USD 1.275 Billion Upfront for a Phase 1 Molecule: Gilead Buys PepStar's Oral IL-23R Macrocycle and Opens Peptides' Second Demand Curve

On 22 October Gilead agreed to pay USD 1.275 billion in cash upfront (about RMB 8.5 billion) for Shanghai-based PepStar Therapeutics' oral IL-23R inhibitor programme, whose lead asset HCXT-2001 is still in Phase 1. The seller was founded in April 2023 and moved a molecule to a billion-dollar upfront in just over three years. The valuation is only part of the story: if oral macrocyclic peptides work, they open a second volume curve for peptide CDMOs beyond GLP-1.

On 22 October Gilead Sciences announced a definitive agreement with Shanghai-based PepStar Therapeutics to acquire its oral interleukin-23 receptor (IL-23R) inhibitor programme, led by HCXT-2001. The consideration is USD 1.275 billion in cash upfront (about RMB 8.5 billion), payable at closing, with the deal expected to close in the fourth quarter of 2026 and to give Gilead worldwide rights. Gilead also said the transaction would reduce Q4 GAAP and non-GAAP earnings per share by roughly USD 0.79 to 0.83.

Put the three numbers together and the character of the deal is clear: a Phase 1 molecule, USD 1.275 billion all in cash upfront, the entire consideration paid at the front. This is not a milestone-inflated headline total; it is money leaving the balance sheet this year.

The molecule: a macrocycle that survives the gut

HCXT-2001 is an oral macrocyclic peptide IL-23R inhibitor that PepStar differentiated by extending half-life. It is in Phase 1 in healthy volunteers. Registered studies include two assessing food effects on absorption (entered China's trial registry in August and September 2026) and a third in healthy adults in New Zealand testing two tablet formulations. The multiple-ascending-dose stage will evaluate both once-daily and once-weekly dosing.

In Gilead's framing, HCXT-2001 can be used alone or combined with its own α4β7 pathway inhibitor emvistegrast as a fully oral regimen, with inflammatory bowel disease as the target and potential in psoriasis. That combination logic is central to BMO Capital Markets' positive read: emvistegrast recently impressed in Phase 2, and pairing it with an oral IL-23R peptide could produce an all-oral option in an IBD market that has long depended on injectable biologics.

The competitive landscape explains the pricing. AbbVie's Skyrizi generated close to USD 10 billion in the first half of the year and is the cash cow of the IL-23 pathway, but requires subcutaneous injection. Johnson & Johnson's icotrokinra (Icotyde) became the first approved oral IL-23R peptide in March 2026, once daily for plaque psoriasis. Merck's oral peptide in the same class carries the same restriction: no food for at least 30 minutes after dosing.

Food effect is the shared weakness. A macrocycle must stay stable in the gastrointestinal tract and cross the epithelium, leaving narrow chemical room to manoeuvre, and absorption is highly sensitive to food. Whoever can move from once-daily to once-weekly, or build a formulation that does not require fasting, holds the differentiation lever. HCXT-2001's half-life engineering aims exactly there.

The seller: zero to a billion-dollar upfront in three years

PepStar Therapeutics was founded in April 2023 in the Shanghai free trade zone by Haibing Guo. Early backers include Lilly Asia Ventures, Sequoia China's seed fund and Eight Roads, with an angel-plus round joined by Taifu Capital in 2024; LongRiver Investments participated across three rounds from seed. The company's stated core asset is its proprietary oral peptide platform.

Three years and change from founding to a ten-figure-dollar upfront for its lead asset. That speed is itself a footnote on Chinese innovation in 2026: pricing power no longer depends on clinical stage but on whether the asset solves something others could not. A Phase 1 molecule commanding USD 1.275 billion in cash shows that diligence and trust in Chinese early pipelines have moved ahead of the clinic.

But a high upfront is not certainty. Three variables will decide whether HCXT-2001 delivers: whether the therapeutic window is wide enough (a narrow window turns Phase 3 into dose-finding), whether there is head-to-head room against existing IL-23 biologics and oral competitors, and how much adherence is lost to food-effect constraints. None of these is answered by Phase 1 data.

For CDMOs: a second volume curve is forming

The relevance to CXOs is larger than to the biotech crowd, and the reason is dosage form.

Injectable peptide volumes have already been validated once by GLP-1. Semaglutide and tirzepatide pushed solid-phase peptide synthesis capacity from tens of thousands of litres into the hundred-thousand-litre range: WuXi AppTec is moving SPPS from 100,000 to 130,000 litres, Asymchem's peptide capacity is full in 2027 with new build planned, and Jiuzhou Pharma expanded solid-phase peptide capacity by 300% year on year in the first half yet remains capacity-constrained. That is the first curve.

Oral peptides are the second. Oral bioavailability is typically far lower than subcutaneous injection, so achieving equivalent exposure requires materially more API per patient; combined with once-daily, long-term use in chronic disease, the pull on API tonnage and drug product capacity is multiplicative. If oral IL-23R succeeds in Phase 3, peptides expand from a weight-loss and diabetes narrative into a second front in autoimmune chronic disease.

The process requirements are also harder. A macrocycle is not a linear peptide: cyclisation after synthesis raises purification difficulty and complicates the impurity profile; improving oral absorption often requires permeability engineering such as N-methylation, making non-natural amino acids and modified monomers a supply bottleneck; and the drug product side must run bioavailability and food-effect studies across two tablet strengths, pulling formulation work forward. Orders therefore land not only on SPPS reactors but on cyclisation and purification capability, non-natural amino acid supply, and oral solid dosage development.

Few Chinese CDMOs hold all three. Many can make peptide API; far fewer can deliver an integrated macrocycle-plus-oral-product package. That is precisely the commercial logic behind Asymchem's drug product expansion (cartridges, prefilled syringes, blow-fill-seal) and WuXi AppTec's continued TIDES investment — beyond capacity, you need the engineering ability to turn a molecule into something a patient can swallow.

The other half of the same week: capital changing channels

Put the deal back into 22 October and another thread appears: the same day, the US Treasury published proposed outbound investment rules adding biotechnology to controlled sectors while explicitly exempting cash-paid, non-equity technology in-licensing and fee-for-service R&D collaborations.

Gilead's USD 1.275 billion cash upfront sits exactly on that exemption line. That is not coincidence. As the equity channel narrows, the asset channel gets more crowded and upfront ratios get pushed higher. For Chinese biotechs this means two things: assets that can command a good price still will; but the old path of forming a NewCo, raising, then listing becomes more expensive, and deal structures will converge faster on direct sales to MNCs.

For CDMOs this is a mildly positive structural shift. When an asset moves to an MNC it enters global multi-centre trials and global registration, and manufacturing orders become more certain — an MNC's Phase 3 and commercial batches are more predictable than those of a biotech living round to round.

Editor's view: the upfront is paying for a platform option

Strictly speaking Gilead did not buy a molecule; it bought the first ticket on a platform. PepStar's founder framed it as validation of the company's proprietary oral peptide platform — an MNC paying all cash at Phase 1 is paying for a platform that can keep producing molecules, not for one molecule's data.

That also explains why 2026's large upfronts cluster around platform assets: CSPC and AstraZeneca's long-acting peptide platform (up to USD 18.5 billion in potential value), Hengrui and Bristol Myers Squibb's multi-therapeutic early portfolio (USD 15.2 billion), Hansoh and Roche's ADC platform (over USD 15.5 billion). The object of evaluation has shifted from a molecule to a production line that can repeat.

The lesson for CDMOs: platform assets generate batch demand, not single-batch demand. A platform means multiple molecules entering the clinic in sequence, creating continuous orders for the same process — macrocycle synthesis, cyclisation, oral formulation. Whoever proves the process on a client's first molecule inherits the default-supplier position for the rest of the line. That is worth more than winning one commercial batch.