Akeso registers Phase 3 test of cadonilimab against nivolumab in gastric cancer
The 900-patient trial pits Akeso's PD-1/CTLA-4 bispecific against a chemo-plus-nivolumab arm, a head-to-head bar cadonilimab has not faced in this setting.
Executive Summary
- Akeso has registered a new Phase 3 trial that raises the comparator bar for its lead checkpoint bispecific, testing it against an active PD-1 inhibitor rather than chemotherapy alone.
- The result would replicate or fail to replicate the survival benefit cadonilimab already showed in an earlier completed gastric cancer trial, this time against a tougher standard-of-care arm.
- Gastric cancer's PD-1 landscape is dense and has produced failures across multiple sponsors, so this trial adds one more entrant to a field where checkpoint blockade has not uniformly worked in Phase 3.
- Akeso already runs a broad cadonilimab program in this indication, and its overall trial completion rate gives the new study an execution baseline to be judged against.
The trial
The study, registered as NCT07700979, is not yet recruiting and lists a start date of August 20, 2026, with a primary completion date of August 18, 2029. It targets 900 participants across the United States, China, Germany, and Poland, and its primary endpoint is overall survival, measured from randomization to death from any cause. Secondary endpoints cover objective response rate, progression-free survival, duration of response, disease control rate, and safety, tracked over roughly 8 to 18 months depending on the measure. NCT07700979Phase 3 Study of Cadonilimab Combined With Chemotherapy vs Chemotherapy in Combination With or Without Nivolumab for the 1L Treatment of Participants With HER2-negative, Previously Untreated, Unresectable or Metastatic Gastric/GEJ AdenocarcinomaNCT07700979
What's different
Cadonilimab, a PD-1/CTLA-4 bispecific antibody, already has one completed Phase 3 trial in this same first-line gastric and GEJ adenocarcinoma population, NCT05008783, which enrolled 610 patients against a chemotherapy-plus-placebo comparator and completed in December 2024. This new trial changes the comparator: one arm gets chemotherapy alone, and the other gets chemotherapy combined with nivolumab, a marketed PD-1 inhibitor, allowing cadonilimab to be tested against an active checkpoint-inhibitor regimen rather than chemotherapy by itself. That is a different question than the one its earlier completed trial answered. NCT07700979Phase 3 Study of Cadonilimab Combined With Chemotherapy vs Chemotherapy in Combination With or Without Nivolumab for the 1L Treatment of Participants With HER2-negative, Previously Untreated, Unresectable or Metastatic Gastric/GEJ AdenocarcinomaNCT07700979
The competitive field
PD-1-targeted therapy in gastric adenocarcinoma is a dense field: 51 active industry trials are studying PD-1 blockade in this indication, and Phase 3 readouts in this target-indication pairing have failed 27% of the time across 11 completed or terminated trials, with 10 distinct sponsors recording a failure. Direct comparators using the same PD-1 target include Innovent's sintilimab and BeOne Medicines' tislelizumab, both also in Phase 2 or Phase 3 gastric adenocarcinoma programs. Bispecific PD-1/CTLA-4 antibodies remain a minority approach in this indication: bispecific antibodies have been used in only 26 of the trials studying gastric adenocarcinoma, against a much larger base of monoclonal antibody programs.
Sponsor track record
Akeso runs eight active or completed trials of cadonilimab in gastric and GEJ adenocarcinoma, including two other Phase 3 studies not yet recruiting or recruiting, and one completed Phase 3 study. Across its full pipeline of 160 trials, Akeso has completed 65 and terminated 8, for an 89% completion rate on resolved trials. The new trial's protocol carries no amendment history yet and is flagged as low operational risk at this early stage.
This analysis was produced using AI-assisted reporting systems, AppliedXL data, and official public records. These systems undergo editorial review, quality checks, and regular audits by human experts. Errors may still occur, as with any automated system. Always consult the linked primary sources. Read our AI Editorial Policy.
