DESTINY-Gastric05 phase III trial of first-line trastuzumab deruxtecan, chemotherapy, and pembrolizumab in HER2-positive gastric or gastroesophageal junction cancer.

DESCRIPTION OF PROTOCOL

Background
Gastric or gastroesophageal junction (GEJ) cancer is the fifth most frequently diagnosed cancer and cause of cancer deaths, globally.1 In advanced (unresectable or metastatic) stages of gastric or GEJ cancer, the survival rate is poor, with a 5-year survival rate of ∼5%.2 Human epidermal growth factor receptor 2 (HER2) is a validated target in up to 17% of patients with gastric or GEJ cancer and trastuzumab is the only first-line HER2-directed treatment option available for patients with HER2-positive tumors.3, 4, 5, 6, 7
The recommended first-line treatment option for patients with HER2-positive [immunohistochemistry (IHC) 3+ or IHC 2+/FISH positive] unresectable locally advanced or metastatic gastric or GEJ adenocarcinoma is combination chemotherapy and trastuzumab, with pembrolizumab for programmed death-ligand 1 (PD-L1) combined positive score (CPS) ≥1.6,8,9 The trastuzumab plus chemotherapy treatment combination reflects the regimen used in the ToGA trial, which demonstrated improved overall survival (OS) in patients with HER2-positive advanced gastric or GEJ cancer.10 The addition of pembrolizumab [an anti-programmed cell death protein 1 (PD-1) monoclonal antibody] was approved in 2021 for patients whose tumors are also PD-L1 positive (CPS ≥1), based on results from the phase III KEYNOTE-811 trial (NCT03615326), which showed that the addition of pembrolizumab to first-line platinum-based chemotherapy and trastuzumab improved progression-free survival (PFS) and OS in patients with HER2-positive gastric or GEJ cancer with a PD-L1 CPS ≥1.11, 12, 13, 14
Despite improved survival demonstrated with current standard of care, more effective HER2-directed first-line treatment options are still needed for patients with HER2-positive gastric or GEJ cancer regardless of PD-L1 expression. Standard-of-care platinum chemotherapy is associated with toxicities including neutropenia, anemia, nausea and vomiting, diarrhea, neurotoxicity, and nephrotoxicity,15 which may limit continued therapy. Other HER2-directed therapies, with different mechanisms of action, are in development in this setting and have shown promising preliminary efficacy.16, 17, 18, 19
Trastuzumab deruxtecan (T-DXd) is an antibody–drug conjugate consisting of a humanized anti-HER2 monoclonal antibody linked to a topoisomerase I inhibitor payload by means of a tetrapeptide-based cleavable linker.20 Based on results from phase II and III trials, T-DXd is approved as monotherapy in the second-line or later setting in >65 countries worldwide for the treatment of patients with advanced or metastatic HER2-positive [IHC 3+ or IHC 2+/in situ hybridization (ISH) positive] gastric or GEJ cancer who have received a prior trastuzumab-based regimen.3,21, 22, 23, 24 The phase III DESTINY-Gastric04 (NCT04704934) trial showed that treatment with T-DXd led to a significantly longer median OS than ramucirumab plus paclitaxel (14.7 months versus 11.4 months, respectively) in patients with HER2-positive metastatic gastric or GEJ cancer.3 T-DXd is reported to modulate the tumor microenvironment and enhance the activity of immunotherapy via immunostimulatory activity, increasing major histocompatibility complex class I expression (and cytotoxic activity) on tumor cells and up-regulating dendritic cell activation markers.25,26
In the first-line setting in DESTINY-Gastric03 (NCT04379596), a multicenter, open-label, dose-escalation/dose-expansion, phase Ib/II trial, combinations of T-DXd and fluoropyrimidine [5-fluorouracil (5-FU) or capecitabine] with or without pembrolizumab showed promising antitumor activity in patients with gastric or GEJ cancer, regardless of PD-L1 CPS.27, 28, 29 In patients who received T-DXd 5.4 mg/kg plus fluoropyrimidine and pembrolizumab in DESTINY-Gastric03, the confirmed objective response rate (ORR) by investigator assessment was 75.0% [95% confidence interval (CI) 56.6% to 88.5%], the median duration of response was 12.3 months (95% CI 3.9 months-not estimable), the median PFS was 12.1 months (95% CI 7.0 months-not estimable), the median OS was 15.7 months (95% CI 12.1 months-not estimable), and the safety profile was considered manageable. These results provide rationale for further investigation of T-DXd combined with a fluoropyrimidine and immunotherapy as a potentially improved, platinum-free, first-line treatment in HER2-positive gastric or GEJ cancer.

Objective
DESTINY-Gastric05 is a phase III trial to investigate first-line treatment with T-DXd 5.4 mg/kg plus chemotherapy, with or without pembrolizumab, versus trastuzumab plus platinum-based chemotherapy, with or without pembrolizumab, in patients with HER2-positive metastatic gastric or GEJ cancer (Table 1). The aim of DESTINY-Gastric05 is to bring potentially improved treatment outcomes with a T-DXd plus platinum-free chemotherapy treatment approach for all patients with HER2-positive gastric or GEJ cancer, regardless of PD-L1 expression.

Endpoints
The primary endpoint is PFS by blinded independent central review (BICR), defined as the time interval from the date of randomization to the date of radiographic disease progression or death due to any cause (Table 2).
The key secondary endpoint in DESTINY-Gastric05 is OS. Additional prespecified secondary endpoints include ORR, PFS by investigator assessment [per Response Evaluation Criteria in Solid Tumours, version 1.1 (RECIST v1.1)], duration of response, time to response by BICR (per RECIST v1.1), time to second PFS event (time from randomization to progression on next line of therapy or death due to any cause), treatment-emergent adverse events (TEAEs) and other safety parameters, and patient-reported outcomes [PROs; Functional Assessment of Cancer Therapy—Gastric (FACT-GA) subscale, FACT-GA Physical Well-being subscale, and EuroQol 5-dimension, 5-level (EQ-5D-5L) visual analogue scale]. Exploratory endpoints include health economics and outcomes research, hospital admissions, other PROs, and correlation of PD-L1 expression levels and other biomarkers with efficacy and safety.
Exploratory analysis of relevant biomarkers will be carried out to investigate the mechanism of action of T-DXd or the correlation of biomarkers with response or safety parameters. Biomarker analyses may be used to observe the effect of treatment at the molecular and cellular levels in addition to determining how changes in biomarkers might relate to T-DXd exposure and clinical outcomes.

Trial design and statistical analyses
DESTINY-Gastric05 is an open-label, randomized, multicenter, phase III trial of T-DXd 5.4 mg/kg as first-line treatment for patients with metastatic HER2-positive gastric or GEJ cancer. The study will be conducted at 246 study sites in 26 countries across 4 regions (North America, South America, Europe, Asia/Pacific) (Figure 1). The trial commenced on 27 February 2025 and is estimated to be completed by 1 February 2030.
The trial will include four periods consisting of tissue prescreening, screening, treatment, and follow-up. The tissue prescreening period may occur any time before screening, and the screening period must be completed within 28 days of cycle 1, day 1. The treatment period will begin after randomization and continue until discontinuation of all study treatment. For all patients, follow-up safety and efficacy assessments will be collected starting from the end-of-treatment visit, which should occur within 7 days after the last dose of study treatment or at the time of discontinuation of all study treatment if this occurs >7 days after the last dose. Additionally, there will be a 40-day (+7 days) follow-up visit, which will occur after the last dose or before starting a new anticancer treatment, whichever comes first. Subsequent long-term follow-up visits will occur every 3 months (±14 days) and are planned to start 40 days (+7 days) after the last dose and through to the end of the trial.
The primary efficacy analysis will compare the PFS between treatment arms based on data from the full analysis set (FAS; including all patients randomly assigned to a treatment group according to the treatment to which they are randomly assigned). The primary PFS analysis will be carried out separately for each cohort, and the null hypothesis will be tested. PFS will be presented graphically using the Kaplan–Meier method, and median PFS and the two-sided 95% CIs will be calculated using the Brookmeyer and Crowley method.
The key secondary efficacy analysis will compare OS between the treatment arms based on data from the FAS. The same analysis conducted for PFS will be carried out for OS in each cohort.
The sample size for the main cohort has been designed to retain at least 90% and 80% power for the PFS and OS comparisons, respectively. It is hypothesized that treatment with T-DXd in combination with fluoropyrimidine and pembrolizumab will result in a 29% reduction in risk for PFS and a 24% reduction in risk for OS. Based on these assumptions, 406 PFS events and 421 OS events are needed to provide the necessary power for this analysis. In order to observe the target number of PFS and OS events, a total of 576 patients will need to be randomized. The sample size of the exploratory cohort has been determined based on feasibility, and ∼150 participants will be randomized in total.
To adjust for multiplicity of the overall type I error level within a cohort, a hierarchical testing procedure will be applied to address multiplicity between endpoints. Firstly, statistical testing will be conducted for PFS, and only if a statistically significant improvement in PFS is observed will statistical testing be conducted for OS. The type I error level for the interim and final analyses will be controlled using a Haybittle–Peto approach for PFS and a Lan-DeMets (O’Brien-Fleming) alpha spending function for OS.

Patients
Approximately 576 patients with PD-L1 CPS ≥1 will be randomly assigned to the main cohort, and ∼150 patients with PD-L1 CPS <1 will be randomly assigned to the exploratory cohort. Patients who have received prior treatment in the perioperative and/or adjuvant setting or have received treatment with immuno-oncology [i.e. anti-PD-(L)1] therapy in the (neo)adjuvant setting will be able to enroll provided there is >6 months between the end of perioperative or neoadjuvant treatment and the diagnosis of recurrent disease. Patients with prior exposure to other HER2-targeting therapies (including antibody–drug conjugates) will be excluded. Table 3 summarizes key inclusion and exclusion criteria.

Trial treatment
Patients will be randomly assigned to treatment arms using interactive response technology and stratified by HER2 status [IHC 3+ versus IHC 2+/ISH positive (as determined by central laboratory)] and by geographic region (Japan/Republic of Korea versus the rest of Asia versus North America/Europe/rest of the world). In the main cohort, patients with PD-L1 CPS ≥1 will be randomly assigned in a 1 : 1 ratio to receive either T-DXd in combination with 5-FU or capecitabine plus pembrolizumab (arm M1) or to receive trastuzumab in combination with platinum-based chemotherapy (either cisplatin and 5-FU or oxaliplatin and capecitabine) plus pembrolizumab (arm M2). The trial will also include an exploratory cohort of patients with PD-L1 CPS <1. Similar to the main cohort, patients in the exploratory cohort will be randomly assigned in a 1 : 1 ratio to receive either T-DXd with 5-FU or capecitabine (arm E1) or to receive trastuzumab with platinum-based chemotherapy (either cisplatin and 5-FU or oxaliplatin and capecitabine) (arm E2). Each treatment cycle will be 21 days.
Patients will continue to receive treatment until radiographic disease progression as assessed per RECIST v1.1 by the investigator and verified by BICR, clinical progression, unacceptable TEAEs, patient or physician decision, pregnancy, loss to follow-up, trial termination by the sponsor, or death.
T-DXd 5.4 mg/kg was chosen as the treatment dose in DESTINY-Gastric05 based on preliminary safety data from the DESTINY-Gastric03 trial. Previous results from the dose-escalation phase of DESTINY-Gastric03, which examined T-DXd plus 5-FU or capecitabine every 3 weeks, and results from the dose-expansion phase, which examined T-DXd 6.4 mg/kg or 5.4 mg/kg plus 5-FU or capecitabine plus pembrolizumab every 3 weeks, demonstrated improved tolerability for 5.4-mg/kg dosing over 6.4-mg/kg dosing when used in first-line combination therapy. Additionally, time-matched analysis from DESTINY-Gastric03 showed that lowering the starting doses of T-DXd and capecitabine improved tolerability of the triplet combination of T-DXd with fluoropyrimidine and pembrolizumab without decreasing the ORR.28,29
Patients randomly assigned to arm M1 will be treated with T-DXd 5.4 mg/kg intravenously every 3 weeks on day 1 with either 5-FU 600 mg/m2/day intravenously on days 1-5 or capecitabine 750 mg/m2 orally twice daily on days 1-14 and in combination with pembrolizumab 200 mg intravenously every 3 weeks on day 1. Patients randomly assigned to arm M2 will be treated with trastuzumab, loading dose of 8 mg/kg intravenously followed by 6 mg/kg intravenously every 3 weeks, in combination with platinum-based chemotherapy (cisplatin 80 mg/m2/day intravenously on day 1 plus 5-FU 800 mg/m2/day intravenously on days 1-5 or oxaliplatin 130 mg/m2/day intravenously on day 1 plus capecitabine 1000 mg/m2 orally twice daily on days 1-14) and pembrolizumab 200 mg intravenously every 3 weeks on day 1. Patients randomly assigned to exploratory arm E1 will be treated with T-DXd 5.4 mg/kg intravenously every 3 weeks on day 1 plus either 5-FU 600 mg/m2/day intravenously on days 1-5 or capecitabine 750 mg/m2 orally twice daily on days 1-14. Patients randomly assigned to exploratory arm E2 will be treated with trastuzumab, loading dose of 8 mg/kg intravenously followed by 6 mg/kg intravenously every 3 weeks, in combination with platinum-based chemotherapy (cisplatin 80 mg/m2/day intravenously on day 1 plus 5-FU 800 mg/m2/day intravenously on days 1-5 or oxaliplatin 130 mg/m2/day intravenously on day 1 plus capecitabine 1000 mg/m2 orally twice daily on days 1-14), which reflects the ToGA regimen.10

Tumor and safety assessment
A complete set of scans is required for all patients. Baseline tumor assessment will be carried out within 28 days of the date of randomization (tumor assessment carried out for the assessment of disease progression on a prior therapy is acceptable as the baseline assessment if carried out within 28 days of randomization). Tumor assessment will be carried out using computed tomography (CT) or magnetic resonance imaging (MRI) and include all known or suspected disease sites. Imaging must include an MRI or CT scan of the chest, abdomen, pelvis, and any other sites of disease. Antitumor activity will be assessed at baseline (screening), every 6 weeks (±7 days) for the first 48 weeks, and every 9 weeks thereafter (±7 days). All assessments of response will be made using RECIST v1.1 and assessed centrally. For patients with brain metastases at baseline, an MRI (preferred) or CT scan of the brain is mandatory and will follow the same schedule as the antitumor activity assessments. For patients without brain metastases at baseline, additional brain scans for tumor assessment are not needed unless clinically indicated.
Safety analyses will be carried out in the safety analysis set (SAS; all patients who receive at least one dose of trial treatment), and patients will be analyzed according to the trial treatment they receive. There will be no inferential statistical analysis for safety data; however, descriptive statistics will be calculated for quantitative safety data, and frequency counts and percentages will be determined for classification of qualitative safety data. All percentages will be calculated based on the number of participants in the SAS, unless determined otherwise.
Interstitial lung disease (ILD)/pneumonitis is a known and important risk associated with T-DXd treatment,31 and specific guidance for the management of ILD/pneumonitis is therefore provided in the DESTINY-Gastric05 trial protocol. Briefly, if a participant develops radiographic changes or new or worsening signs or symptoms that are potentially consistent with ILD/pneumonitis, study treatment should be interrupted for further evaluation to rule out ILD/pneumonitis. In the event of grade 1 ILD/pneumonitis, T-DXd treatment must be interrupted and can be restarted only if the event is fully resolved within 18 weeks. For grade 2 ILD/pneumonitis, T-DXd treatment must be permanently discontinued; however, pembrolizumab treatment can be restarted if the ILD/pneumonitis event improves to grade ≤1 within 12 weeks. For grade 3 or 4 ILD/pneumonitis, all trial interventions must be permanently discontinued.

Patient-reported outcomes
PROs will be evaluated using the FACT-GA questionnaire, the EQ-5D-5L questionnaire, the Patient Global Impressions scale, the patient-reported outcomes version of the Common Terminology Criteria for Adverse Events questionnaire, and the FACT—General item GP5 questionnaire.

Conclusion
Although progress has been made, an unmet medical need remains for more effective treatment options for patients with HER2-positive gastric or GEJ cancer.19,32,33 Given the promising preliminary safety and efficacy data for T-DXd observed in the DESTINY-Gastric03 trial, DESTINY-Gastric05 will explore the feasibility of using T-DXd as a part of first-line therapy in combination with chemotherapy with or without immuno-oncology therapy.

CRediT author statement
YYJ: investigation, data curation, writing—review & editing, project administration. ES: investigation, data curation, writing—review & editing, project administration. LS: investigation, data curation, writing—review & editing, project administration. JL: investigation, data curation, writing—review & editing, project administration. PMH: investigation, data curation, writing—review & editing, project administration. SL: investigation, data curation, writing—review & editing, project administration. DB: conceptualization, methodology, writing—review & editing. KK: conceptualization, methodology, writing—review & editing. YO: conceptualization, methodology, formal analysis, writing—review & editing. TK: conceptualization, methodology, writing—review & editing. KS: investigation, data curation, writing—review & editing, project administration.