IBI343: a gastric cancer ADC already “validated globally”, whose real test is its second tumor type

To understand IBI343’s strategic value, first understand where the company holding it stands. In 2025, Innovent Biologics’ total revenue reached about RMB 13 billion, up 38.4% year on year, of which product revenue was about RMB 11.9 billion, up 44.6%; the company achieved IFRS net profit of about RMB 814 million, formally crossing the break-even line, with year-end cash reserves of about RMB 24.3 billion. From its first product launch in 2019 to 18 products on the market and revenue exceeding RMB 10 billion took Innovent just 7 years.

Behind this string of numbers is a proven China oncology commercial team. With the PD-1 inhibitor Tyvyt (sintilimab) at its core, plus biosimilars such as bevacizumab and rituximab and a batch of in-licensed small-molecule targeted drugs, more than 3,000 new oncology patients start Innovent treatments every day. For a gastric cancer ADC, this means it does not need to build a GI oncology hospital network and medical promotion system from scratch at launch — that system is already running.

In the CLDN18.2 ADC race, most competitors’ payloads are MMAE (a microtubule inhibitor). Innovent took another road — an exatecan-class topoisomerase inhibitor. The difference this choice makes is not in how dazzling the absolute response rate is, but in the toxicity profile.

In the Phase 1 study published in Nature Medicine (NCT05458219; CLDN18.2 high expression defined as membrane staining intensity ≥2+ in ≥75% of tumor cells), the 6 mg/kg cohort (n=31) had a confirmed objective response rate of 29.0% (32.3% after data update), disease control rate of 90.3%, median progression-free survival of 5.5 months and median overall survival of 10.8 months; the 8 mg/kg cohort had a higher confirmed response rate (47.1%), but 6 mg/kg Q3W was chosen as the recommended Phase 2 dose. What really deserves to be circled is the safety in 116 gastric/GEJ patients: grade ≥3 nausea was only 1.7%, and no interstitial lung disease was observed. The main grade ≥3 adverse events were myelosuppression (neutropenia 28.4%, leukopenia 25.9%, anemia 16.4%), within the manageable range.

The Phase 3 G-HOPE-001 (NCT06238843) is an international multicenter, randomized, open-label study in China and Japan, enrolling patients with CLDN18.2-positive advanced gastric/GEJ adenocarcinoma who had received at least two prior lines of systemic therapy, comparing IBI343 monotherapy with “treatment of investigator’s choice”, with dual primary endpoints of progression-free survival and overall survival. The June 4, 2026 announcement made clear that the first interim analysis met the pre-specified progression-free survival (PFS) primary endpoint, officially described as “excellent efficacy, good overall safety and a low incidence of GI-related toxicity”; but overall survival, hazard ratios and full survival data have not yet been published — exactly what needs close watching when the complete data read out.

The CLDN18.2 race: a crowded timetable

Lining up the main players makes IBI343’s real position clearer — it is not sprinting across empty ground, but fighting for tempo in a lane packed with monoclonal antibodies, ADCs and cell therapies.

Note: The table is a compilation of public progress as of June 2026; lines and status are subject to each company’s latest disclosures.

This table reveals a repeatedly validated Chinese pattern: once an originator target is validated, domestic pipelines pour in extremely fast, compressing the window for differentiation. IBI343’s response is “trading position for space” — since it is hard to pull away from rivals in first and second line, first secure the label of “first ADC to market” in third line and beyond, then use it as leverage into earlier lines and new tumor types. It is a clear-eyed playbook belonging to a mature pharma company.

IBI343 has collected almost every expedited tool available in China. Both the gastric and pancreatic cancer indications received NMPA Breakthrough Therapy Designation; the gastric cancer new drug application was accepted by NMPA with priority review on June 4, 2026; and in pancreatic cancer it has also obtained FDA Fast Track designation. Choosing “treatment of investigator’s choice” as the comparator in third-line-plus gastric cancer both matches the clinical reality of no uniform standard of care in this line and avoids ethical controversy — a well-thought-out registration design.

Even more noteworthy is the “native” nature of the data. From the Phase 1 first-in-human study to the Phase 3 registration study, IBI343’s core data were all generated in China (and Japan), with no reliance on overseas bridging, thus bypassing common sources of delay such as ethnic differences and supplementary PK/PD studies. Add that zolbetuximab has already paved China’s regulatory precedent for the CLDN18.2 target, and IBI343’s review pathway is relatively clear. At the current pace, the gastric cancer indication is expected to land in 2026–2027.

In October 2025, Innovent and Takeda entered a global strategic collaboration. The deal was a “bundle”: it included the PD-1/IL-2α bispecific fusion protein IBI363, the CLDN18.2 ADC IBI343 and an option on an early-stage bispecific ADC, IBI3001. Innovent received a US$1.2 billion upfront payment (including a US$100 million equity investment component), plus milestones and royalties.

Specific to IBI343, the rights split is very clear: Takeda is responsible for global development, manufacturing and commercialization outside Greater China, while Innovent retains Greater China rights to self-commercialize. This is a classic “do China ourselves, hand overseas to a top multinational” structure. Its benefits are obvious — overseas registration, global supply chain and US/European commercialization, the links where Innovent was weak, are filled in one go; while the hardest market with the most CLDN18.2 patients, China, stays in its own hands. Completing the BD around the Phase 3 readout, the window of highest valuation, was also almost textbook timing.

If gastric cancer is the “certainty” IBI343 has essentially locked in, pancreatic cancer is the bet not yet paid out but with higher odds.

First, the market backdrop. China is the absolute main battlefield for gastric cancer — about 360,000 new cases a year (2022 data), close to 40% of the global total, of which about 30–40% are CLDN18.2-positive. This means that even in later lines, the gastric cancer patient base is large enough to support an ADC commercially. But the problem with gastric cancer is “too many people splitting the pie”: first line is held by zolbetuximab, second line is pressed by CMG901, and the third-line-plus window IBI343 truly has to itself is a rather small corner of this big market. Pancreatic cancer is a completely different situation — about 120,000 new cases a year in China, over 90% carrying KRAS mutations, almost every past targeted attempt has failed, and later-line patients have essentially no drugs available. In such a “blue ocean inside a red ocean”, whoever first delivers a regimen that extends survival will almost monopolize the entire lane.

Pancreatic ductal adenocarcinoma is recognized as one of the hardest solid tumors to crack, with almost no effective targeted options in later lines. In the Phase 1 dose-expansion cohort, patients with CLDN18.2-positive (1+/2+/3+ ≥60%) advanced pancreatic cancer receiving 6 mg/kg monotherapy had a confirmed objective response rate of about 23%, disease control rate of about 81%, median progression-free survival of about 5.3–5.4 months and median overall survival of about 9.1 months. In a tumor type where overall survival is usually counted in “months” and past targeted attempts have repeatedly failed, these are weighty numbers.

That is why pancreatic cancer monotherapy received NMPA Breakthrough Therapy Designation and FDA Fast Track. More critically, the confirmatory pancreatic cancer Phase 3 G-HOPE-002 (NCT07066098) has started and is recruiting — a randomized, double-blind study comparing IBI343 monotherapy plus best supportive care with placebo plus best supportive care, enrolling CLDN18.2-positive advanced pancreatic cancer after at least two prior lines, with overall survival as the primary endpoint. If this study confirms the benefit seen in Phase 1, IBI343 has a chance to become the first CLDN18.2 therapy to actually land in pancreatic cancer — this “first” currently appears open, but it is necessary to keep close watch on whether any same-target competitor files first in pancreatic cancer. For Innovent, gastric cancer locks in the base; pancreatic cancer determines the molecule’s ceiling.

ADC conjugation, purification and quality control are inherently more complex than for monoclonal antibodies, and process scale-up risks are higher. But on this point IBI343 happens to benefit from the partnership structure: manufacturing outside Greater China is Takeda’s responsibility, and Innovent only needs to ensure supply for Greater China. This effectively transfers the bulk of the pressure of global scale-up and multi-region GMP compliance to a multinational with a mature global supply system, significantly reducing Innovent’s own supply chain burden. Given that Innovent’s overall gross margin has reached 87.2%, the higher manufacturing cost of an ADC is still within its tolerable range.

This is the point most easily overlooked in analyzing IBI343, yet the one that most needs to be spelled out. In the bundled deal with Takeda, the heavier piece is actually IBI363 — the PD-1/IL-2α bispecific, for which Innovent not only retained US co-commercialization rights (Takeda/Innovent share costs 60/40 and split US profit and loss 60/40, with Takeda leading), but also participates in global co-development. By comparison, IBI343 overseas is a relatively “clean” license: hand it over, collect the money, take the royalties.

This means that in Innovent’s oncology portfolio ranking, IBI343 is an asset that has been well monetized and substantially de-risked, but not necessarily the one receiving the most company attention and resources. The good side is that it does not conflict by target with the portfolio’s existing PD-1 (Tyvyt), and can instead create combination and sequencing synergies in GI tumors; the less good side is that when resources, clinical teams and commercial focus need to be allocated between IBI363 and IBI343, the former will most likely take priority. For the pace of value realization in China, this is an internal variable that needs ongoing observation.

IBI343 is a case of “getting most of the key decisions right”: it chose a differentiated payload, designed a registration study it could win, went global in the most valuable window, and kept the hardest market, China, for its own mature commercial platform. In the gastric cancer battle, it has already taken the lead in terms of Phase 3 progression-free survival and NDA acceptance — though formal approval and complete overall survival data are still some distance away.

But its story is not finished. What really determines its value ceiling is whether pancreatic cancer delivers, whether it can bite off share of the first-line market from zolbetuximab and CMG901, whether Chinese and Japanese data can withstand scrutiny from US and European regulators, and whether it can secure resources within Innovent’s portfolio commensurate with its potential. For a molecule already “validated globally”, all the remaining suspense lies in these questions.