ASCO 2026: From Replicating PD-1 to "Defining the Next-Generation Framework", Three Routes of China's Immunotherapy

If we were to write the history of tumor immunotherapy (Immuno-Oncology, IO) over the past decade, PD-1/PD-L1 inhibitors would undoubtedly be the absolute protagonists. From Merck's Keytruda and Bristol Myers Squibb's Opdivo to a series of Chinese products such as Sintilimab, Tislelizumab, and Camrelizumab, this class of drugs, with the core mechanism of "releasing the immune brake," has reshaped the first-line treatment landscape for multiple cancer types, including non-small cell lung cancer (NSCLC), melanoma, kidney cancer, gastric cancer, esophageal cancer, and liver cancer.

However, at the same time, the "ceiling" of PD-1/PD-L1 monoclonal antibodies has gradually emerged in clinical practice:

● Limited objective response rate (ORR): In most cancer types, the ORR of PD-1 monotherapy is only around 20% - 30%, which means that most patients cannot directly benefit from single-agent immunotherapy.

● Primary and secondary resistance: Some patients have no response to PD-1 from the beginning (primary resistance), and some patients initially respond well but experience tumor progression again after using the drug for some time (secondary resistance).

● Weaker benefits for patients with low/negative PD-L1 expression: The lower the PD-L1 expression level on the surface of tumor cells, the worse the efficacy of PD-1 monoclonal antibodies usually is. This group of patients has always lacked better immunotherapy options.

For this reason, the core proposition in the global IO field has changed in recent years - "Adding on the PD-1 framework" to improve the response rate, overcome resistance, and cover a wider population. Chinese pharmaceutical companies have presented three clear technological routes in immunotherapy after PD-1, and ASCO 2026 is an important observation window for this proposition.

First, the three routes target different populations and different clinical scenarios. They are not necessarily in direct competition - PD-1 × VEGF is more targeted at large first-line cancer types and competes directly with PD-1 monoclonal antibodies; PD-1 × IL-2 differentiates itself in the weak areas of PD-1; PD-1 × VEGF × CTLA-4 explores synergy in a broader range of cancer types and treatment lines. This means that the future pattern of next-generation IO is likely not a "winner-takes-all" situation but a coexistence stratified by population and cancer type.

Second, China is already in the global first echelon in these three routes. Ivonescimab entered the ASCO 2026 Plenary, IBI363 was licensed to Takeda, and CS2009 is a globally first-in-class trispecific antibody. This means that Chinese IO research and development has moved from the stage of "copying PD-1" to the stage of "defining the next-generation framework," which is the most important change at the industrial level.

Third, the core contradiction of "adding on" is always the balance between efficacy and toxicity. The more mechanisms are combined, the higher the risk of potential toxicity accumulation. CTLA-4 is particularly typical - historically, its combination with PD-1 has often failed due to toxicity issues. The good safety performance of CS2009 in Phase I data is a highlight, but whether it can maintain this in Phase III and longer follow-up is one of the core focuses of this route.

Route 1: The Hottest Track, PD-1 × VEGF Bispecific Antibody

Due to its synergistic anti-tumor mechanism of "immune activation + anti-angiogenesis," the PD-1/VEGF bispecific antibody is regarded as a promising next-generation therapy to replace PD-1 monoclonal antibodies and is also a key pipeline for MNCs to acquire.

PD-1/PD-L1 can be understood as the "brake" of the immune system. Under normal circumstances, T cells receive the "stop attacking" signal through the PD-1 molecule to avoid accidentally harming normal tissues. However, many tumor cells "cunningly" express PD-L1 on their surface and actively press the brake of T cells, thereby escaping immune attack. The role of PD-1/PD-L1 inhibitors is to forcefully release this brake.

VEGF (vascular endothelial growth factor) is a key signaling molecule that promotes the formation of new blood vessels. Tumors often secrete a large amount of VEGF to "build their own blood supply lines." Anti-VEGF drugs (such as Bevacizumab) cut off this blood supply. In recent years, studies have also found that VEGF not only "feeds" tumors but also participates in shaping a tumor microenvironment that suppresses immune cells - this is an important mechanistic basis for the combination of PD-1 × VEGF.

The core idea of the PD-1 × VEGF bispecific antibody is to integrate two mechanisms that have been individually proven effective into one molecule:

● One end binds to PD-1: Releases the immune brake of T cells;

● The other end binds to VEGF: Inhibits tumor angiogenesis and remodels the immunosuppressive microenvironment.

These two mechanisms seem separate but actually echo each other - tumors with high VEGF expression often have a "colder" immune environment (difficult for T cells to infiltrate, enriched in regulatory T cells and inhibitory myeloid cells), and simply releasing PD-1 sometimes has no effect. Combining with anti-VEGF can transform the tumor microenvironment into a "hot tumor" and allow the effects of PD-1 inhibitors to be released.

Ivonescimab, a globally first-in-class PD-1/VEGF bispecific antibody developed by Kangfang Biotech is one of the most concerned products on this route and is also the only Chinese study to enter the Plenary Session of ASCO 2026, which was officially reported at the conference on June 1st.

Data released by The Lancet on May 31st showed that in the Phase III HARMONi-6/AK112-306 study, the combination of 【Ivonescimab and chemotherapy】 compared with 【Tislelizumab and chemotherapy】 for the first-line treatment of advanced squamous non-small cell lung cancer (sq-NSCLC) showed a significant positive result in OS (overall survival). A total of 532 subjects were enrolled in the HARMONi-6 study. The proportion of central squamous cell carcinoma was about 63%, the proportion of PD-L1 TPS < 1% was 39.0%, and the proportion of patients with multi-site tumor metastasis/liver metastasis/brain metastasis was about 33.8%.

The results of the pre-specified interim analysis evaluated by the IDMC showed that the study reached the primary endpoint, all key secondary endpoints, and all secondary endpoints, including achieving positive results in PFS and OS with clinical and statistical benefits. The safety profile was controllable and consistent with previous study results. The study results showed that compared with Tislelizumab combined with chemotherapy, Ivonescimab combined with chemotherapy could significantly prolong the OS of patients. Compared with PD-1 combined with chemotherapy, the risk of lung cancer death was further reduced by 34% with Ivonescimab combined with chemotherapy.

Previously, as the world's first and only approved first-in-class PD-1/VEGF bispecific antibody, Kangfang Biotech's Ivonescimab won in a head-to-head trial against Pembrolizumab, becoming the world's first and only drug to prove significantly superior efficacy to Keytruda in a single-agent head-to-head Phase III clinical study. This laid a key safety foundation for the combination of bispecific antibodies and ADCs and added fuel to the PD-1/VEGF track.

From a commercial perspective, in 2025, Kangfang Biotech achieved new drug sales revenue of 3.033 billion yuan, mainly from Ivonescimab and Cadonilimab. In the indication of non-small cell lung cancer resistant to EGFR-TKI, Ivonescimab is currently the only immunotherapy globally to achieve significant positive results in both PFS (progression-free survival) and OS (overall survival). At the end of 2025, its indication for the first-line treatment of PD-L1-positive non-small cell lung cancer was successfully added to the national medical insurance, and it is expected to see further growth in 2026. During JPM 2026 in January this year, Summit said that it had submitted a marketing application for Ivonescimab to the US FDA.

In the BD field, the PD-1/VEGF bispecific antibody is almost the most popular and highest-selling pipeline type in the history of cross-border BD of domestic innovative drugs, forming a typical "in China for Global" pattern. According to Caijing magazine, as of July 2025, among the 13 PD-1/VEGF bispecific antibodies in development globally, 8 were developed by Chinese companies, and among the 7 in the clinical stage, 6 were from China; among the 15 PD-L1/VEGF bispecific antibodies in development globally, 10 were from China, and all 6 new drugs in the clinical stage were developed by Chinese companies.

Among the top ten global MNCs, 4 have already entered this track through BD transactions, forming a fierce "second echelon" pattern:

● Lixin Pharma/Merck (LM-299): In November 2024, it was licensed to Merck with a down payment of $588 million and milestone payments of up to $2.7 billion.

● 3SBio/Pfizer (SSGJ-707): In May 2025, it was licensed to Pfizer with a down payment of $1.25 billion and a total transaction value of $6.05 billion.

● Primis Biotech/BioNTech/BMS (PM8002): In June 2025, BMS reached a cooperation with BioNTech on Primis' PD-L1/VEGF bispecific antibody. The down payment was up to $1.5 billion, and the potential total transaction value was $11.1 billion, with a resale difference of up to $9 billion.

● Rongchang Biotech/AbbVie (RC148): In January 2026, with a down payment of $650 million and milestone payments of up to $495 million, a total transaction value of up to $5.6 billion, AbbVie was granted exclusive rights for development, production, and commercialization outside the Greater China region.

Route 2: PD-1 × IL-2 Fusion Protein

IBI363 (TAK-928) of Innovent Biologics takes another route - a dual mechanism of "checkpoint blockade + cytokine activation."

IL-2 (Interleukin-2) is a cytokine that can strongly activate and expand T cells and is the "accelerator" of the immune system. Early high-dose IL-2 did show complete remission effects in a few cancer types, but it had great toxicity, which limited its clinical application. The next-generation design idea is α-bias - only activating the IL-2 receptor subunits that are beneficial for expanding effective T cells and bypassing the branches that bring toxicity.

IBI363 is a globally first-in-class PD-1/IL-2α-biased bispecific antibody fusion protein/biased bispecific antibody, focusing on the treatment bottlenecks of PD-1-resistant, cold tumors, and tumors with low PD-L1 expression. Its two ends have their own functions:

● One end binds to PD-1, releasing the immune brake of T cells: The PD-1 binding arm can simultaneously block PD-1 and selectively deliver IL-2.

● The other end "α-bias" activates the IL-2 pathway, stepping on the accelerator and expanding effective tumor-specific T cells: The IL-2 arm of IBI363 has been designed and modified to retain its affinity for IL-2 Rα but weaken its binding ability to IL-2Rβ and IL-2Rγ, thereby reducing toxicity.

At ASCO 2026, IBI363 presented two sets of data in completely different directions but equally crucial:

(1) Aiming at patients with refractory NSCLC in the later lines without treatment options, showing long-term survival benefits

The study showed that in patients with advanced squamous and adenocarcinoma NSCLC who had failed PD-1/PD-L1 treatment, IBI363 showed strong OS benefits. For patients with advanced NSCLC without driver gene mutations who had progressed after previous immunotherapy, the subsequent treatment options were limited, and clinically, chemotherapy regimens such as docetaxel were usually relied on. If a new mechanism immunotherapy drug can show longer survival benefits in this population, it means that it has "targeted areas that PD-1 cannot reach."

The updated ASCO abstract showed that as of November 20, 2025, a total of 136 NSCLC subjects had received IBI363 monotherapy. The patients in this dose group achieved remarkable survival results: the median progression-free survival (PFS) reached 10.1 months, the median overall survival (OS) reached 18.2 months, and the 24-month overall survival rate was as high as 47.8%. In other words, nearly half of the patients were still alive two years after treatment. As the follow-up time extended, the survival curve gradually flattened in the later stage, showing the characteristics of typical long-term benefits of immunotherapy.