds returns above the cost of that capital. It may. But the burden of proof sits with management, and the disclosure does not carry it.

On the credibility ledger more broadly, the record is better than the capital allocation episode suggests. The narrative across filings has been consistent for years: innovation as the core driver, R&D spend rising annually, out-licensing as a deliberate strategy rather than an opportunistic scramble. The company said it would move the revenue mix toward innovative medicines and it did — 68% in 2023, 77.3% in 2024, 82.2% in 2025 — with each step disclosed against the prior year. It said the ADC assets had global potential and four multinationals validated that with cash. It has not made aggressive forward guidance that later required explanation, largely because, in the Hong Kong convention, it does not give much numerical guidance at all.

That last point is itself a limitation. Hansoh's investor communication is thinner than a US-listed peer's: no quarterly reporting, no formal guidance, limited transcript availability for analyst Q&A. Where an investor would most want to hear management pressed — on how much of forward revenue growth depends on new signings, on whether the deal cadence is repeatable, on the pricing gap versus Western comparables — the public record is largely silent. Disclosure of senior operating management's tenure and prior track record beyond titles is similarly limited. Some of this is convention; some of it is a choice, and it is a choice a controlling family can make without consequence.

One data point that cuts the other way: as of the 2025 results announcement, Hansoh's MSCI ESG rating had been upgraded to AAA and the company was included in S&P Global's Sustainability Yearbook, ranking in the top 1% of the Chinese pharmaceutical industry.⁶ Governance ratings are imperfect instruments, but a controlled company achieving a top-band external assessment is not nothing.

The related-party dimension deserves one more paragraph, because it is genuinely unusual and rarely discussed. Zhong Huijuan controls Hansoh; her husband, Sun Piaoyang, is the founding figure behind Hengrui. These are the two largest domestic innovative-drug companies in China, competing directly in oncology, both bidding for the same targets, the same scientists, and the same Big Pharma partners. In December 2025 they transacted with each other: Hansoh in-licensed the calcium-sensing receptor modulator HS-10568 from Hengrui for the PRC.⁶ Nothing about the disclosure suggests anything improper, and each company has its own listed shareholder base and independent directors. But a skeptical investor is entitled to ask how competitive dynamics, pipeline overlap, talent movement and asset allocation are actually managed between two firms controlled by one household — and to note that the answer is not something minority shareholders of either company can independently verify.

The final management-credibility observation is about what has not happened. Across the transition, Hansoh has not restated results, has not changed auditors under contentious circumstances, has not issued and then walked back ambitious targets, has reported no material contingent liabilities and has pledged no assets.⁶ For a Chinese company listed offshore — a category where investors have learned to price accounting risk explicitly — a clean, unremarkable filing history over seven public years is itself a meaningful data point.

↑ Back to Top

VIII. What's Still There: The Generics and Legacy CNS Base

It would be easy to write the legacy business out of the story entirely. That would be a mistake, though not for the reason companies usually claim.

After VBP and five years of innovative-drug growth, the non-innovative portion of Hansoh's revenue is now approximately 17.8% of the total — roughly RMB2.7 billion in 2025, down from about 22.7% in 2024 as the innovative side grew faster.⁶ The CNS therapeutic area, which includes both the innovative XINYUE and the legacy psychiatric portfolio of olanzapine, paliperidone and agomelatine, contributed RMB1,310 million.⁶

The decline is arithmetic, not collapse. Most of Hansoh's significant generic molecules have already been through VBP, which means the worst of the repricing is behind them; the sell-side view has been that the remaining generic business should be broadly stable going forward, having already absorbed the shock.⁹ What is left is a low-growth, low-margin, cash-converting utility.

Its value to the equity story is not its P&L contribution. It is three things the innovative business would otherwise have to buy. First, the manufacturing base — GMP facilities, quality systems and regulatory relationships built over three decades, which is why Hansoh's capex requirement is a modest RMB458 million a year rather than the hundreds of millions a scaling biotech would need.⁶ Second, the hospital sales organization, which is the reason a newly approved Hansoh indication reaches formulary quickly. Third, a floor: a business that does not need external financing to survive a bad clinical year.

Compare that to a pure-play biotech. Innovent reached its first full year of profitability only in 2025, sixteen years after founding.¹⁴ Hansoh has been profitable throughout its transition, which is why it could fund a rising R&D budget out of operating cash flow rather than serial equity raises. That is the generics inheritance doing quiet work.

There is a subtler benefit as well, and it shows up in how Hansoh negotiates. A company that must sell an asset to make payroll sells on the buyer's timetable. A company with a legacy business covering its fixed costs can walk away from a term sheet. The rising upfronts across Hansoh's deal sequence are consistent with a seller that gained leverage over time — though, as noted, the initial terms suggest that leverage was not fully exercised early.

The honest framing, then, is that the legacy base is neither a hidden asset nor a drag. It is a diminishing but useful ballast — and its remaining VBP exposure is a modest, well-understood risk rather than a live threat. The growth and the valuation now sit almost entirely elsewhere. The one scenario in which it matters again is a bad one: if the innovative pipeline stumbled badly and the licensing engine stalled, this is the business that would still be generating cash while management rebuilt. Ballast is invisible until the weather turns.

↑ Back to Top

IX. Playbook: Business & Investing Lessons

Four transferable ideas come out of this story, and the interesting thing about all four is that they generalize well beyond Chinese pharmaceuticals.

Policy can be the best strategic consultant you never hired — if you were already halfway there. VBP forced a transformation that Hansoh's board might have deferred for a decade under commercial logic alone. But the reason the forcing function produced a transformation rather than a liquidation is that the R&D infrastructure already existed. The lesson is not "regulation drives innovation." It is that optionality has to be bought before you need it, because when the shock arrives there is no time to build. Companies facing slow-moving structural threats — energy transition, AI disruption, tariff regimes — should read Hansoh's R&D line from 2010 to 2018 as the price of having a second act available.

Asset-light globalization is a real strategy, not a euphemism for weakness. Building a US commercial organization for an oncology drug costs hundreds of millions of dollars and years, and it is the single largest destroyer of capital in the biotech industry. Licensing lets a company convert scientific output into cash without ever leaving home. The trade-off is explicit and quantifiable: you exchange the majority of the terminal economics for the elimination of commercial execution risk and capital intensity. For a company with a deep pipeline and no Western infrastructure, that is often the right trade. For a company with one great asset, it is usually the wrong one — which is why BeOne's opposite choice was also defensible.

Learn to separate the annuity from the option. The single most useful analytical habit an investor can bring to Hansoh — and to any company with material licensing, settlement, contract or one-time income — is to decompose the income statement into revenue that recurs because customers keep buying and revenue that arrived because a negotiation concluded. Both are real. Only one compounds. Hansoh discloses the split; the market frequently reads the total.

Founder control is a volatility trade, not a quality signal. Concentrated ownership removes the short-termism problem and removes the accountability mechanism at the same time. It produces both the willingness to spend through a crisis and the ability to issue equity into a rally without explaining why. Investors buying controlled companies are, in effect, buying the founder's judgment unhedged. That is sometimes an excellent purchase. It is never a diversified one.

Being early to a repricing is worth less than being credible during one. Hansoh signed the first of its major ADC deals in 2023, before Chinese assets were fashionable, and took terms that look thin against what the same category commands in 2026. First movers in a repricing market frequently capture the least value from it, because the price they accept becomes the comparable everyone else improves upon. The generalizable point for any company selling into an inefficient market — intellectual property, spectrum, carbon credits, private assets — is that the returns accrue to whoever transacts once the market has learned to price, not to whoever transacts first. Hansoh's compensation for going early was credibility and repeat business. Whether that was worth the discount is an open question, and reasonable people can disagree.

↑ Back to Top

X. Porter's Five Forces & 7 Powers Lens

Run Hansoh through the standard frameworks and the picture that emerges is a company with real, narrow advantages sitting in a structurally difficult industry.

Buyer power is the defining force, and it is extreme. In China, the buyer is effectively a single entity: the National Healthcare Security Administration, which runs both VBP for off-patent drugs and the National Reimbursement Drug List negotiations for innovative ones. NRDL inclusion is the gateway to volume — Hansoh notes that the third and fourth indications of Ameile were added to the 2025 list, and that Saint Luolai and Hengmu were renewed.⁶ The price of admission is a negotiated discount, typically steep, renegotiated periodically. This is categorically unlike Western payer dynamics, where fragmented insurers and pharmacy benefit managers negotiate against a manufacturer with genuine walk-away power. In China the manufacturer's walk-away option is to forgo the market. That is not pricing power; it is volume-for-price arbitrage administered by the state.

Supplier power is low and not a material part of the story — APIs, CRO capacity and manufacturing inputs are abundantly available in China.

Rivalry is intense and getting worse. Hengrui, Innovent, BeOne, Junshi, Akeso and dozens of smaller players are developing overlapping ADC, bispecific and metabolic assets against many of the same targets. Crucially, they compete not only for Chinese patients but for the same finite pool of Big Pharma business development attention. When 186 cross-border out-licensing deals close in a single year, the marginal Chinese ADC is no longer scarce.²⁴ Differentiation now depends on data quality and speed to a randomized readout — which is exactly why the ARTEMIS-008 result matters so much, and why a company with three plausible ADCs and no Phase 3 data is in a much weaker negotiating position than it was in 2023.

Threat of substitution operates at the modality level. ADCs are currently displacing conventional chemotherapy in solid tumors; radioligand therapies, T-cell engagers and next-generation immuno-oncology combinations may in time displace ADCs. In metabolic disease, the oral GLP-1 field Hansoh entered via Merck is among the most crowded in pharmaceutical history.

New entrants are not a meaningful threat in Chinese commercialization — the regulatory and sales barriers are high — but they are a severe threat in the licensing market, where the barrier to becoming a seller of a preclinical asset is a competent chemistry team and a plausible data package.

On Hamilton Helmer's 7 Powers, three apply with varying strength:

Process power is Hansoh's most credible claim. The capability to run discovery, chemistry, manufacturing, clinical development and Chinese commercialization inside one organization was built over roughly fifteen years and cannot be bought. The evidence is repeat business from sophisticated counterparties: GSK licensed twice within sixty days, and four separate multinationals conducted independent diligence and concluded the assets were worth eight- and nine-figure upfronts.

Scale economics apply in a specific, limited way — 2,300 research staff across four sites and 9,347 total employees give Hansoh the throughput to run more than 70 trials simultaneously, which a smaller company cannot.⁶ But Hengrui runs a bigger R&D operation, so this is a scale advantage over the mid-tier, not over the leader.

Counter-positioning is the subtlest and most interesting. Hansoh competes against Western-listed biotechs that cannot match its cost per experiment, and against Chinese pure-play biotechs that cannot match its cash flow cushion. A venture-backed competitor with two years of runway must sell its best asset when the market offers, not when the data justify the price. Hansoh, generating RMB6.7 billion of operating cash flow annually, can wait.⁶ Whether it consistently does wait is precisely the question the GSK deal terms raise.

The powers Hansoh conspicuously lacks: no network economies, no switching costs beyond physician habit, no branding power in the pharmaceutical sense, and — critically — no cornered resource. Its patents protect individual molecules for finite terms, which is a legal monopoly on one product, not a durable structural advantage over an industry.

Two of Hansoh's most-cited advantages deserve explicit demotion, because they are commonly mistaken for moats.

The first is the R&D cost advantage. Running a clinical trial in China is genuinely cheaper than running one in the United States or Western Europe — lower site costs, faster enrollment in some indications, lower salaries for research staff. But this is a national advantage, shared by every Chinese competitor and shrinking as domestic biotech salaries rise and trial demand outstrips high-quality site capacity. It explains why Chinese assets exist. It explains nothing about why a partner should choose Hansoh's over Hengrui's.

The second is the patent portfolio. Hansoh filed 40 formal patent applications in China and 128 overseas during 2025 and was granted 80 globally.⁶ Patent counts are a proxy for research activity, not for defensibility. A pharmaceutical patent protects one composition of matter; it does not stop a competitor from finding a structurally distinct molecule that hits the same target, which is precisely what happens in every crowded target class — as the population of B7-H3 and GLP-1 programmes worldwide demonstrates.

What is genuinely defensible reduces to one thing: an integrated organization that can take a target from idea to Phase 3 data faster and cheaper than most, repeatedly, funded from its own cash flow. That is real. It is also the sort of advantage that erodes quietly if execution slips, because there is no structural mechanism keeping it in place.

↑ Back to Top

XI. Bull vs. Bear: The Investment Case

Before the two cases, it is worth clearing away three claims that circulate widely about this company and do not survive contact with the filings.

Myth: Hansoh is a Chinese biotech. Reality: it is a profitable, cash-generative specialty pharmaceutical company with a large research operation. It has never had a loss-making year as a listed entity, funds its R&D from operating cash flow, and carries no meaningful debt.⁶ Investors who bucket it with the pre-revenue Hong Kong biotech cohort are mispricing both the risk and the character of the business.

Myth: the out-licensing deals prove Hansoh's science is world-class. Reality: they prove Hansoh's science is good enough to buy at a discount. Four sophisticated buyers conducted diligence and paid real money, which is genuine evidence. They also paid substantially less than they have paid for comparable Western-origin assets, which is genuine counter-evidence.²²⁰⁵ Both facts are true at once, and any narrative that keeps only the flattering half is incomplete.

Myth: the pivot away from generics was a management masterstroke. Reality: it was a policy-forced repricing that Hansoh happened to be prepared for, because R&D spending had been ramping for years beforehand for reasons that had little to do with anticipating volume-based procurement.⁹ Preparedness is a real virtue. Prescience is a different claim, and the timeline does not support it.

With that cleared, the two cases.

The bull case starts with a transition that is no longer a promise. Innovative medicines and collaborative products moved from 68% of revenue in 2023 to 77.3% in 2024 to 82.2% in 2025, and that shift was accompanied by accelerating profitability rather than the margin destruction that usually accompanies a pivot.⁹⁶ Seven innovative medicines generate product sales, spread across oncology, CNS, metabolic and anti-infective franchises, so the business is not a single-molecule story.

The clinical validation arrived, and recently. On July 10, 2026 — one month before this writing — Hansoh announced that ARTEMIS-008, a Phase 3 study of risvutatug rezetecan versus topotecan in relapsed small-cell lung cancer, met its primary endpoint of overall survival with statistically significant and clinically meaningful improvement, with consistent benefit on progression-free survival. GSK described it as the first positive Phase 3 overall survival data for a B7-H3-targeted ADC in any tumor type.²⁶ That transforms HS-20093 from a plausible asset into a de-risked one, and it does so for the partner as well: GSK's global Phase 3 EMBOLD SCLC-301 trial has pivotal data expected in 2027.²⁶ The molecule already carries three NMPA breakthrough designations, two FDA breakthrough designations, FDA orphan drug designation, and EMA PRIME and orphan designations.⁶

The regulatory internationalization is real. Aumolertinib is approved in the UK and the EU, which means a Chinese company has now cleared the two most demanding non-US regulatory bars with a self-discovered small molecule. The metabolic pipeline gives exposure — via Merck and Regeneron, two credible partners — to obesity and diabetes, arguably the largest addressable market in pharmaceuticals. The balance sheet is close to unassailable: RMB31.5 billion of cash, 11.4% gearing, no material contingent liabilities, no pledged assets.⁶ And the controlling family's economic interest is aligned with the share price to an unusual degree.

The bear case begins with the same income statement, read differently.

Roughly 14% of 2025 revenue was collaboration income recognized at a point in time, and it grew 34.5% year over year — faster than product sales.⁶ The margin structure and the growth rate both depend on a deal cadence that has no contractual guarantee. If 2026 or 2027 passes without a major signing, reported growth decelerates sharply even if every prescription-driven line grows exactly as planned. This is the single most important thing to understand about the quality of Hansoh's earnings, and it is not adequately reflected in a consolidated growth rate.

The pricing evidence on those deals cuts against the "world-class R&D" narrative. Hansoh's B7-H3 ADC cleared at $185 million upfront while a comparable-stage B7-H3 ADC from Daiichi Sankyo sat inside a package that commanded billions at signing.²²⁰ Either Hansoh's assets are genuinely worth an order of magnitude less — which contradicts the bull case — or Hansoh sold cheap, which speaks to negotiating position rather than science. Neither reading is flattering, and the sector data on Chinese upfront discounts suggests the answer is mostly the second.⁵

Governance is a live issue rather than a theoretical one. A combined chair-CEO, a board including the founder's daughter, two-thirds family ownership, a related-party in-licensing transaction with a company run by the founder's husband, and an equity placement conducted while sitting on cash that was subsequently 89% unspent four months later — each is defensible individually, and collectively they describe a company where minority shareholders have influence but no leverage.⁶

Sector re-rating risk is the largest macro exposure. Hansoh's shares rose roughly 70% in the period preceding November 2025, and the fortune of its founder rose 67% alongside them.⁸ Much of that move was a repricing of the entire China-biotech complex rather than Hansoh-specific news. The stock has drifted lower in 2026 — market capitalization of HK$209.4 billion on August 10, 2026 against HK$218.4 billion at the end of 2025.⁷ If the licensing wave cools, or if a marquee partnered asset disappoints in a global Phase 3, the multiple compression would be felt across the sector and by Hansoh acutely.

Geopolitics is the tail risk that cannot be modeled. Hansoh's out-licensing strategy depends on US and European pharmaceutical companies being willing and legally able to license Chinese assets. Legislative proposals in the United States aimed at restricting biotech cooperation with Chinese entities have surfaced repeatedly; none has yet materially impeded deal flow, but the deal flow is the business model. A meaningful restriction would not reduce Hansoh's growth — it would remove a revenue line.

Finally, execution risk in the transition from seller of molecules to owner of global economics. Every licensing deal converts a potential product franchise into a royalty stream. That is a rational trade today, but it means Hansoh's long-run terminal value depends on partners executing, not on Hansoh executing. GSK, Merck, Roche and Regeneron are excellent operators — and they will prioritize their own portfolios.

That last risk has a specific, underappreciated form: shelving. A licensed asset can be strategically deprioritized inside a large pharmaceutical company for reasons that have nothing to do with its merits — a portfolio reshuffle, a competing internal programme, a change of therapeutic-area leadership. When that happens, the milestone schedule simply stops advancing. Hansoh retains contractual protections in these agreements, as any competent licensor would, but the practical remedy for a shelved asset is slow and rarely restores the original timeline. Six of Hansoh's most valuable programmes now sit inside other companies' portfolios, subject to other companies' priorities.

The activist question. If a well-capitalized skeptic took a position and wrote a letter tomorrow, what would it say? Probably four things: that the August 2025 placement was unnecessary and dilutive given RMB31.5 billion of cash and RMB6.7 billion of annual operating cash flow; that combining the chairlady and CEO roles in a company with two-thirds family ownership is indefensible on any reading of the governance code; that disclosure around the split between recurring and non-recurring revenue, and around milestone realization rates, is thinner than the business's complexity warrants; and that the early ADC deal terms represent hundreds of millions of dollars of value transferred to GSK that a better-advised seller would have retained.

The company's answers to the first three are available and reasonable. The answer to the fourth is that 2023 was 2023, and hindsight is not a negotiating strategy. None of these are existential complaints. They are the ordinary friction of a controlled company that has grown large enough to attract scrutiny it is not structurally obliged to answer.

So: why does Hansoh win from here, and what breaks the case?

The credible answer to "why win" has three legs, and only one is fully load-bearing. The strongest is the discovery engine's demonstrated ability to produce assets that independent, sophisticated buyers pay for — validated four separate times by four separate diligence processes, and now backed by randomized Phase 3 survival data. The second is the cash-flow cushion, which lets Hansoh fund R&D internally and negotiate without a runway clock. The third — commercial infrastructure in China — is real but shrinking in relative importance as the value migrates offshore.

What breaks it: a decisive clinical failure in the GSK-partnered ADCs or the Regeneron metabolic asset, which would simultaneously kill milestone economics and puncture the platform narrative; a structural narrowing of the China cost advantage as domestic R&D salaries rise and competition for the same targets intensifies; regulatory or political interruption of the cross-border licensing channel; or, most quietly, simple deal-cadence exhaustion — a year or two in which nothing gets signed and the market discovers how much of the growth rate was BD.

↑ Back to Top

XII. Current Watch List: KPIs, Risks, What's Next

If you track only a few things about this company, track these.

KPI one: the split between product revenue and collaboration revenue. Hansoh disclosed pharmaceutical product sales of RMB12,913 million and collaboration revenue of RMB2,116 million for 2025, against RMB10,688 million and RMB1,573 million respectively for 2024.⁶ This single disclosure line determines whether reported growth is durable or episodic. Watch the product line's own growth rate independently of the total; that is the annuity. The collaboration line tells you how the option book is performing.

KPI two: milestone realization versus headline deal value. The published figures — up to $1.485 billion from GSK, up to $2.01 billion from Regeneron, up to $2.18 billion from Avere — are ceilings, achieved only if every development, regulatory and sales trigger fires.¹⁶¹⁸ Industry-wide, most never do. The number that matters is cash actually received against each agreement over time, disclosed in the collaboration revenue and receivables detail. A rising ratio of realized to announced value would be the strongest possible evidence that the deal engine is a business rather than a headline generator.

KPI three: late-stage clinical readouts on the partnered pipeline. ARTEMIS-008 was the proof that Hansoh's ADC chemistry produces survival benefit in a randomized setting.²⁶ The next tests are GSK's global EMBOLD SCLC-301 trial, with pivotal data expected in 2027; the Phase 3 program for the B7-H4 ADC in platinum-resistant ovarian cancer; and the Phase 3 obesity and type 2 diabetes studies of HS-20094, the asset Regeneron licensed.⁶²⁶ These are binary events with asymmetric consequences: a win advances milestone payments and validates the platform; a failure removes both the payments and the narrative.

On the risk radar, four items are material enough to price.

Policy risk has not gone away, it has moved. The remaining generic exposure is small and largely repriced, but innovative medicines face NRDL negotiation, which extracts price in exchange for volume on a recurring basis. Hansoh's own management framed 2025 as the year Chinese innovative pharma shifted from an R&D-investment-driven model to one driven by "clinical value prioritization and commercialization efficiency," with intensifying competition and "higher requirements" on differentiated positioning and cost control.⁶ That is management language for: the easy phase is over.

Clinical risk is now embedded in the equity in a way it was not five years ago. When 82.2% of revenue comes from innovative medicines and a meaningful slice of the growth comes from milestone-bearing partnerships, trial outcomes move the valuation directly.

Competitive crowding in ADCs and bispecifics is the most likely mechanism by which future deal terms deteriorate. Rising average deal sizes across the sector are encouraging on price, but the supply of Chinese assets is growing faster than the number of buyers.²⁴

Geopolitical and regulatory risk to US-bound licensing remains the fat tail.

One risk that is frequently listed and does not belong here: financing and refinancing. With gearing at 11.4%, no pledged assets, no material contingent liabilities, cash of RMB31.5 billion and operating cash flow approaching RMB7 billion, Hansoh has essentially no cost-of-capital exposure.⁶ Whatever breaks this company, it will not be a balance sheet.

Currency is a mild, real exposure rather than a headline one. Licensing receipts are dollar-denominated while costs and most product revenue are in renminbi; the 2025 accounts recorded net foreign exchange losses of approximately RMB44 million within other expenses, small in the context of the business but a reminder that the offshore revenue stream carries translation risk as it grows.⁶

Near-term, three things are worth watching specifically. Hansoh's board was scheduled to meet on August 26, 2026 to approve interim results for the six months ended June 30, 2026 and consider an interim dividend — the first reported period fully after the ARTEMIS-008 readout, and the first opportunity to see whether product revenue growth is holding up independent of new signings.²⁷ The Avere/NextCure structure will begin to show whether taking equity rather than cash produces better returns than the pure-licensing model. And the commercial ramp of aumolertinib in Europe, following the February 2026 EU approval and the December 2025 Glenmark agreement, will be the first real read on whether a China-developed oncology drug can build a Western prescription base.⁶

↑ Back to Top

XIII. Epilogue

Thirty-one years after a chemistry teacher left a classroom in Lianyungang, the company she built occupies a genuinely unusual position — and the most interesting thing about it is how unresolved that position is.

Hansoh is no longer a generics manufacturer with an R&D budget. That question was settled by policy and by the numbers: the legacy business is under a fifth of revenue and the innovative business funds itself. But it is not yet a global innovative pharmaceutical franchise either. It sells its best assets to companies that have what it lacks — Western commercial organizations, global regulatory machinery, the ability to run a Phase 3 program across forty countries. Its molecules reach American and European patients under someone else's logo.

The licensing wave bought Hansoh time and capital: roughly RMB2.1 billion of collaboration revenue in 2025 alone, a cash pile above RMB31 billion, and an R&D budget that has grown every year for over a decade.⁶ What it has not yet bought is proof that the underlying science produces durable global economics rather than well-timed deal premiums. ARTEMIS-008 was the first hard evidence in favor. The next few years of readouts — GSK's global SCLC trial, the ovarian cancer program, the metabolic Phase 3s — will supply the rest of the answer, one way or the other.

It is worth naming what would constitute genuine proof, because it is a narrower set of outcomes than the current narrative implies. Not another licensing deal — those now demonstrate market conditions as much as company quality. Not another Chinese approval — Hansoh has proven it can execute domestically. Proof would be a Hansoh-originated molecule generating substantial recurring revenue in Western markets, whether through a partner's sales or its own, at Western prices, in a randomized-data-driven category. Aumolertinib in Europe is the first candidate. The GSK-partnered B7-H3 ADC, if EMBOLD reads out positively in 2027, would be the second and far larger one.

Until then, an investor is underwriting a company whose value rests substantially on the future decisions of GSK, Merck, Roche, Regeneron and a newly merged Nasdaq shell — none of which Hansoh controls, all of which it selected.

There is a version of the next decade in which Hansoh's royalty streams compound into a self-funding global franchise and the company graduates to building its own Western presence on the back of a genuinely validated platform. There is another in which the China discount persists, the deal terms compress as domestic competition thickens, and Hansoh settles into a comfortable, profitable, permanently mid-tier role: the reliable supplier of molecules to companies that capture the value.

Both are consistent with the evidence available today. That ambiguity is not a failure of analysis; it is the actual state of the investment case, and it is the state of Chinese innovative pharmaceuticals more broadly. An industry was told by its government to stop copying and start inventing. It did. Whether invention at Chinese cost can command Western prices — durably, on its own terms, without a discount — is the question the next five years will settle. Hansoh is one of the best places to watch it get answered.

References

1. GSK enters exclusive license agreement with Hansoh for HS-20089 — GSK, 2023-10-20 ↩↩↩

2. GSK enters exclusive license agreement with Hansoh for HS-20093 — GSK, 2023-12-20 ↩↩↩↩

3. Merck Enters into Exclusive Global License Agreement with Hansoh Pharma for Investigational Oral GLP-1 Receptor Agonist — Merck.com, 2024-12 ↩↩

4. China's Hansoh Pharmaceutical Group beats net profit expectations in 2025 — PM360 / Reuters, 2026-03-29 ↩

5. China now sources 32% of global biopharma out-licensing deals, up from 5% in 2020 — PharmaSource, 2026-05-12 ↩↩↩↩

6. Annual Results Announcement for the Year Ended December 31, 2025 — Hansoh Pharmaceutical Group / HKEXnews, 2026-03-29 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

7. Hansoh Pharmaceutical Group Company (HKG:3692) Market Cap & Net Worth — StockAnalysis, 2026-08-10 ↩↩

8. China's Richest Self-Made Woman Amasses $19.7 Billion Fortune Amid Biotech Boom — Forbes, 2025-11-05 ↩↩↩↩↩↩↩↩

9. Hansoh Pharma (3692 HK): Leading innovative biopharma company — CMB International Global Markets, 2024-08-26 ↩↩↩↩↩↩↩↩↩↩↩

10. Global Offering prospectus — Hansoh Pharmaceutical Group / HKEXnews, 2019-05-31 ↩

11. The impact of Chinese volume-based procurement on pharmaceutical market concentration — Frontiers in Pharmacology ↩

12. Aumolertinib Approved by UK's MHRA — Hansoh Pharmaceutical Group, 2025-06 ↩

13. Hengrui Pharma Announces Strong 2025 Annual Results — PR Newswire, 2026-03-25 ↩↩↩

14. Innovent Announces 2025 Annual Results and Business Updates — PR Newswire, 2026-03-26 ↩↩↩

15. BeOne Medicines Announces Fourth Quarter and Full Year 2025 Financial Results — Business Wire, 2026-02-26 ↩

16. Merck & Co. boosts GLP-1 portfolio, paying $112M upfront for Hansoh's preclinical drug — Fierce Biotech ↩

17. China's Hansoh and Leads Biolabs clinch US$2.5 billion in global drug licensing deals — South China Morning Post ↩

18. S&C Advises Hansoh Pharma on Exclusive License Agreement and Strategic Investment with Avere Therapeutics — Sullivan & Cromwell LLP, 2026-07-15 ↩↩↩

19. Biotheus Expanded Their Partnership with Hansoh Pharma for Developing EGFR/cMET Bispecific Antibody-Drug Conjugates — PR Newswire, 2024-03-14 ↩

20. Daiichi Sankyo and Merck Announce Global Development and Commercialization Collaboration for Three Daiichi Sankyo DXd ADCs — Merck.com, 2023-10-19 ↩↩↩

21. Takeda Announces Oncology Partnership with Innovent — Takeda, 2025-10-21 ↩

22. Bristol Myers Squibb and Hengrui Pharma Announce Strategic Agreements to Advance Innovative Medicines Across Oncology, Hematology, and Immunology — Bristol Myers Squibb, 2026-05-12 ↩

23. Interim Results Announcement for the Six Months Ended June 30, 2025 — Hansoh Pharmaceutical Group / HKEXnews, 2025-08-18 ↩↩

24. China biopharma out-licensing surges to record $137.7B in 2025; 2026 on pace to break it again — PharmaSource, 2026-02-20 ↩↩↩

25. Hansoh Pharma in HK$3.9 Billion Share Placement — Cleary Gottlieb, 2025-08 ↩

26. GSK's licensor Hansoh Pharma announces positive phase III results for Ris-Rez in China patient population — GSK, 2026-07-10 ↩↩↩↩

27. Hansoh Pharmaceutical Schedules Board Meeting to Approve Interim Results and Dividend — TipRanks ↩