Akeso, Inc. has presented real-world data from a retrospective cohort study at the 2026 ASCO Gastrointestinal Cancers Symposium showing that cadonilimab plus chemotherapy significantly outperformed PD-1 inhibitor plus chemotherapy regimens in PD-L1-low (CPS <5) gastric and gastroesophageal junction cancers. The findings, now accepted for publication in the Journal of Gastrointestinal Oncology, support earlier trial results from the Phase III COMPASSION-15 study and strengthen cadonilimab’s positioning as a potential new standard of care for first-line treatment of PD-L1-low gastric cancer, a segment long underserved by current checkpoint inhibitors.

Why this real-world comparison may mark a turning point for PD-L1-low gastric cancer patients

The new study delivers something rarely seen in the immuno-oncology field: head-to-head real-world comparative effectiveness data favoring a bispecific antibody over established PD-1 inhibitors in a population known to under-respond to current immunotherapy. While PD-1 inhibitors remain a mainstay in the treatment of PD-L1-high gastric cancer, their diminished efficacy in PD-L1-low and negative tumors has created a clinical gap that most immune checkpoint strategies have failed to bridge.

Cadonilimab, a PD-1/CTLA-4 bispecific antibody, appears to have made a meaningful step forward in this regard. The reported hazard ratios—0.49 for overall survival and 0.43 for progression-free survival—translate into risk reductions of 51 percent and 57 percent, respectively, compared to standard PD-1 plus chemotherapy regimens. Notably, these survival benefits were achieved without an increase in high-grade adverse events, indicating the drug’s dual-targeting mechanism may offer improved efficacy without compromising tolerability.

For patients with PD-L1 CPS <5, who comprise nearly half of all advanced gastric cancer cases, this data signals a new level of clinical relevance. It also reopens the conversation about immunotherapy options for populations previously left out of regulatory approvals and global guidelines due to poor response data.

What this reveals about bispecific strategies versus single-target checkpoint blockade

Cadonilimab’s performance reinforces growing industry interest in bispecific or multispecific checkpoint inhibitors. Unlike single-target PD-1 antibodies that depend on PD-L1 expression as a predictor of efficacy, cadonilimab’s mechanism may bypass this dependency by concurrently modulating two immune checkpoints. This could account for its ability to deliver benefit in a biologically colder PD-L1-low tumor microenvironment, a setting where PD-1 blockade alone has repeatedly underperformed in trials.

Industry observers note that the study’s methodology—a propensity-matched, real-world comparison with consistent baseline characteristics—provides a more pragmatic reflection of clinical effectiveness than idealized Phase III settings. With median follow-up of 11 months, the data holds up not just in terms of statistical significance, but also in practical endpoints that influence treatment guidelines and reimbursement decisions.

The implications extend beyond cadonilimab. The study adds to a broader narrative in oncology suggesting that bispecific and dual-targeting modalities may become more than second-line innovations. They are increasingly being evaluated as first-line contenders, especially in populations where single-target agents have biologically hit a wall.

What global regulators and guideline committees may watch next

Cadonilimab is already approved in China for the first-line treatment of advanced gastric cancer across all PD-L1 expression levels. Its inclusion on China’s National Reimbursement Drug List beginning in 2026 positions it for rapid adoption domestically. However, the real litmus test for cadonilimab’s global ambitions will be the success of its ongoing Phase III international registrational study, COMPASSION-37.

That trial, comparing cadonilimab plus chemotherapy versus nivolumab plus chemotherapy, was approved by the United States Food and Drug Administration in late 2025 and will include global enrollment. If results align with both COMPASSION-15 and the current real-world study, the evidence package could support a regulatory submission outside China—potentially reshaping the landscape in jurisdictions where PD-1 inhibitors are currently restricted to PD-L1-positive populations.

Regulatory watchers suggest that while real-world data is increasingly accepted as supportive evidence, especially in post-marketing or label expansion contexts, it is unlikely to be sufficient on its own for full approval in the United States or Europe. The availability of randomized, head-to-head prospective trial data will remain critical.

What could challenge cadonilimab’s path to broader adoption

While the data is promising, several potential obstacles remain. First, bispecific antibodies often present complex manufacturing challenges, from production yields to formulation stability. Akeso’s ability to scale cadonilimab production for global markets will be watched closely, especially as manufacturing bottlenecks have hampered rollout for other bispecific therapies.

Second, the economic argument for cadonilimab will need to be compelling. In regions where health technology assessments play a central role in reimbursement, such as the United Kingdom or Germany, price-to-benefit ratios will be critical—especially given that PD-1 inhibitors are already entrenched in formularies for PD-L1-high patients.

Third, while the safety profile in the real-world study appeared manageable, the long-term immunotoxicity profile of PD-1/CTLA-4 bispecifics remains a question mark in broader populations. Even small imbalances in immune-related adverse events could have implications for guideline placement, especially in frontline settings where risk tolerance is lower.

Finally, there is the issue of competing innovations. The gastric cancer pipeline remains active, with other companies exploring personalized immunotherapy, antibody-drug conjugates, and CAR-T modalities. Cadonilimab may lead for now in the PD-L1-low segment, but it will need to hold that ground as newer modalities emerge.

Why this could catalyze a shift in real-world oncology trial design

Beyond the drug itself, the study’s design and execution offer a signal about the growing role of real-world evidence in oncology. The retrospective, propensity-matched methodology allowed for comparative insights without the time and resource demands of a prospective trial. While not a replacement for randomized data, such designs are increasingly being viewed as essential complements, especially in fast-moving therapeutic areas.

Clinicians tracking the field believe that if regulators and payers continue to accept real-world data as a valuable adjunct, companies with agile trial infrastructure like Akeso could move more quickly from local approval to global strategy. It may also encourage other developers to invest in real-world comparative studies to bolster clinical claims in patient subsets where standard-of-care data remains thin.

Final takeaways: momentum, not a milestone

Cadonilimab’s performance in this real-world setting is not the final word on first-line therapy for PD-L1-low gastric cancer—but it represents a significant momentum shift. With supportive data from multiple sources, a favorable tolerability profile, and a regulatory foothold in China, Akeso has positioned cadonilimab as a serious global contender.

However, the ultimate shift in treatment paradigms will depend on Phase III results, manufacturing scalability, payer willingness, and comparative innovation from rivals. For now, cadonilimab has done what few others have: deliver meaningful outcomes in one of the most difficult-to-treat immunotherapy populations. The oncology world will be watching what comes next.

  Akeso, cadonilimab, COMPASSION-15, gastric cancer, gastroesophageal junction cancer, immunotherapy, oncology, PD-1 inhibitor, PD-L1-low