COMPASSION-37's first patient enrollment is a trial initiation, not a data readout — and that distinction is the sharpest verdict available. Cadonilimab is a PD-1/CTLA-4 bispecific antibody entering a first-line HER2-negative G/GEJ adenocarcinoma market already occupied by three approved anti-PD-1 agents: nivolumab (CheckMate 649, Phase 3 RCT), pembrolizumab (KEYNOTE-859, Phase 3 RCT, median OS 13.0 vs. [1] 11.4 months, HR 0.74 in CPS ≥1), and tislelizumab (RATIONALE-305, Phase 3 RCT). A nine-trial network meta-analysis found no statistically significant OS differences between any ICI-containing regimens in this indication, establishing that the market has reached a mechanistic plateau that only a genuinely differentiated approach can break. Cadonilimab's dual-checkpoint mechanism is that differentiation hypothesis — but it carries a critical safety precedent: the nivolumab plus ipilimumab arm of CheckMate 649, the closest functional analogue for dual PD-1/CTLA-4 blockade in this exact population and line, was closed due to increased toxicity and premature deaths. [2] The bispecific format may produce a different pharmacodynamic profile than two co-administered antibodies, but this must be demonstrated, not assumed. The comparator arm — chemotherapy with or without nivolumab — is materially harder than the chemotherapy-alone control CheckMate 649 faced, and its heterogeneous composition (some patients receiving nivolumab, some not) introduces analytical complexity that must be pre-specified in the statistical analysis plan. [3] HTA bodies including the G-BA, NICE, and PBAC have all signaled that new first-line G/GEJ entrants must demonstrate incremental benefit over established immunotherapy combinations, not merely over chemotherapy alone, and that cost-effectiveness versus nivolumab-containing regimens is the operative bar. No precedent clears the full mechanistic-fit bar for a PD-1/CTLA-4 bispecific in this indication — the closest available (nivolumab, Phase 3 RCT, partial PD-1 overlap) fails the CTLA-4 component test. [4] The sharpest risk is that the dual-checkpoint safety signal from CheckMate 649's closed arm, combined with the absence of any efficacy data for cadonilimab in G/GEJ, leaves the entire investment thesis resting on an unproven mechanistic hypothesis against a harder comparator than any prior approval in this space has cleared.
COMPASSION-37 has enrolled its first patient with no efficacy or safety data for cadonilimab in G/GEJ adenocarcinoma in the retrieved evidence. [5] The dual-checkpoint safety precedent from CheckMate 649's closed nivolumab-plus-ipilimumab arm (Phase 3 RCT, same population) is the only mechanistically adjacent outcome data available, and it is a negative safety signal.
| Indication | Gastric or Gastroesophageal Junction Adenocarcinoma, HER2-negative, First-Line |
| Drug | Cadonilimab |
| Mechanism of Action | PD-1/CTLA-4 bispecific antibody |
| Company | Akeso, Inc. |
| Trial Phase | Phase III |
| Trial Acronym | COMPASSION-37 |
| Category | Clinical Trial Event |
| Sub Category | Trial Initiation / First Patient In (FPI) |
| Therapeutic Area | Oncology |
| Comparator Arm | Chemotherapy with or without nivolumab |
| Patient Subpopulation | HER2-negative, previously untreated, unresectable or metastatic gastric or gastroesophageal junction (G/GEJ) adenocarcinoma |
| Line of Therapy | First-line |
| Biomarker Status | HER2-negative, PD-L1 expression (high, low, negative, CPS ≥5, CPS <5) |
| Global Principal Investigators | Dr. Yelena Y. Janjigian, Professor Markus Möhler, Professor Jiafu Ji, Professor Lin Shen |
| Cadonilimab China Approvals | Approved in China for first-line gastric cancer (Sep 2024), first-line cervical cancer (May 2025), and recurrent/metastatic cervical cancer (Jun 2022) |
| Previous Phase III Trial | COMPASSION-15 (China) |
| Concurrent Phase II Trial | NCT07631000 (US, perioperative setting) |
Akeso Enrolls First Patient in Global Phase III Cadonilimab Trial
Akeso, Inc. announced the enrollment of the first patient in COMPASSION-37 (AK104-311), a global, multicenter Phase III head-to-head trial. This study evaluates cadonilimab, Akeso's first-in-class PD-1/CTLA-4 bispecific antibody, in combination with chemotherapy against chemotherapy with or without nivolumab. The trial targets HER2-negative, previously untreated, unresectable or metastatic gastric or gastroesophageal junction (G/GEJ) adenocarcinoma as a first-line treatment. The primary goal is to assess cadonilimab's efficacy and safety across diverse patient populations and to determine if dual PD-1/CTLA-4 immune checkpoint blockade can improve outcomes beyond the current standard of care, particularly for patients with low PD-L1 expression.
- COMPASSION-37 is a global, multicenter, head-to-head Phase III trial designed to rigorously compare cadonilimab plus chemotherapy against chemotherapy with or without nivolumab as a first-line treatment for HER2-negative advanced gastric or G/GEJ adenocarcinoma. The study aims to validate cadonilimab's potential to improve patient outcomes, especially in those with low PD-L1 expression, addressing a significant global unmet medical need.
- Cadonilimab is a first-in-class PD-1/CTLA-4 bispecific antibody developed by Akeso, engineered to deliver synergistic antitumor activity with a favorable safety profile compared to conventional combination regimens. Its development is supported by positive Phase III results from the COMPASSION-15 study in China, which demonstrated significant overall survival benefits in HER2-negative advanced gastric cancer, including a promising trend in patients with low PD-L1 expression.
- The trial addresses a critical question in gastric cancer immunotherapy: how to broaden the population that benefits from treatment, particularly beyond patients with high PD-L1 expression. While PD-1 inhibitor plus chemotherapy is a standard first-line strategy, its efficacy is often linked to PD-L1 levels. COMPASSION-37 seeks to determine if dual PD-1/CTLA-4 blockade can offer more effective and broadly applicable first-line options across different PD-L1 levels and patient populations worldwide.
Addressing Unmet Needs in First-Line HER2-Negative Gastric Cancer
Despite meaningful advances in first-line systemic therapy for HER2-negative advanced gastric/gastroesophageal junction (G/GEJ) adenocarcinoma, significant clinical and biological challenges persist. The expanding therapeutic landscape — spanning immune checkpoint inhibitors (ICIs), anti-Claudin 18.2 agents, and cytotoxic chemotherapy — has introduced new complexities in patient selection, sequencing, and resistance management that demand strategic navigation.
Biomarker co-expression and treatment selection ambiguity: In patients with tumors co-expressing CLDN18.2 and PD-L1, the optimal first-line strategy remains unresolved. ICIs may offer greater long-term survival benefit, particularly in patients with high PD-L1 combined positive score (CPS), while zolbetuximab may be preferred in patients with negative-to-low CPS or ICI contraindications. A practical decision-making threshold of PD-L1 CPS ≥ 10 has been proposed to favor ICI-based regimens, but treatment selection must be further tailored to comorbidities, comprehensive molecular characterization, and available subsequent options.
Intratumoural heterogeneity and biomarker instability: CLDN18.2 expression demonstrates high prevalence of heterogeneous intratumoral distribution — 87% of CLDN18.2-positive cases (81/93) exhibited heterogeneous expression patterns, including superficial, random, and invasive-front patterns. Biopsy-surgery concordance decreased to 73% in patients with heterogeneous expression, and was particularly low (65%) in the superficial pattern. Furthermore, among patients who underwent neoadjuvant chemotherapy, only 4 of 11 initially CLDN18.2-positive cases remained positive after treatment, highlighting the risk of biomarker loss following chemotherapy and its implications for patient selection for targeted agents such as zolbetuximab.
Tolerability challenges with zolbetuximab: Nausea and vomiting were reported in more than three-quarters of patients receiving zolbetuximab, occurring most frequently and severely during the first infusion due to the loading dose and faster infusion rate. Optimal management requires strict adherence to high-risk antiemetic protocols — including NK-1/5-HT3 antagonists and corticosteroids — and careful infusion rate control using a stop-and-go strategy, underscoring the need for specialized supportive care infrastructure to maintain treatment adherence.
Resistance driven by extrachromosomal DNA (ecDNA) dynamics: Intratumoural genetic heterogeneity in receptor tyrosine kinases (RTKs) such as FGFR2, driven by dynamic quantitative and qualitative changes in ecDNA, contributes to heterogeneity in RTK protein expression and resistance to RTK inhibitors. Resistant cells demonstrated diverse ecDNA changes — including chimeric ecDNA, large ecDNA, and increased ecDNA numbers — associated with high expression and rephosphorylation of FGFR2, representing a mechanistic challenge for targeted therapy strategies in this setting.
Comparative efficacy differentiation across immunotherapy-chemotherapy combinations: While immunotherapy plus chemotherapy is associated with greater PFS and OS benefits compared with chemotherapy alone in HER2-negative advanced gastric cancer, differentiating between regimens remains complex. In a network meta-analysis of 25 studies encompassing 14,389 patients and 23 first-line treatments, sintilimab plus capecitabine plus oxaliplatin (Sint-XELOX) demonstrated the best OS (SUCRA 81%) and PFS (SUCRA 96%), yet head-to-head data across all regimens are limited, complicating definitive treatment algorithm construction.
Prognostic heterogeneity within biomarker-selected subgroups: Even within molecularly defined populations, clinical outcomes vary substantially. Among patients with deficient mismatch repair (d-MMR) gastric cancer treated with nivolumab plus chemotherapy, those with a high neutrophil-to-lymphocyte ratio (NLR ≥ 3.80) had markedly worse PFS benefit (HR 0.58) compared to those with a lower NLR (HR 0.10), and survival outcomes were poor even with immunotherapy in the high-NLR subgroup. PD-L1 CPS did not differentiate survival outcomes among d-MMR patients treated with nivolumab chemotherapy, indicating that current biomarker frameworks remain insufficient to fully predict benefit within selected populations.
COMPASSION-37: Evaluating Cadonilimab in First-Line Gastric Cancer
Several pivotal phase 3 trials have defined the first-line treatment landscape for HER2-negative gastric or gastroesophageal junction (GEJ) adenocarcinoma, evaluating immunotherapy combinations, targeted agents, and chemotherapy switch strategies across diverse global populations. The trials below span PD-1/PD-L1 inhibitor combinations, CLDN18.2-targeted therapy, and antiangiogenic switch maintenance, with progression-free survival and overall survival as the dominant endpoints.
| Trial | Intervention | Control | Population | Phase | Primary Endpoint | Key Efficacy Results |
|---|---|---|---|---|---|---|
| KEYNOTE-859 (Japan subgroup) | Pembrolizumab 200 mg IV Q3W + chemotherapy (≤35 cycles) | Placebo + chemotherapy | Untreated locally advanced unresectable or metastatic HER2-negative gastric or GEJ adenocarcinoma; n=101 (Japan) | III | OS | Median OS: 16.8 months (pembrolizumab) vs. 13.3 months (placebo); HR 0.71 (95% CI 0.44–1.13); Grade 3–4 TRAEs: 41.7% vs. 39.6% |
| SPOTLIGHT | Zolbetuximab (800 mg/m² loading, then 600 mg/m² Q3W) + mFOLFOX6 | Placebo + mFOLFOX6 | CLDN18.2-positive (≥75% tumour cells, moderate-to-strong staining), HER2-negative, untreated locally advanced unresectable or metastatic gastric or GEJ adenocarcinoma; n=565 | III | PFS by independent review committee (all randomised patients) | Median PFS: 10.61 months vs. 8.67 months; HR 0.75 (95% CI 0.60–0.94; p=0.0066); Median OS HR 0.75 (95% CI 0.60–0.94; p=0.0053); Grade ≥3 AEs: 87% vs. 78% |
| ARMANI | Paclitaxel 80 mg/m² (days 1, 8, 15) + ramucirumab 8 mg/kg (days 1, 15) Q28D (switch maintenance) | Continuation of FOLFOX or CAPOX × 12 weeks, then fluoropyrimidine monotherapy maintenance | Advanced HER2-negative gastric or GEJ cancer with disease control after 3 months of FOLFOX or CAPOX; n=280; ECOG PS 0–1 | III | PFS (intention-to-treat) | Median PFS: 6.6 months vs. 3.5 months; HR 0.61 (95% CI 0.48–0.79; p=0.0002); 24-month restricted mean PFS: 8.8 months vs. 6.1 months (p=0.0010); Grade 3–4 neutropenia: 26% vs. 10% |
| Meta-analysis (PD-1 inhibitor + chemo, 6 RCTs) | PD-1 inhibitor + chemotherapy | Chemotherapy alone | HER2-negative advanced GC/GEJC; n=6,294 | III (pooled) | OS, PFS, ORR | OS HR 0.79 (95% CI 0.75–0.84; P<.00001); PFS HR 0.75 (95% CI 0.70–0.80; P<.00001); ORR RR 1.22 (95% CI 1.15–1.29; P<.00001); MSI-H subgroup OS HR 0.35 (95% CI 0.21–0.59; P<.0001) |
| Meta-analysis (PD-L1 cutoff, 12 studies) | ICI monotherapy or combined immunotherapy | Various comparators | Gastric cancer; n=6,488 | Prospective trials (pooled) | ORR, PFS HR, OS HR by PD-L1 CPS subgroup | ICI monotherapy: OS HR 0.84 (95% CI 0.74–0.96) in CPS ≥1; PFS HR 1.38 (95% CI 0.91–2.09) in CPS ≥1; Combined immunotherapy: OS HR 0.81 (95% CI 0.71–0.92), PFS HR 0.77 (95% CI 0.69–0.86) in CPS ≥1 |
Cadonilimab's Expanding Role Across Multiple Cancer Types
Cadonilimab, a first-in-class bispecific PD-1/CTLA-4 antibody, is under active investigation across a broad range of solid tumor indications beyond HER2-negative advanced gastric or gastroesophageal junction adenocarcinoma. The trials span both randomized controlled and single-arm designs, reflecting the drug's versatility across tumor types with distinct immunological profiles.
Recurrent or Metastatic Cervical Cancer: Cadonilimab is being evaluated in combination with chemotherapy with or without bevacizumab. Real-world prospective observational data from 51 patients reported an ORR of 72.5% and a DCR of 90.2% at first tumor evaluation, with a median PFS of 7.0 months (IQR: 4.0–10.0). The intervention model encompasses three regimens: cadonilimab plus chemotherapy plus bevacizumab (n=22), cadonilimab plus chemotherapy (n=24), and cadonilimab monotherapy (n=5). Cadonilimab has also received regulatory approval for cervical cancer, representing one of eight bispecific antibodies approved for solid tumors.
Advanced/Metastatic Non-Small Cell Lung Cancer (NSCLC) with STK11 Mutation: A single-center, open-label, single-arm Phase II trial is evaluating cadonilimab (10 mg/kg on Day 1) plus chemotherapy as front-line treatment in treatment-naïve patients harboring STK11 genetic aberration. The chemotherapy backbone is pemetrexed (500 mg/m²) and carboplatin (AUC=5) for nonsquamous NSCLC, or abraxane (100 mg/m²) and carboplatin (AUC=5) for squamous NSCLC, for 4 cycles, followed by maintenance with cadonilimab plus pemetrexed or abraxane. The primary endpoint is ORR and safety.
Unresectable Locally Advanced or Metastatic Pancreatic Ductal Adenocarcinoma (PDAC): A single-center retrospective real-world study evaluated cadonilimab (10 mg/kg, Q3W) combined with AG chemotherapy in 14 patients. The DCR reached 85.7% and ORR was 14.3%, with a median PFS of 7.87 months and a median OS of 11.45 months. Grade 3 treatment-related adverse events occurred in 57.1% of patients, with no grade 4 events observed.
Advanced Cholangiocarcinoma (Second-Line): A single-arm, prospective Phase II trial is investigating cadonilimab (6 mg/kg on Day 1 of each 21-day cycle) combined with liposomal irinotecan (70 mg/m²), leucovorin (400 mg/m²), and fluorouracil (400 mg/m²) in 51 patients with locally advanced or metastatic biliary tract cancer. The primary endpoint is ORR, with secondary endpoints including OS, PFS, DCR, and adverse event incidence rate (registered as NCT06438822).
Cadonilimab's Pivotal Bid to Redefine First-Line G/GEJ Immunotherapy
The initiation of Akeso's COMPASSION-37 Phase III trial represents a pivotal moment for the treatment of HER2-negative, unresectable or metastatic gastric and gastroesophageal junction (G/GEJ) adenocarcinoma. This global, multicenter study is designed to rigorously evaluate cadonilimab, a first-in-class PD-1/CTLA-4 bispecific antibody, in combination with chemotherapy against a current benchmark: chemotherapy with or without nivolumab. This head-to-head comparison is crucial because while PD-1 inhibitors plus chemotherapy have become a standard, their efficacy, particularly in patients with low PD-L1 expression, remains a significant unmet need. Existing evidence indicates that immunochemotherapy benefits can be subtle in these low CPS subgroups, highlighting the potential for a bispecific approach to offer a more pronounced advantage.
Cadonilimab has already demonstrated its potential, having significantly improved overall survival (14.1 vs 11.1 months) and progression-free survival in a prior Phase III study when combined with chemotherapy versus chemotherapy alone. The current trial seeks to build on this by directly challenging established PD-1 inhibitor regimens. If successful, cadonilimab could redefine the first-line standard of care, offering a new, potentially more effective option for a broader patient population, including those who currently derive limited benefit from existing immunotherapies. However, several factors warrant close consideration:
Safety Profile: Previous data showed a higher incidence of Grade ≥3 treatment-related adverse events with cadonilimab plus chemotherapy compared to chemotherapy alone. The competitive landscape demands a comparable or superior safety profile against nivolumab-based regimens.
Efficacy in Low PD-L1: While the trial specifically targets improving outcomes for low PD-L1 patients, achieving a significant advantage in this subgroup, where other immunotherapies have shown subtle or no clear benefit, will be a critical test.
Evolving Landscape: The G/GEJ treatment arena is dynamic, with other novel combinations emerging. Cadonilimab must demonstrate a clear and sustained advantage to carve out a dominant market position.
Ultimately, the results of COMPASSION-37 will provide essential insights into the role of bispecific immune checkpoint inhibitors in G/GEJ cancer, potentially ushering in a new era of more effective and inclusive first-line treatment strategies.
Frequently Asked Questions
References
- [1] Jiang H, Sun N et al.. Evaluating Safety and Clinical Activity of Front-line Treatment with Cadonilimab plus Chemotherapy in Advanced/Metastatic Nonsmall Cell Lung Cancer Harboring STK11 Genetic Aberration: A Protocol of Phase II Study. Clinical lung cancer. 2024 Dec. 39095236
- [2] Dent S, Cortés J et al.. Phase III randomized study of taselisib or placebo with fulvestrant in estrogen receptor-positive, PIK3CA-mutant, HER2-negative, advanced breast cancer: the SANDPIPER trial. Annals of oncology : official journal of the European Society for Medical Oncology. 2021 Feb. 33186740
- [3] Rha SY, Chung HC. Breakthroughs in the Systemic Treatment of HER2-Positive Advanced/Metastatic Gastric Cancer: From Singlet Chemotherapy to Triple Combination. Journal of gastric cancer. 2023 Jan. 36751001
- [4] Zhou J, Ye W et al.. Comparison of drug regimens for recurrent or metastatic cervical cancer: a systematic review and network meta-analysis. Frontiers in immunology. 2026. 41822503
- [5] Yang H, Li L et al.. Second-line treatment of PD-1 and CTLA-4 blockade combined with liposomal irinotecan plus leucovorin and fluorouracil for advanced cholangiocarcinoma: study protocol of a single-arm, prospective phase II trial. Therapeutic advances in medical oncology. 2024. 39492841
- [6] Picheta N, Piekarz J et al.. Zolbetuximab in the treatment of advanced gastric and gastroesophageal junction cancer: a systematic review. Frontiers in immunology. 2026. 42534935
- [7] Nagano S, Kurokawa Y et al.. Intratumoral heterogeneity and chemotherapy-induced alteration of CLDN18.2 expression in resectable gastric cancer. International journal of clinical oncology. 2026 Apr. 41604118
- [8] Rosanu NM, Gervaso L et al.. Navigating the clinical challenges of zolbetuximab in patients with claudin positive advanced gastric cancer. European journal of cancer (Oxford, England : 1990). 2025 Jun 25. 40466467
- [9] Cai T, Liang L et al.. Comparative efficacy and tolerability of first-line treatments for untreated, HER2-negative, advanced gastric cancer: systematic review and network meta-analysis. Critical reviews in oncology/hematology. 2024 Jan. 38029945
- [10] Lian X, Qi H et al.. Real-world efficacy, safety, and associated biomarkers of cadonilimab in cervical cancer: a prospective observational study. Frontiers in immunology. 2026. 42125658
- [11] Qin W, Du Y et al.. Efficacy and safety of cadonilimab combined with AG chemotherapy in patients with unresectable locally advanced or metastatic pancreatic ductal adenocarcinoma: a retrospective real-world study. Frontiers in immunology. 2025. 41142819
- [12] Liang X, Xue J et al.. Safety, tolerability, and pharmacokinetics/pharmacodynamics of JMT103 in patients with bone metastases from solid tumors. Frontiers in oncology. 2022. 35992822
- [13] Shitara K, Lordick F et al.. Zolbetuximab plus mFOLFOX6 in patients with CLDN18.2-positive, HER2-negative, untreated, locally advanced unresectable or metastatic gastric or gastro-oesophageal junction adenocarcinoma (SPOTLIGHT): a multicentre, randomised, double-blind, phase 3 trial. Lancet (London, England). 2023 May 20. 37068504
- [14] Randon G, Lonardi S et al.. Ramucirumab plus paclitaxel as switch maintenance versus continuation of first-line oxaliplatin-based chemotherapy in patients with advanced HER2-negative gastric or gastro-oesophageal junction cancer (ARMANI): a randomised, open-label, multicentre, phase 3 trial. The Lancet. Oncology. 2024 Dec. 39557058
- [15] Diamond JR, Becerra C et al.. Phase Ib clinical trial of the anti-frizzled antibody vantictumab (OMP-18R5) plus paclitaxel in patients with locally advanced or metastatic HER2-negative breast cancer. Breast cancer research and treatment. 2020 Nov. 32803633
- [16] Xie T, Zhang Z et al.. Appropriate PD-L1 Cutoff Value for Gastric Cancer Immunotherapy: A Systematic Review and Meta-Analysis. Frontiers in oncology. 2021. 34540656
- [17] Wang J, Ding B et al.. Cadonilimab plus chemotherapy for patients with HER2-negative advanced gastric or gastroesophageal junction adenocarcinoma in China and the United States: a cost-effectiveness analysis. Therapeutic advances in medical oncology. 2026. 41835338
- [18] Wang Y, Li Y et al.. Immune checkpoint inhibitors plus chemotherapy in the first-line treatment of young unresectable gastric cancer patients: a multicentre real-world study. Frontiers in oncology. 2025. 40291906
- [19] Desnoyers A, Nadler MB et al.. Comparison of treatment-related adverse events of different Cyclin-dependent kinase 4/6 inhibitors in metastatic breast cancer: A network meta-analysis. Cancer treatment reviews. 2020 Nov. 32861975
- [20] Kim IH. Emerging Targets for Systemic Treatment of Gastric Cancer: HER2 and Beyond. Journal of gastric cancer. 2024 Jan. 38225765
Contact Us
Address
One Research Ct, Suite 450
Rockville, MD 20850
For General Inquiry
info@pienomial.com
