Randomized phase II trial of zolbetuximab as second-line treatment in patients with claudin 18.2-positive advanced gastric/GEJ cancer with previous zolbetuximab exposure: ZELDA Authors: C. Inagaki, E. Oki, K. Ando, H. Hirano, M. Furuta, Y. Narita, T. Esaki, K. Murotani, H. Kawakami, K. Muro Published: 2026 ABSTRACT BackgroundZolbetuximab, an anti-claudin 18.2 (CLDN18.2) monoclonal antibody, combined with fluoropyrimidine- and platinum-based chemotherapy, is established as first-line standard of care for patients with CLDN18.2-positive, human epidermal growth factor receptor 2-negative advanced gastric or gastroesophageal junction (GEJ) cancer, based on the phase III SPOTLIGHT and GLOW trials. However, evidence regarding the efficacy and safety of continuing zolbetuximab beyond disease progression remains lacking. Given reports that CLDN18.2 expression is maintained after chemotherapy and the absence of clearly established resistance pathways, continued CLDN18.2-targeted therapy in the second-line setting warrants investigation.AimTo investigate whether the addition of zolbetuximab to second-line nab-paclitaxel plus ramucirumab improves overall survival (OS) in patients with CLDN18.2-positive advanced gastric or GEJ adenocarcinoma who have progressed with first-line zolbetuximab-based therapy.Trial designZELDA (jRCTs071250038) is a multicenter, open-label, randomized phase II (rP2) trial conducted by the West Japan Oncology Group and Kyushu Study Group of Clinical Cancer in Japan. A safety lead-in part (n = 6) confirms the tolerability of the triplet combination (zolbetuximab + nab-paclitaxel + ramucirumab), followed by a rP2 trial (n = 210) with a 1 : 1 allocation to nab-paclitaxel + ramucirumab (arm A, control) or zolbetuximab + nab-paclitaxel + ramucirumab (arm B, experimental). The primary endpoint is OS in the full analysis set. Integrated translational research (ZELDA-TR) is also planned. Description of protocol Background Gastric and gastroesophageal junction (GEJ) cancers are the fifth most common malignancy and the third leading cause of cancer-related mortality worldwide, causing over 40 000 deaths annually in Japan alone.1,2 In the advanced or metastatic setting, prognosis remains poor, underscoring the need for more effective therapeutic strategies. Claudin 18 isoform 2 (CLDN18.2) is a tight-junction protein selectively expressed in normal gastric mucosa that becomes aberrantly exposed on the tumor cell surface following disruption of cell polarity during malignant transformation.3 Approximately 30%-50% of gastric cancers are CLDN18.2-positive, defined as moderate-to-strong membranous CLDN18 staining in ≥75% of tumor cells by immunohistochemistry.4 Zolbetuximab is a chimeric immunoglobulin G1 monoclonal antibody targeting CLDN18.2 that mediates antitumor activity primarily through antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC).3 Two pivotal phase III trials established zolbetuximab as a first-line standard of care for patients with CLDN18.2-positive, human epidermal growth factor receptor 2 (HER2)-negative advanced gastric or GEJ adenocarcinoma.5, 6, 7, 8, 9 The SPOTLIGHT trial (zolbetuximab + FOLFOX versus placebo + FOLFOX) demonstrated substantial improvements in progression-free survival [PFS 10.6 versus 8.7 months; hazard ratio (HR) 0.75, P = 0.0066] and overall survival (OS) (18.2 versus 15.5 months; HR 0.75, P = 0.0053).5,6 Similarly, the GLOW trial (zolbetuximab + CapeOX versus placebo + CapeOX) showed improvements in PFS (8.2 versus 6.8 months; HR 0.68; P = 0.0007) and OS (14.4 versus 12.2 months; HR 0.77; P = 0.0118).7 Based on these results, zolbetuximab was approved in Japan in March 2024. Despite the clinical benefit of first-line zolbetuximab-based therapy, most patients ultimately develop progressive disease and the efficacy of currently available second-line treatments remains limited. The RAINBOW trial established ramucirumab plus weekly paclitaxel as a standard second-line treatment (OS 9.6 versus 7.4 months; HR 0.807).10 The ABSOLUTE trial demonstrated the noninferiority of weekly nab-paclitaxel (100 mg/m2) to paclitaxel in the second-line setting (OS 11.1 versus 10.9 months; HR 0.97),11 and a subsequent phase II study confirmed encouraging activity with nab-paclitaxel plus ramucirumab, with an objective response rate (ORR) of 54.8% and median PFS of 7.6 months.12 Additional prospective and retrospective studies suggest that nab-paclitaxel plus ramucirumab has efficacy comparable to that of paclitaxel plus ramucirumab.13, 14, 15 However, prospective evidence for second-line treatment after a first-line zolbetuximab-based regimen is lacking and dedicated clinical investigation is warranted to define the optimal second-line approach for this population, including whether CLDN18.2-targeted therapy should be continued beyond progression. The rationale for continuing zolbetuximab beyond progression is supported by two key biological considerations. Firstly, CLDN18.2 expression appears to be maintained after chemotherapy in patients with CLDN18.2-positive gastric cancer.16 Particularly, CLDN18.2 expression may reappear at disease progression after first-line treatment including zolbetuximab, in contrast to the well-documented attenuation of HER2 expression following trastuzumab treatment.17 Secondly, zolbetuximab-mediated ADCC and CDC are mechanistically distinct from classical inhibition of intracellular signaling pathways18, 19, 20; therefore, established resistance mechanisms, such as gatekeeper mutations or bypass pathway activation, may be less applicable.21 Furthermore, preclinical evidence suggests that taxane-induced immunogenic cell death and vascular endothelial growth factor receptor 2 blockade may complement zolbetuximab-mediated immune activation through remodeling of the tumor microenvironment,22 providing a biologically plausible basis for therapeutic synergy. Collectively, these observations provide compelling biological plausibility for sustaining CLDN18.2-targeted therapy in the second-line setting. Nonetheless, clinical evidence regarding CLDN18.2 dynamics after zolbetuximab exposure remains limited. Antigen downregulation may constitute a potential resistance mechanism and the extent of intratumoral and temporal heterogeneity in CLDN18.2 expression has not been fully elucidated. Given the practical difficulty of obtaining tumor tissue in the second-line setting, the Randomized Phase II trial Evaluating the Efficacy of ZolbEtuximab in the second-line for CLaudin18.2 positive advanceD GAstric/EGJ Cancer Patients Previously Treated with Zolbetuximab and Platinum-Based Chemotherapy (ZELDA) trial did not mandate confirmation of CLDN18.2 positivity at the time of enrollment. These biological uncertainties warrant further investigation and are prospectively addressed through integrated translational research (see ‘Translational research’ section). Objective The primary objective of ZELDA is to explore whether the addition of zolbetuximab to the standard second-line therapy, nab-paclitaxel plus ramucirumab, improves OS in patients with CLDN18.2-positive, HER2-negative, unresectable advanced/recurrent gastric or GEJ adenocarcinoma who are refractory or intolerant to first-line zolbetuximab, fluoropyrimidine, and platinum-based chemotherapy (Table 1).Table 1Specifications tableSubject areaMedical oncologyMore specific subject areaGastric cancer: systemic chemotherapyName of your trial in progressRandomized phase II trial of zolbetuximab as second-line treatment in claudin18.2-positive advanced gastric/GEJ cancer patients with previous zolbetuximab exposure: ZELDA trialReagents/toolsInvestigational drug: zolbetuximab (anti-CLDN18.2 monoclonal antibody, 400 mg/m2, i.v., days 1 and 15 of each 28-day cycle), nab-paclitaxel (100 mg/m2, days 1, 8, and 15), and ramucirumab (8 mg/kg, days 1 and 15).Standard-of-care drugs: nab-paclitaxel (100 mg/m2, days 1, 8, and 15) and ramucirumab (8 mg/kg, days 1 and 15).Trial designZELDA is a multicenter, open-label, randomized phase II trial with a safety lead-in part (n = 6) followed by a randomized part (n = 210), allocating patients 1 : 1 to nab-paclitaxel + ramucirumab (arm A, control) versus zolbetuximab + nab-paclitaxel + ramucirumab (arm B, experimental). Stratification factors include Eastern Cooperative Oncology Group performance status (ECOG PS; 0 versus 1), primary tumor site (stomach versus GEJ), and prior gastrectomy. The primary endpoint is overall survival in the full analysis set.Trial registrationjRCTs071250038EthicsThe trial protocol was approved by the Certified Review Board (Kyushu University Certified Institutional Review Board for Clinical Trials) and will be conducted in accordance with the Declaration of Helsinki and the Japanese Clinical Trials Act. Written informed consent will be obtained from all participants before enrollment.Value of the trial in progress•Testing the beyond progression concept for CLDN18.2-targeted therapy: This is the first prospective randomized trial to evaluate whether sustained CLDN18.2 inhibition after first-line treatment failure confers a survival benefit, thereby addressing an unmet clinical need.•Integrated translational research: ZELDA-TR prospectively evaluates longitudinal changes in CLDN18.2 expression, CLDN18–ARHGAP fusions, and plasma biomarkers, including cell-free DNA and proteomic profiles, to clarify resistance mechanisms and identify predictive biomarkers. Table 1Specifications tableSubject areaMedical oncologyMore specific subject areaGastric cancer: systemic chemotherapyName of your trial in progressRandomized phase II trial of zolbetuximab as second-line treatment in claudin18.2-positive advanced gastric/GEJ cancer patients with previous zolbetuximab exposure: ZELDA trialReagents/toolsInvestigational drug: zolbetuximab (anti-CLDN18.2 monoclonal antibody, 400 mg/m2, i.v., days 1 and 15 of each 28-day cycle), nab-paclitaxel (100 mg/m2, days 1, 8, and 15), and ramucirumab (8 mg/kg, days 1 and 15).Standard-of-care drugs: nab-paclitaxel (100 mg/m2, days 1, 8, and 15) and ramucirumab (8 mg/kg, days 1 and 15).Trial designZELDA is a multicenter, open-label, randomized phase II trial with a safety lead-in part (n = 6) followed by a randomized part (n = 210), allocating patients 1 : 1 to nab-paclitaxel + ramucirumab (arm A, control) versus zolbetuximab + nab-paclitaxel + ramucirumab (arm B, experimental). Stratification factors include Eastern Cooperative Oncology Group performance status (ECOG PS; 0 versus 1), primary tumor site (stomach versus GEJ), and prior gastrectomy. The primary endpoint is overall survival in the full analysis set.Trial registrationjRCTs071250038EthicsThe trial protocol was approved by the Certified Review Board (Kyushu University Certified Institutional Review Board for Clinical Trials) and will be conducted in accordance with the Declaration of Helsinki and the Japanese Clinical Trials Act. Written informed consent will be obtained from all participants before enrollment.Value of the trial in progress•Testing the beyond progression concept for CLDN18.2-targeted therapy: This is the first prospective randomized trial to evaluate whether sustained CLDN18.2 inhibition after first-line treatment failure confers a survival benefit, thereby addressing an unmet clinical need.•Integrated translational research: ZELDA-TR prospectively evaluates longitudinal changes in CLDN18.2 expression, CLDN18–ARHGAP fusions, and plasma biomarkers, including cell-free DNA and proteomic profiles, to clarify resistance mechanisms and identify predictive biomarkers. Specifications table •Testing the beyond progression concept for CLDN18.2-targeted therapy: This is the first prospective randomized trial to evaluate whether sustained CLDN18.2 inhibition after first-line treatment failure confers a survival benefit, thereby addressing an unmet clinical need. Testing the beyond progression concept for CLDN18.2-targeted therapy: This is the first prospective randomized trial to evaluate whether sustained CLDN18.2 inhibition after first-line treatment failure confers a survival benefit, thereby addressing an unmet clinical need. •Integrated translational research: ZELDA-TR prospectively evaluates longitudinal changes in CLDN18.2 expression, CLDN18–ARHGAP fusions, and plasma biomarkers, including cell-free DNA and proteomic profiles, to clarify resistance mechanisms and identify predictive biomarkers. Integrated translational research: ZELDA-TR prospectively evaluates longitudinal changes in CLDN18.2 expression, CLDN18–ARHGAP fusions, and plasma biomarkers, including cell-free DNA and proteomic profiles, to clarify resistance mechanisms and identify predictive biomarkers. Study design The ZELDA trial, which will evaluate zolbetuximab as second-line treatment for patients with CLDN18.2-positive advanced gastric or gastroesophageal cancer, is an investigator-initiated, multicenter, open-label, randomized phase II (rP2) trial. The trial is designed to evaluate the efficacy and safety of adding zolbetuximab to nab-paclitaxel plus ramucirumab in patients with CLDN18.2-positive, unresectable, advanced or recurrent gastric or GEJ adenocarcinoma previously treated with an anti-CLDN18.2 antibody and platinum-based chemotherapy. The trial is sponsored by the West Japan Oncology Group and the Kyushu Study Group of Clinical Cancer. The ZELDA trial consists of two sequential parts: a safety lead-in (SLI) part and a rP2 part. Zolbetuximab was previously evaluated in combination with nab-paclitaxel in a pancreatic cancer trial, in which the combination showed an acceptable safety profile.23 However, the triplet regimen of zolbetuximab, nab-paclitaxel, and ramucirumab has not yet been investigated. Therefore, a SLI part has been incorporated into the trial design to confirm the tolerability of this novel combination before proceeding to the rP2 part. SLI part Six patients receive the experimental combination of zolbetuximab, nab-paclitaxel, and ramucirumab. Tolerability is evaluated according to a prespecified stopping rule: the trial proceeds to the rP2 part if ≤2 of 6 patients experience a dose-limiting toxicity (DLT). If ≥3 DLTs are observed, the study will be halted. Although a formal contingency plan was not prespecified beyond the DLT-based stopping rule, the overall safety profile will be comprehensively reviewed before proceeding to the rP2 part and protocol amendments including dose modifications may be implemented as needed. rP2 part Upon completion of the SLI part, eligible patients in the rP2 part will be randomly assigned in a 1 : 1 ratio to arm A (control) or arm B (experimental). Randomization will be carried out using a minimization method via electronic data capture, stratified by Eastern Cooperative Oncology Group performance status (ECOG PS; 0 versus 1), primary tumor site (stomach versus GEJ), and prior gastrectomy (yes versus no). A total of 210 patients, with 105 patients per arm, is planned to be enrolled. Patients In the SLI and rP2 parts, eligible patients must have histologically confirmed unresectable advanced or recurrent adenocarcinoma of the stomach or GEJ. Tumors must be CLDN18.2-positive, defined as moderate-to-strong membranous staining in at least 75% of tumor cells, as determined by CLDN18 immunohistochemistry using the VENTANA CLDN18 [43-14A] RxDx Assay (Roche Diagnostic Solutions, Tucson, AZ). Patients must have experienced disease progression during or after first-line combination chemotherapy consisting of a fluoropyrimidine, platinum agent, and zolbetuximab, or must have been intolerant to agents in the regimen other than zolbetuximab. In the rP2 part, prior treatment with ASP2138,24 a bispecific antibody targeting CLDN18.2 and CD3, is also permitted in place of zolbetuximab. Additional key eligibility criteria include an ECOG PS of 0 or 1, age of at least 18 years, measurable or evaluable disease according to RECIST version 1.1,25 an estimated life expectancy of at least 3 months, adequate organ function within 14 days before enrollment, negative hepatitis B surface antigen (HBsAg) status or receipt of nucleotide analog therapy in accordance with guidelines, and provision of written informed consent. The detailed inclusion and exclusion criteria are summarized in Table 2.Table 2Inclusion and exclusion criteriaInclusion criteriaaHistologically confirmed adenocarcinoma of the stomach or esophagogastric junction that is unresectable advanced or recurrent (including cases in which adenocarcinoma comprises ≥50% of the tumor).bCLDN18.2-positive, defined as moderate-to-strong membranous staining in ≥75% of tumor cells by CLDN18 immunohistochemistry using the VENTANA CLDN18 (43-14A) RxDx Assay (Roche Diagnostic Solutions; Tucson, AZ). Re-evaluation of CLDN18.2 status at recurrence or progression on first-line treatment is not mandatory.cRefractory to first-line combination chemotherapy consisting of a fluoropyrimidine, a platinum agent, and zolbetuximab, or intolerant to agents other than zolbetuximab in the regimen. In the randomized phase II part, prior treatment with ASP2138 is also permitted in place of zolbetuximab.dAt least one measurable or evaluable lesion per RECIST version 1.1.eAge ≥18 years.fECOG PS of 0 or 1.gEstimated life expectancy of ≥3 months from the date of enrollment.hAdequate organ function within 14 days before enrollment (results from the same day of the week 2 weeks prior are acceptable):i.Neutrophils ≥1500/mm3ii.Hemoglobin ≥8.0 g/dliii.Platelets ≥10.0 × 104/mm3iv.Total bilirubin ≤1.5 mg/dlv.AST ≤120 U/l (≤200 U/l with liver metastases)vi.ALT ≤120 U/l (≤200 U/l with liver metastases)vii.Creatinine ≤1.5 mg/dl or CrCl ≥30 ml/minviii.Urinary protein ≤1+ (if ≥2+: 24-h urine protein <2 g or UPC <2)iHBsAg-negative. If HBsAg-positive, the patient must be receiving nucleoside/nucleotide analog therapy per hepatitis B guidelines. Hepatologist consultation is preferable.jWomen of childbearing potential: agreement to highly effective contraception before treatment.Men: surgically sterile or agreement to highly effective contraception during treatment and for 6 months after end of treatment.kWritten informed consent obtained from the patient after full explanation of the trial, before enrollment.Exclusion criteriaaActive double cancer: synchronous or metachronous with disease-free interval ≤3 years. Exceptions: carcinoma in situ cured by local treatment, intramucosal carcinoma, nonmetastatic prostate cancer not requiring systemic treatment, or other solid cancers not requiring treatment and not expected to affect prognosis.bPrior treatment with any antiangiogenic agent.cPrior taxane therapy for an incurable condition.Exception: perioperative taxane with last dose ≥24 weeks before enrollment.dIntolerance to zolbetuximab.eHER2-positive tumor.fComplication of uncontrolled diabetes mellitus.gDeep vein thrombosis, pulmonary embolism, or other substantial thromboembolism within 3 months before enrollment (port/catheter thrombosis and superficial vein thrombosis are not considered substantial).hGrade 3-4 hemorrhagic disease, vasculitis, or gastrointestinal hemorrhage within 3 months before enrollment.iMyocardial infarction, unstable angina, cerebrovascular accident, or transient ischemic attack within 6 months before enrollment.jGastrointestinal perforation/fistula within 6 months before enrollment or high risk of perforation.kMajor surgery, serious unhealed wounds/injuries, peptic ulcer, or bone fracture within 4 weeks before enrollment. Minor procedures (lymph node biopsy, needle biopsy, port placement) are excluded from this criterion.lBlood transfusion within 2 weeks before enrollment.mSymptomatic heart failure (NYHA class II-IV) or symptomatic/uncontrolled arrhythmia.nUncontrolled hypertension despite antihypertensives: systolic blood pressure >160 mmHg or diastolic blood pressure >100 mmHg sustained for ≥4 weeks.oContinuous systemic steroid therapy.pComplication of hepatic cirrhosis.qSymptomatic brain metastases or meningeal dissemination.rActive infection.sPericardial effusion, pleural effusion, or ascites requiring frequent drainage.tIntolerance, severe allergy, or anaphylactic reaction to antibody preparations or human serum albumin.uInterstitial lung disease on CT. Exception: radiation-induced fibrosis within the irradiation field.vPsychosis or psychiatric symptoms judged to preclude trial participation.wPregnant, possibly pregnant, or lactating female patients.xDeemed ineligible by the attending physician for any other reason.ALT, alanine aminotransferase; AST, aspartate aminotransferase; CLDN, claudin; CT, computed tomography; ECOG PS, Eastern Cooperative Oncology Group performance status; HBsAg, hepatitis B surface antigen; HER2, human epidermal growth factor receptor 2; NYHA, New York Heart Association; UPC, urine protein-to-creatinine ratio. Table 2Inclusion and exclusion criteriaInclusion criteriaaHistologically confirmed adenocarcinoma of the stomach or esophagogastric junction that is unresectable advanced or recurrent (including cases in which adenocarcinoma comprises ≥50% of the tumor).bCLDN18.2-positive, defined as moderate-to-strong membranous staining in ≥75% of tumor cells by CLDN18 immunohistochemistry using the VENTANA CLDN18 (43-14A) RxDx Assay (Roche Diagnostic Solutions; Tucson, AZ). Re-evaluation of CLDN18.2 status at recurrence or progression on first-line treatment is not mandatory.cRefractory to first-line combination chemotherapy consisting of a fluoropyrimidine, a platinum agent, and zolbetuximab, or intolerant to agents other than zolbetuximab in the regimen. In the randomized phase II part, prior treatment with ASP2138 is also permitted in place of zolbetuximab.dAt least one measurable or evaluable lesion per RECIST version 1.1.eAge ≥18 years.fECOG PS of 0 or 1.gEstimated life expectancy of ≥3 months from the date of enrollment.hAdequate organ function within 14 days before enrollment (results from the same day of the week 2 weeks prior are acceptable):i.Neutrophils ≥1500/mm3ii.Hemoglobin ≥8.0 g/dliii.Platelets ≥10.0 × 104/mm3iv.Total bilirubin ≤1.5 mg/dlv.AST ≤120 U/l (≤200 U/l with liver metastases)vi.ALT ≤120 U/l (≤200 U/l with liver metastases)vii.Creatinine ≤1.5 mg/dl or CrCl ≥30 ml/minviii.Urinary protein ≤1+ (if ≥2+: 24-h urine protein <2 g or UPC <2)iHBsAg-negative. If HBsAg-positive, the patient must be receiving nucleoside/nucleotide analog therapy per hepatitis B guidelines. Hepatologist consultation is preferable.jWomen of childbearing potential: agreement to highly effective contraception before treatment.Men: surgically sterile or agreement to highly effective contraception during treatment and for 6 months after end of treatment.kWritten informed consent obtained from the patient after full explanation of the trial, before enrollment.Exclusion criteriaaActive double cancer: synchronous or metachronous with disease-free interval ≤3 years. Exceptions: carcinoma in situ cured by local treatment, intramucosal carcinoma, nonmetastatic prostate cancer not requiring systemic treatment, or other solid cancers not requiring treatment and not expected to affect prognosis.bPrior treatment with any antiangiogenic agent.cPrior taxane therapy for an incurable condition.Exception: perioperative taxane with last dose ≥24 weeks before enrollment.dIntolerance to zolbetuximab.eHER2-positive tumor.fComplication of uncontrolled diabetes mellitus.gDeep vein thrombosis, pulmonary embolism, or other substantial thromboembolism within 3 months before enrollment (port/catheter thrombosis and superficial vein thrombosis are not considered substantial).hGrade 3-4 hemorrhagic disease, vasculitis, or gastrointestinal hemorrhage within 3 months before enrollment.iMyocardial infarction, unstable angina, cerebrovascular accident, or transient ischemic attack within 6 months before enrollment.jGastrointestinal perforation/fistula within 6 months before enrollment or high risk of perforation.kMajor surgery, serious unhealed wounds/injuries, peptic ulcer, or bone fracture within 4 weeks before enrollment. Minor procedures (lymph node biopsy, needle biopsy, port placement) are excluded from this criterion.lBlood transfusion within 2 weeks before enrollment.mSymptomatic heart failure (NYHA class II-IV) or symptomatic/uncontrolled arrhythmia.nUncontrolled hypertension despite antihypertensives: systolic blood pressure >160 mmHg or diastolic blood pressure >100 mmHg sustained for ≥4 weeks.oContinuous systemic steroid therapy.pComplication of hepatic cirrhosis.qSymptomatic brain metastases or meningeal dissemination.rActive infection.sPericardial effusion, pleural effusion, or ascites requiring frequent drainage.tIntolerance, severe allergy, or anaphylactic reaction to antibody preparations or human serum albumin.uInterstitial lung disease on CT. Exception: radiation-induced fibrosis within the irradiation field.vPsychosis or psychiatric symptoms judged to preclude trial participation.wPregnant, possibly pregnant, or lactating female patients.xDeemed ineligible by the attending physician for any other reason.ALT, alanine aminotransferase; AST, aspartate aminotransferase; CLDN, claudin; CT, computed tomography; ECOG PS, Eastern Cooperative Oncology Group performance status; HBsAg, hepatitis B surface antigen; HER2, human epidermal growth factor receptor 2; NYHA, New York Heart Association; UPC, urine protein-to-creatinine ratio. Inclusion and exclusion criteria i.Neutrophils ≥1500/mm3 Neutrophils ≥1500/mm3 ii.Hemoglobin ≥8.0 g/dl Hemoglobin ≥8.0 g/dl iii.Platelets ≥10.0 × 104/mm3 Platelets ≥10.0 × 104/mm3 iv.Total bilirubin ≤1.5 mg/dl Total bilirubin ≤1.5 mg/dl v.AST ≤120 U/l (≤200 U/l with liver metastases) AST ≤120 U/l (≤200 U/l with liver metastases) vi.ALT ≤120 U/l (≤200 U/l with liver metastases) ALT ≤120 U/l (≤200 U/l with liver metastases) vii.Creatinine ≤1.5 mg/dl or CrCl ≥30 ml/min Creatinine ≤1.5 mg/dl or CrCl ≥30 ml/min viii.Urinary protein ≤1+ (if ≥2+: 24-h urine protein <2 g or UPC <2) Urinary protein ≤1+ (if ≥2+: 24-h urine protein <2 g or UPC <2) ALT, alanine aminotransferase; AST, aspartate aminotransferase; CLDN, claudin; CT, computed tomography; ECOG PS, Eastern Cooperative Oncology Group performance status; HBsAg, hepatitis B surface antigen; HER2, human epidermal growth factor receptor 2; NYHA, New York Heart Association; UPC, urine protein-to-creatinine ratio. Treatment SLI part In the SLI part, the patients will receive zolbetuximab (400 mg/m2) intravenously every 2 weeks on days 1 and 15, in combination with standard-dose nab-paclitaxel (100 mg/m2) on days 1, 8, and 15 and ramucirumab (8 mg/kg) on days 1 and 15 in 28-day cycles (Table 3). Zolbetuximab is administered as an intravenous infusion over ≥2 h (and within <12 h) before nab-paclitaxel and ramucirumab. These doses correspond to the approved or commonly used doses in advanced gastric cancer.5,7,12 Because the combination of zolbetuximab with ramucirumab has not previously been evaluated, the SLI phase is included to confirm the tolerability of the triplet regimen before initiation of the rP2 part.Table 3Treatment regimens in the ZELDA randomized trialDrugDoseAdministrationScheduleArm A (control)Nab-paclitaxel + ramucirumab (q28d)Nab-paclitaxel100 mg/m2IV infusion over 30 minDays 1, 8, 15Ramucirumab8 mg/kgIV infusion over ≥30 minDays 1, 15Safety lead-in part/arm B (experimental)Zolbetuximab + Nab-paclitaxel + ramucirumab (q28d)Zolbetuximab400 mg/m2IV infusion over ≥2 h (before chemotherapy)Days 1, 15Nab-paclitaxel100 mg/m2IV infusion over 30 minDays 1, 8, 15Ramucirumab8 mg/kgIV infusion over ≥30 minDays 1, 15 Table 3Treatment regimens in the ZELDA randomized trialDrugDoseAdministrationScheduleArm A (control)Nab-paclitaxel + ramucirumab (q28d)Nab-paclitaxel100 mg/m2IV infusion over 30 minDays 1, 8, 15Ramucirumab8 mg/kgIV infusion over ≥30 minDays 1, 15Safety lead-in part/arm B (experimental)Zolbetuximab + Nab-paclitaxel + ramucirumab (q28d)Zolbetuximab400 mg/m2IV infusion over ≥2 h (before chemotherapy)Days 1, 15Nab-paclitaxel100 mg/m2IV infusion over 30 minDays 1, 8, 15Ramucirumab8 mg/kgIV infusion over ≥30 minDays 1, 15 Treatment regimens in the ZELDA randomized trial rP2 part Treatment will be administered in 28-day cycles and will continue until protocol-defined discontinuation criteria are met. In arm A (control), patients will receive nab-paclitaxel plus ramucirumab. Nab-paclitaxel must be administered on days 1, 8, and 15 and ramucirumab will be administered on days 1 and 15 (Table 3). In arm B (experimental), patients will receive nab-paclitaxel, ramucirumab, and zolbetuximab. Zolbetuximab will be administered as an intravenous infusion over ≥2 h and within <12 h before nab-paclitaxel and ramucirumab on days 1 and 15. Nab-paclitaxel will be administered on days 1, 8, and 15 and ramucirumab will be administered on days 1 and 15 (Table 3). Patients will continue treatment until radiographic disease progression according to RECIST version 1.1,25 unacceptable toxicity, patient or physician decision to discontinue treatment, or occurrence of protocol-specified treatment termination criteria. If any single agent is discontinued because of an adverse event, treatment with the remaining agents may be continued at the investigator’s discretion if continued treatment is considered clinically beneficial. Dose reductions of nab-paclitaxel and ramucirumab are permitted according to prespecified criteria. Dose reduction will not be permitted for zolbetuximab; instead, zolbetuximab will be withheld or permanently discontinued according to the adverse event. Study endpoint and outcome measures The primary endpoint of the SLI part is tolerability, defined as the incidence of DLTs during cycle 1 (days 1-28) (Table 4). DLT is defined as protocol-specified hematological or nonhematological toxicities of grade ≥ 3 evaluated as possibly related to zolbetuximab, including grade 4 neutropenia persisting ≥7 days, febrile neutropenia, grade ≥3 thrombocytopenia with clinically substantial bleeding or grade 4 thrombocytopenia and any treatment-related adverse event requiring a dose delay of >14 days. The DLT evaluation period extends from the first dose to day 1 of cycle 2 (28 days) or up to day 15 of cycle 2 (42 days) if the criteria for starting cycle 2 are not met on cycle 2 day 1.Table 4Primary and secondary endpoints by study partEndpointSLI partrP2 partPrimary endpointTolerability: incidence of DLTs during cycle 1OS in the FASSecondary endpoints—PFSTTFORRDCRDoRDpRETSIncidence and severity of adverse events (NCI-CTCAE v5.0)Patient-reported QOL (EORTC QLQ-C30, QLQ-STO22)DCR, disease control rate; DLT, dose-limiting toxicity; DoR, duration of response; DpR, deepness of response; EORTC, European Organization for Research and Treatment of Cancer; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ETS, early tumor shrinkage; FAS, full analysis set; NCI-CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; QLQ, quality of life; QLQ-STO22, Gastric Cancer-Specific Supplementary Module; rP2, randomized phase II; SLI, safety lead-in; TTF, time to treatment failure. Table 4Primary and secondary endpoints by study partEndpointSLI partrP2 partPrimary endpointTolerability: incidence of DLTs during cycle 1OS in the FASSecondary endpoints—PFSTTFORRDCRDoRDpRETSIncidence and severity of adverse events (NCI-CTCAE v5.0)Patient-reported QOL (EORTC QLQ-C30, QLQ-STO22)DCR, disease control rate; DLT, dose-limiting toxicity; DoR, duration of response; DpR, deepness of response; EORTC, European Organization for Research and Treatment of Cancer; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ETS, early tumor shrinkage; FAS, full analysis set; NCI-CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; QLQ, quality of life; QLQ-STO22, Gastric Cancer-Specific Supplementary Module; rP2, randomized phase II; SLI, safety lead-in; TTF, time to treatment failure. Primary and secondary endpoints by study part DCR, disease control rate; DLT, dose-limiting toxicity; DoR, duration of response; DpR, deepness of response; EORTC, European Organization for Research and Treatment of Cancer; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ETS, early tumor shrinkage; FAS, full analysis set; NCI-CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; QLQ, quality of life; QLQ-STO22, Gastric Cancer-Specific Supplementary Module; rP2, randomized phase II; SLI, safety lead-in; TTF, time to treatment failure. The primary endpoint of the rP2 part is OS in the full analysis set, defined as the time from randomization to death from any cause (Table 4). OS was selected as the primary endpoint as PFS has not been established as a surrogate for OS in advanced gastric cancer. A meta-analysis demonstrated only a moderate correlation between treatment effects on PFS and OS (R2 = 0.61) and could not confirm PFS as a valid surrogate.26 This limitation is particularly relevant in the second-line setting, where PFS has been shown to correlate insufficiently with OS.27 Secondary endpoints include OS in the intention-to-treat set and per-protocol set; PFS, defined as the time from randomization to the first occurrence of radiographic disease progression per RECIST version 1.1 or death from any cause25; time to treatment failure; ORR; disease control rate; duration of response; deepness of response; early tumor shrinkage; and incidence and severity of adverse events assessed per Common Terminology Criteria for Adverse Events version 5.0; and patient-reported quality of life. Disease assessment will be carried out using contrast-enhanced computed tomography (CT) of the thoracoabdominal pelvic region. Plain CT or magnetic resonance imaging may be substituted for patients with a history of contrast media allergy and the same imaging modality must be used throughout the trial. A baseline scan will be obtained within 28 days before enrollment, followed by imaging every 8 weeks. Responses will be evaluated using RECIST version 1.1. Patients who discontinue study treatment for reasons other than disease progression will continue to undergo disease assessments every 8 weeks until disease progression or withdrawal of consent, including after the initiation of subsequent therapy. Safety will be monitored throughout the trial. The incidence, severity, causality, and outcomes of adverse events will be assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0. Complete blood counts and blood chemistry assessments will be carried out on days 1, 8, and 15 of each 28-day cycle, and urinary protein will be assessed on days 1 and 15. The Efficacy and Safety Evaluation Committee reviews the accumulation of safety data at prespecified intervals. Patient-reported quality of life will be assessed using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 and the Gastric Cancer-Specific Supplementary Module at baseline, 1 month after treatment initiation, 3 months after treatment initiation, and at treatment discontinuation. For patients who consent to participate in the translational research component, ZELDA-TR, blood samples will be collected at the start and completion of protocol treatment. Tumor tissue samples, including archival biopsy specimens obtained before treatment and optional fresh biopsy specimens collected at treatment discontinuation, will also be obtained according to the assessment schedule. Planned sample size The ‘Safety lead-in’ section will include six patients. This sample size was not based on formal hypothesis testing but was selected in accordance with the conventional six rolling design widely used in early-phase oncology trials.28,29 Tolerability will be evaluated according to a prespecified stopping rule. The trial will proceed to the rP2 part if ≤2 of 6 patients experience a DLT. If ≥3 DLTs are observed, the trial will be halted. For the rP2 trial, the planned sample size is 210 patients. This sample size will provide 80% power to detect an OS HR of 0.77 at a one-sided significance level of 20%, assuming ∼190 OS events at the final analysis. The target HR was derived from the OS benefit observed in the first-line SPOTLIGHT (HR 0.75)5,6 and GLOW (HR 0.77) trials.7 A one-sided significance level of 20% was adopted in accordance with the randomized screening design for phase II trials, as this study is designed as a signal-seeking trial to inform the decision regarding a subsequent confirmatory phase III study. The median OS in the control arm was estimated to be 10.0 months, based on results from the RAINBOW trial of paclitaxel plus ramucirumab.10 Study duration The trial was initiated in August 2025, with the SLI phase starting at seven institutions. After tolerability is confirmed in the SLI part, enrollment in the rP2 will commence. As of March 2026, 51 institutions across Japan are participating in the trial. The planned enrollment period is 2 years, followed by a 2-year follow-up period; therefore, the total trial duration is estimated to be 5 years. Translational research To explore biomarkers for the prediction of efficacy and treatment selection, as well as to identify resistance mechanisms, a concurrent prospective translational trial (ZELDA-TR; UMIN000059075) will be conducted with consenting participants from the ZELDA trial (Table 5). The target enrollment is ∼170 patients, corresponding to ∼80% of rP2 trials. Tumor tissues and blood samples will be collected prospectively at predefined time points during the trial. Tissue-based biomarker analyses will include immunohistochemical assessments of CLDN18.2 and caveolin-1, including intratumoral heterogeneity and longitudinal changes, as well as next-generation sequencing for CLDN18–ARHGAP fusions, somatic mutation profiling, and gene expression profiling. Blood-based biomarker analyses will include plasma cell-free DNA assessment of circulating tumor DNA dynamics, genomic alterations, and DNA methylation profiles as well as plasma proteomic profiling and enzyme-linked immunosorbent assay-based measurements of plasma CLDN18 levels. Biomarker data and clinical outcomes will be analyzed to examine the associations between pretreatment CLDN18.2 expression and treatment efficacy, to evaluate longitudinal biomarker changes during treatment, and to explore potential mechanisms of resistance. The primary endpoint of ZELDA-TR is the association between pretreatment CLDN18.2 expression and treatment efficacy. Secondary endpoints include the associations of other tissue- and plasma-based biomarkers with treatment outcomes and exploration of resistance mechanisms.Table 5Biomarker sample collection schedule for ZELDA-TRSampleBiomarkerPretreatmentOn-treatment (every 8 weeks)Post-treatmentBloodMRD/methylation○○○Gene alteration○○Proteome (Orbitrap)○○CLDN18.2 (ELISA)○○TissueCLDN18 (IHC)○a○CAV-1 (IHC)○a○CLDN18–ARHGAP fusion (NGS)○a○CAV-1, caveolin-1; CLDN18.2, claudin 18.2; ELISA, enzyme-linked immunosorbent assay; IHC, immunohistochemistry.aIncluding archival. Table 5Biomarker sample collection schedule for ZELDA-TRSampleBiomarkerPretreatmentOn-treatment (every 8 weeks)Post-treatmentBloodMRD/methylation○○○Gene alteration○○Proteome (Orbitrap)○○CLDN18.2 (ELISA)○○TissueCLDN18 (IHC)○a○CAV-1 (IHC)○a○CLDN18–ARHGAP fusion (NGS)○a○CAV-1, caveolin-1; CLDN18.2, claudin 18.2; ELISA, enzyme-linked immunosorbent assay; IHC, immunohistochemistry.aIncluding archival. Biomarker sample collection schedule for ZELDA-TR CAV-1, caveolin-1; CLDN18.2, claudin 18.2; ELISA, enzyme-linked immunosorbent assay; IHC, immunohistochemistry. Including archival.