Content area
Summary Background
Zanidatamab, a dual human epidermal growth factor receptor 2 (HER2)-targeted bispecific antibody, previously demonstrated encouraging antitumour activity and a manageable safety profile in patients with treatment-refractory HER2-expressing gastro-oesophageal adenocarcinoma. Here, we evaluated the antitumour activity and safety of zanidatamab plus chemotherapy in first-line HER2-positive advanced gastro-oesophageal adenocarcinoma.
MethodsThis phase 2 trial enrolled patients in Canada, South Korea, and the USA who were aged 18 years and older with untreated, metastatic, or advanced HER2-positive gastro-oesophageal adenocarcinoma (HER2 IHC 3+ or 2+ by local or central assessment [part 1]; HER2 IHC 3+ or 2+ with FISH+ by central assessment [part 2]). Eligible patients, with an Eastern Cooperative Oncology Group performance status of 0 or 1 received zanidatamab intravenously plus standard chemotherapy (CAPOX [capecitabine plus oxaliplatin], FP [5-fluorouracil [5-FU] plus cisplatin], or modified FOLFOX6 [mFOLFOX6; leucovorin, 5-FU, and oxaliplatin]). In our study, part 1 aimed to characterise the safety and tolerability of zanidatamab and find the recommended dose when administered with combination chemotherapy and part 2 aimed to evaluate the antitumour activity of zanidatamab administered with combination chemotherapy in patients receiving first-line treatment for HER2-expressing advanced gastro-oesophageal adenocarcinoma. Two dosing schemes for zanidatamab were used in this study: a weight-based regimen and a two-tiered flat dosing regimen. In the CAPOX and FP groups, patients received either 30 mg/kg zanidatamab or 1800 mg or 2400 mg (patients weighing <70 kg and ≥70 kg, respectively) every 3 weeks. In the CAPOX group, patients also received 1000 mg/m
Between Aug 29, 2019, and Feb 18, 2022, 46 patients were enrolled (39 [85%] were male; seven [15%] were female; 28 [61%] were white, 17 [37%] were Asian, and 43 [93%] were not Hispanic or Latino). Median follow-up was 47·9 months (IQR 39·2–53·7); eight (17%) patients were on treatment and 19 (41%) were in survival follow-up. The confirmed objective response rate was 76·2% (95% CI 60·5–87·9) with a median duration of response of 18·7 months (95% CI 10·4–44·1). The median progression-free survival was 12·5 months (95% CI 8·2–21·8) and median overall survival was 36·5 months (23·6–not estimable). The disease control rate was 88·1% (95% CI 74·4–96·0) and clinical benefit rate was 78·6% (95% CI 63·2–89·7). In part 1, there were no dose-limiting toxicities in six patients treated with zanidatamab plus CAPOX. One (50%) of two patients treated with zanidatamab plus FP had dose-limiting toxicities of diarrhoea and acute kidney injury (both grade 3). Two dose-limiting toxicities of diarrhoea (both grade 3) occurred in 2 (15%) of 13 patients receiving 5-FU 400 mg/m
Zanidatamab plus chemotherapy as first-line treatment of HER2-positive advanced gastro-oesophageal adenocarcinoma demonstrated clinically meaningful and durable antitumour activity, with a manageable safety profile.
FundingJazz Pharmaceuticals, Zymeworks.
Gastro-oesophageal adenocarcinoma represents a group of malignancies comprising gastric, oesophageal, and gastro-oesophageal junction adenocarcinomas. 1 Gastro-oesophageal adenocarcinoma remains one of the most prevalent cancers worldwide, accounting for over 1·4 million new cases and approximately 1 million deaths globally in 2022. 2 Patients with gastro-oesophageal adenocarcinoma are typically diagnosed with advanced disease and have a poor prognosis.
Approximately 20% of patients with gastro-oesophageal adenocarcinoma have human epidermal growth factor receptor 2 (HER2)-positive tumours (defined by immunohistochemistry [IHC] 3+ or IHC 2+ with fluorescence in-situ hybridisation [FISH] positivity). 3–5 In the phase 3 Trastuzumab for Gastric Cancer (ToGA) trial, trastuzumab—a HER2-targeted monoclonal antibody—plus platinum-fluoropyrimidine chemotherapy was associated with an objective response rate of 47% with a median progression-free survival of 6·7 months (95% CI 6–8; hazard ratio [HR] 0·71 vs chemotherapy alone) and a median overall survival of 13·8 months (95% CI 12–16; HR 0·74) as first-line treatment of patients with HER2-positive advanced gastric or gastro-oesophageal junction adenocarcinoma. 6 These results established trastuzumab plus platinum-based and fluoropyrimidine-based chemotherapy as first-line standard of care. 7–9 Moreover, nationally and regionally recognised practice guidelines recommend HER2 testing at the time of diagnosis of metastatic disease to identify patients who might benefit from HER2-targeted therapies. 3,8–10
Research in context
Evidence before this study
We searched PubMed on Feb 11, 2025, from database inception until the search date, for clinical trials published in English evaluating regimens for patients with previously untreated advanced human epidermal growth factor receptor 2 (HER2)-positive gastro-oesophageal adenocarcinoma. The search terms included “HER2-positive”, “gastric”, “gastro-esophageal junction”, “esophageal”, “adenocarcinoma”, “advanced”, “metastatic”, “unresectable”, “previously untreated”, and “first-line”. Results from the published literature show that trastuzumab, a monoclonal antibody that binds to the HER2 juxtamembrane domain, plus platinum-fluoropyrimidine doublet chemotherapy is the standard of care for first-line treatment of patients with newly diagnosed HER2-positive advanced gastro-oesophageal adenocarcinoma worldwide. For patients with HER2-positive and PD-L1–positive advanced gastric or gastro-oesophageal junction adenocarcinoma, the addition of pembrolizumab, an immune checkpoint inhibitor, to the trastuzumab-chemotherapy regimen demonstrated increased survival compared with trastuzumab-chemotherapy alone in a phase 3 trial. As such, pembrolizumab in combination with trastuzumab plus platinum-fluoropyrimidine chemotherapy has become the new first-line standard of care for patients with HER2-positive and PD-L1–positive tumours. Other HER2-targeted agents, alone or in combination, including antibody-drug conjugates and tyrosine kinase inhibitors, have been studied or have reported antitumour activity, but none are currently approved for first-line treatment of patients with HER2-positive advanced gastro-oesophageal adenocarcinoma. For example, the addition of pertuzumab to first-line trastuzumab plus chemotherapy did not significantly improve the overall survival versus trastuzumab plus chemotherapy alone in advanced HER2-positive gastric or gastro-oesophageal junction adenocarcinoma.
Added value of this study
To our knowledge, this is the first global, phase 2 trial of zanidatamab, a dual HER2-targeted bispecific antibody, that simultaneously targets both the HER2 juxtamembrane (ECD4) and dimerisation (ECD2) domains in trans (between molecules), in combination with standard chemotherapy regimens for the first-line treatment of patients with HER2-positive advanced gastro-oesophageal adenocarcinoma. This regimen provided clinically meaningful and sustained antitumour activity, with a robust response rate observed in the subset of patients with HER2-positive tumours (status confirmed by central laboratory assessment). The safety profile of the combination was manageable, with the incidence of grade 3 diarrhoea mitigated after the implementation of mandatory antidiarrhoeal prophylaxis for all patients and omission of the 5-fluorouracil bolus for patients receiving modified FOLFOX (leucovorin, 5-fluorouracil, and oxaliplatin) chemotherapy.
Implications of all the available evidence
This study provides evidence that zanidatamab plus standard chemotherapy offers encouraging antitumour activity with sustained responses and a manageable safety profile in the first-line treatment of patients with HER2-positive advanced gastro-oesophageal adenocarcinoma. These results, in addition to the ongoing global randomised phase 3 trial of zanidatamab plus chemotherapy with or without the PD-1 inhibitor, tislelizumab, as first-line treatment of HER2-positive advanced gastro-oesophageal adenocarcinoma ( NCT05152147), support further research of these combination regimens in this setting.
In 2023, the phase 3 KEYNOTE-811 trial demonstrated that adding pembrolizumab to trastuzumab-based combination regimens in patients with HER2-positive advanced gastric or gastro-oesophageal junction adenocarcinomas that also expressed PD-L1 (combined positive score ≥1) improved survival versus trastuzumab plus chemotherapy alone (median progression-free survival 10·9 months vs 7·3 months, HR 0·73; median overall survival 20·1 months vs 15·7 months, HR 0·79). 11,12 Despite this advancement, a high unmet need remains for first-line treatments that could further improve survival outcomes for patients with HER2-positive advanced gastro-oesophageal adenocarcinoma while also maintaining a manageable safety profile.
Zanidatamab, a dual HER2-targeted bispecific antibody, binds to two non-overlapping domains on HER2 (the juxtamembrane and dimerisation domains) in trans (between molecules) to initiate distinct HER2 reorganisation. 13 In preclinical studies, zanidatamab was shown to drive multiple antitumour mechanisms of action, including facilitation of HER2 internalisation and subsequent degradation, inhibition of HER2 homo-dimerisation and hetero-dimerisation, and activation of immune-mediated effects (complement-dependent cytotoxicity and antibody-dependent cellular cytotoxicity and phagocytosis). 13
In a phase 1 trial, zanidatamab as a monotherapy and in combination with chemotherapy demonstrated encouraging antitumour activity (objective response rates of 32% and 49%, respectively), with a manageable safety profile in heavily pretreated patients with HER2-expressing gastro-oesophageal adenocarcinoma. 14,15
In our study, part 1 aimed to characterise the safety and tolerability of zanidatamab and find the recommended dose when administered with combination chemotherapy, and part 2 aimed to evaluate the antitumour activity of zanidatamab administered with combination chemotherapy in patients receiving first-line treatment for HER2-expressing advanced gastro-oesophageal adenocarcinoma.
Methods Study design and participantsThis was a phase 2, multicentre, open-label, two-part study tested zanidatamab plus standard chemotherapy for first-line treatment of patients with select HER2-expressing gastrointestinal cancer, including gastro-oesophageal adenocarcinoma.
In this Article, we present results from the gastro-oesophageal adenocarcinoma cohort, which enrolled patients in Canada, South Korea, and the USA; results for other cancer types will be reported separately. Eligible patients were aged 18 years and older, with unresectable, locally advanced, recurrent or metastatic HER2-expressing gastro-oesophageal adenocarcinoma, an Eastern Cooperative Oncology Group performance status of 0 or 1, and adequate cardiac left ventricular function (ie, left ventricular ejection fraction equal to or higher than the lower limit of the institutional normal range). Patients with clinically significant interstitial lung disease or cardiac disease; previous or concurrent malignancy; acute or chronic uncontrolled renal disease, pancreatitis, or liver disease; or active infection with hepatitis B, hepatitis C, or HIV were ineligible. Biopsies from either primary or metastatic lesions were permitted for the testing for HER2 status. Patients could participate in part 1 if their gastro-oesophageal adenocarcinoma HER2 IHC test score was 3+ or 2+ with or without HER2 amplification per local or central assessment, regardless of measurable or non-measurable disease; whereas in part 2, only patients with tumours assessed as HER2 IHC 3+ or 2+ with FISH-positive per central laboratory and who had measurable disease per Response Evaluation Criteria in Solid Tumours version 1.1 (RECIST v1.1) 16 were permitted to enrol.
No previous treatment with a HER2-targeted agent was permitted. Patients who had received adjuvant or neoadjuvant therapy with systemic anticancer treatment were eligible, provided treatment was completed 6 months or more before first dose of study treatment. Archival tissue samples were required for retrospective central review. Full eligibility criteria are described in the study protocol ( appendix p 10), and the protocol and all amendments were approved by the relevant institutional review board at each study centre. No cancer survivors or patient representatives were involved in shaping the research question, the study design, the choice of outcome measures, the conduct of the trial, the analysis or interpretation of the data, or in the writing of the report. All patients provided written informed consent. This study was conducted in accordance with the Declaration of Helsinki, Good Clinical Practice guidelines, and International Council for Harmonisation guidelines and overseen by a safety monitoring committee. It was registered at ClinicalTrials.gov ( NCT03929666) and is complete for enrolment.
ProceduresAll patients provided fresh or archival tumour biopsies for central assessment of HER2 status using HercepTest for HER2 protein expression and HER2 IQFISH pharmDx for HER2 amplification (SK001 and K5731 respectively; both Agilent [Dako], Carpinteria, CA, USA).
Part 1 of the study followed a de-escalation design. Zanidatamab doses in the combination regimens determined in part 1 based on the number of dose-limiting toxicities were then used in part 2. Eligible patients received zanidatamab plus standard chemotherapy. Standard chemotherapy regimens were chosen based on findings from landmark studies 6,11 and national guidelines, 10 as well as to provide flexibility to administer physician's choice of chemotherapy. The three regimens were CAPOX (capecitabine plus oxaliplatin), FP (5-fluorouracil [5-FU] plus cisplatin), or modified FOLFOX6 (mFOLFOX6; leucovorin, 5-FU, and oxaliplatin). All patients received mandatory prophylaxis (acetaminophen, diphenhydramine, and corticosteroids) for potential infusion-related reactions as previously reported. 17
Two dosing schemes for zanidatamab were used in this study: a weight-based regimen (scheme 1) and a two-tiered flat dosing regimen (scheme 2). The reason to switch from a weight-based zanidatamab regimen to a flat dosing one was based on similar population pharmacokinetic modelling and more consistent zanidatamab exposure. 18
For each chemotherapy regimen, standard starting doses were selected, and up to two step-down doses could be evaluated if deemed necessary by the safety monitoring committee and the sponsor. Zanidatamab was delivered intravenously. In the CAPOX cohort, the starting dose was 30 mg/kg zanidatamab (scheme 1) or 1800 mg for patients weighing under 70 kg or 2400 mg for patients weighting 70 kg and above (scheme 2), plus 1000 mg/m 2 capecitabine orally twice daily on days 1–14 every 3 weeks, and 130 mg/m 2 oxaliplatin intravenously every 3 weeks. In the FP cohort, the starting dose for zanidatamab was 30 mg/kg (scheme 1); the starting dose was 1800 mg or 2400 mg for patients weighing under 70 kg or 70 kg and above, respectively (scheme 2), every 3 weeks, plus 80 mg/m 2 cisplatin intravenously every 3 weeks, and 800 mg/m 2 5-FU per day continuous intravenous infusion on days 1–5 every 3 weeks. The starting dose for zanidatamab was 20 mg/kg (mFOLFOX6–1), or 1200 mg or 1600 mg for patients weighing under 70 kg or 70 kg and above, respectively (mFOLFOX6–2), every 2 weeks, plus 400 mg/m 2 intravenous leucovorin every 2 weeks, 85 mg/m 2 intravenous oxaliplatin every 2 weeks, and 1200 mg/m 2 per day 5-FU as a continuous intravenous infusion for 48 h every 2 weeks. mFOLFOX6–1 included the administration of a 400 mg/m 2 5-FU on days 1 and 15; mFOLFOX6–2 omitted this 5-FU bolus.
Zanidatamab dose interruptions and reductions due to toxicity were permitted in both schemes, with a minimum dose of 20 mg/kg every 3 weeks, or 10 mg/kg every 2 weeks for weight-based dosing or less than 50% of the starting dose for flat-dosing.
Chemotherapy was required for at least six cycles unless there was evidence of intolerability to chemotherapy, radiographically confirmed disease progression per RECIST v1.1, withdrawal of consent, start of a subsequent therapy, or study termination by the sponsor. Patients who discontinued chemotherapy due to toxicity without disease progression could continue zanidatamab monotherapy until discontinuation criteria were met. For patients receiving CAPOX or mFOLFOX6 who discontinued oxaliplatin due to toxicity unrelated to zanidatamab, 5-FU, or capecitabine could be continued. However, if 5-FU or capecitabine was discontinued, the study protocol required discontinuation of oxaliplatin as well. After six cycles, chemotherapy (including fluoropyrimidine maintenance regimens) could be continued per investigator and patient discretion. Tumour responses were examined every 6 weeks from day 1 of cycle 1 and at end of treatment visit (within 7 days after the last dose of study treatment) if more than 4 weeks since the previous scan using CT, MRI, or both according to RECIST v1.1. 16 Adverse events and concomitant medications were monitored throughout the study, and laboratory evaluation was assessed every 2–3 weeks. Study treatment continued until radiographically confirmed disease progression per RECIST v1.1, investigator's decision, withdrawal of consent, start of a subsequent therapy, or study termination by the sponsor. Patients who discontinued study treatment could remain on study and be followed for disease progression (if this was not the reason for discontinuation) and survival unless they withdrew consent. Patients could withdraw from the study because of adverse events, loss to follow-up, physician's decision, pregnancy, progressive disease, protocol violation, study termination by sponsor, or other non-toxicity reasons.
OutcomesPart 1 assessed the safety and tolerability of zanidatamab plus chemotherapy to establish the recommended dose for part 2. The primary endpoints of part 1 were frequencies of dose-limiting toxicities and dose reductions of zanidatamab and chemotherapy, along with frequency and severity of treatment-emergent adverse events, serious adverse events, the frequency of abnormalities in clinical laboratory evaluations, electrocardiograms, and of left ventricular ejection fraction. For part 2, the primary endpoint was confirmed investigator-assessed objective response rate (defined as the proportion of patients with complete or partial responses that were radiographically confirmed at least 4 weeks following initial documentation of objective response by investigator assessment per RECIST v1.1). Secondary antitumour activity endpoints in part 1 were objective response rate , duration of response (time from first objective response to documented disease progression or death from any cause); disease control rate (proportion of patients achieving a best response of complete or partial response or with stable disease); clinical benefit rate (proportion of patients achieving stable disease for ≥24 weeks or a confirmed, best overall response of a complete or partial response); progression-free survival (time from first dose of study treatment to disease progression or death from any cause), and overall survival (time from first dose of study treatment to death due to any cause); and part 2 secondary antitumour activity endpoints were similar to those of part 1 (disease control rate, duration of response, clinical benefit rate, progression-free survival, and overall survival). Safety and tolerability outcomes were the same as primary endpoints of part 1 and were secondary endpoints in part 2. Additional secondary endpoints (pharmacokinetics and immunogenicity of zanidatamab) will be reported separately.
The severity of adverse events was graded per the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0. 19 Adverse events of special interest included infusion-related reactions, left ventricular dysfunction (defined as absolute decrease in left ventricular ejection fraction of ≥10% from pretreatment baseline and value <50% by echocardiogram or multiple gated acquisition scan results, or grade ≥2 adverse events of preferred term from the Heart Failure Custom Query), and non-infectious pulmonary toxicities (defined by the Broad Interstitial Lung Disease Standardized MedDRA Query).
Statistical analysisNo statistical hypothesis testing or formal sample size calculation were performed for this study. Depending on the number of de-escalations required from the initial recommended doses across all cohorts, the approximate minimum and maximum sample sizes for part 1 were six and 14 patients in whom dose-limiting toxicities could be evaluated, respectively, for each dosing group. All data reported here are from an interim analysis.
Safety and survival endpoints were assessed in the safety analysis set (patients who received any amount of zanidatamab, chemotherapy, or both, in part 1 or part 2). Objective response rate was assessed in the measurable disease set (patients in the safety analysis set with one or more measurable target lesion per RECIST v1.1 at baseline) and in the response evaluable set (patients in the measurable disease set with one or more post-baseline tumour assessment per RECIST v1.1 or discontinuing treatment due to death or unequivocal clinical progression without any post-baseline disease assessment). Patients alive without disease progression at the time of data cut-off were censored for duration of response and progression-free survival analyses at the time of their most recent tumour assessment before subsequent anticancer treatment. If disease progression or death occurred after two or more consecutive missed disease assessments, patients were censored at the last non-progressive disease assessment for progression-free survival and duration of response. Patients who had not died or who were lost to follow-up at the data cut-off date were censored for overall survival analyses at the date they were last known to be alive. Four enrolled patients were not included in antitumour activity analysis due to lack of measurable target lesions per RECIST v1.1 at baseline (n=3) or missing response data during this study (n=1). Outcomes were also analysed post hoc by HER2 status per central laboratory test and date of enrolment pre- versus post-mandatory antidiarrhoeal prophylaxis (implemented on Sept 30, 2020).
Binomial Clopper–Pearson 95% CIs were calculated for categorical response outcomes, and medians for time-to-event outcomes were estimated using the Kaplan–Meier method with the 95% CIs computed using the Brookmeyer and Crowley method with log-log transformation. Proportions of patients without an event at specific time points were estimated using the Kaplan–Meier method with two-sided 95% CIs using the Greenwood estimator. Time to first confirmed response was also analysed in the response evaluable set as a post-hoc outcome. All statistical analyses and graphing were performed using SAS version 9.4.
Role of the funding sourceJazz Pharmaceuticals and Zymeworks funded the study and participated in the study design, data collection, data analyses, data interpretation, and writing and approval of this report for publication in conjunction with the authors.
ResultsBetween Aug 29, 2019, and Feb 18, 2022, 57 patients were screened. After 11 patients (six in part 1 and five in part 2) were found to be ineligible, 46 patients were enrolled (28 [61%] in part 1 and 18 [39%] in part 2; figure 1 ) and treated with zanidatamab plus chemotherapy at 14 sites across three countries (eight [17%] in Canada; 14 [30%] in South Korea; and 24 [52%] in the USA; appendix p 2). 39 patients (85%) were male and seven (15%) were female; 28 (61%) were white and 17 (37%) were Asian; 43 (93%) were not Hispanic or Latino in ethnicity. 38 (83%) patients had advanced disease at diagnosis, and 35 (76%) had gastric or gastro-oesophageal junction adenocarcinomas ( table 1 ). 41 (89%) patients had tumours confirmed as HER2-positive by central assessment (23 [82%] of 28 in part 1; 18 [100%] of 18 in part 2); while IHC 2+ or 3+ were noted by local assessment, four (9%) of all 46 patients had tumours deemed HER2 IHC 0 and FISH negative, IHC 1+ and FISH negative, or IHC 2+ and FISH negative by central assessment despite being enrolled in part 1 based on local assessment.
At data cut-off on 28 July, 2024, the median follow-up was 47·9 months (IQR 39·2–53·7). Eight (17%) patients were continuing to receive treatment, 19 (41% [including 11 who discontinued zanidatamab but continued with the study]) patients were in survival follow-up ( figure 1). The median number of treatment cycles was 13·5 (IQR 7–28) for zanidatamab, six cycles (2–6) for continuous 5-FU infusions, 10·5 cycles (6–18·5) for capecitabine, 4·5 cycles (3–6) for cisplatin, and five cycles (3–6) for oxaliplatin. 33 (72%) patients received one or more subsequent systemic anticancer therapies ( appendix p 3), including 20 (43%) receiving one or more HER2-targeted agents (trastuzumab: 15 [33%]; trastuzumab deruxtecan: six [13%]; and other: three [7%]).
In part 1, there were no dose-limiting toxicities in six patients treated with zanidatamab plus CAPOX. One (50%) of two patients treated with zanidatamab plus FP had dose-limiting toxicities of diarrhoea and acute kidney injury (both grade 3). Two dose-limiting toxicities of diarrhoea (both grade 3) occurred in two (15%) of 13 patients receiving 5-FU 400 mg/m 2 bolus on day 1 and 15 as part of the zanidatamab plus mFOLFOX6–1 regimen. After these 13 patients were treated with mFOLFOX6–1, the regimen was modified to omit the 5-FU bolus (mFOLFOX6–2) per safety monitoring committee recommendation.
Considering 24 (96%) of the first 25 patients who were enrolled and treated had treatment-emergent diarrhoea events of any cause ( appendix p 4), the safety monitoring committee also recommended amending the protocol to implement mandatory antidiarrhoeal prophylaxis (4 mg loperamide twice daily starting on the first treatment day of cycle 1 and continuing for at least 7 days) for all subsequent patients. Continued loperamide treatment after cycle 1 was at the investigator's discretion. Among the seven patients receiving mFOLFOX6–2 and mandatory antidiarrhoeal prophylaxis in part 1, there was one dose-limiting toxicity of grade 3 diarrhoea. The mFOLFOX6–2 regimen with the omission of 5-FU bolus was the recommended regimen for part 2 per the safety monitoring committee approval.
Among the 42 response-evaluable patients, the confirmed objective response rate was 76·2% (95% CI 60·5–87·9), with complete responses in three (7%) patients and partial responses in 29 (69%) patients ( table 2 , figure 2A ). The median duration of response was 18·7 months (95% CI 10·4–44·1) for 32 patients with a confirmed response. The post-hoc median time to first confirmed response was 1·3 months (IQR 1·3–1·4) with 27 (84%) of 32 responses observed at the first tumour assessment after treatment initiation. Eight (17%) patients had ongoing responses at data cut-off ( figure 2B). Responses assessed in patients with measurable disease ( appendix p 5) were consistent with the response-evaluable population.
Among all 46 patients, the median progression-free survival was 12·5 months (95% CI 8·2–21·8; figure 3A ; reasons for censoring are given in the appendix, p 6). Estimated 12-month and 24-month progression-free survival rates were 57% (95% CI 40–70) and 31% (17–46), respectively. The median overall survival was 36·5 months (95% CI 23·6–not estimable; figure 3B). Estimated 12-month and 24-month overall survival rates were 87% (95% CI 72–94) and 65% (49–77), respectively. In a post-hoc subgroup analysis of the 41 treated patients with centrally confirmed HER2-positive tumours, the confirmed objective response rate was 84% (31 of 37 response-evaluable patients; 95% CI 68·0–93·8; table 2; appendix pp 8–9). The median progression-free survival was 15·2 months (95% CI 9·5–33·4) and median overall survival was 36·5 months (23·6–not estimable; figure 3C,D).
In a post-hoc analysis of patients enrolled pre-implementation versus post-implementation of the antidiarrhoeal prophylaxis, 14 (61%) of 23 response-evaluable patients enrolled pre-implementation (95% CI 38·5–80·3) had confirmed responses versus 18 (95%) of 19 enrolled post-implementation (74·0–99·9). Prolonged treatment duration and exposure were observed among patients who received antidiarrhoeal prophylaxis ( appendix p 4).
All-cause treatment-emergent adverse events occurred in 46 (100%) patients, with 38 (83%) having events at grade 3 or above, including one grade 5 event of respiratory failure that was deemed unrelated to treatment per the investigators ( appendix p 7). Adverse events of special interest occurred in 11 (24%) patients: ten (22%) had grade 1 or 2 infusion-related reactions; one (2%) had a grade 2 non-infectious pulmonary toxicity deemed unrelated to zanidatamab; and no patients reported left ventricular dysfunction or grade 2 or higher heart failure. Four (9%) patients had zanidatamab dose reductions due to any-cause adverse events (all four [17%] of 24 patients were on zanidatamab with mFOLFOX6). Six (13%) patients discontinued zanidatamab due to any-cause adverse events (five [21%] of 24 on zanidatamab with mFOLFOX6 and one [5%] of 20 on zanidatamab with CAPOX).
All 46 (100%) patients had treatment-related adverse events related to any study treatment component or components ( table 3 ). Diarrhoea (18 [39%]) and hypokalaemia (10 [22%]) were the most common grade 3 or 4 treatment-related adverse events. There were no grade 5 treatment-related events. Serious treatment-related adverse events occurred in eight (17%) patients, with diarrhoea (3 [7%]) being the most common. These events occurred concurrently in three patients (diarrhoea and acute kidney injury in two patients; diarrhoea, acute kidney injury, and hypokalaemia in one patient).
The median time to onset of any-grade diarrhoea of any cause in the 43 patients experiencing one or more event was 5 days (IQR 3–8), and the median duration was 5 days (2–20), irrespective of the cycle in which the diarrhoea event occurred. In a post-hoc analysis, 13 (52%) of 25 patients enrolled pre-implementation of antidiarrhoeal prophylaxis had grade 3 treatment-related diarrhoea versus five (24%) of 21 patients enrolled post-amendment ( appendix p 4). The median duration of any-cause grade 1 or 2 and grade 3 diarrhoea was 6·5 days (IQR 2–29) and 3 (2–5) days, respectively, in patients enrolled pre-implementation of antidiarrhoeal prophylaxis versus 3 days (2–13) and 4·5 days (3–8) in patients enrolled post-implementation. Two patients enrolled pre-implementation of antidiarrhoeal prophylaxis had zanidatamab dose reductions due to diarrhoea, and one patient enrolled post-implementation had dose reductions due to diarrhoea. Two patients enrolled pre-implementation of antidiarrhoeal prophylaxis discontinued zanidatamab due to diarrhoea, while three had at least one serious diarrhoea event. No patients enrolled post-implementation discontinued due to diarrhoea or had a serious diarrhoea event.
DiscussionIn this phase 2 trial, the confirmed objective response rate in the overall response-evaluable population was 76·2% (95% CI 60·5–87·9); zanidatamab plus chemotherapy showed clinically meaningful antitumour activity and promising survival outcomes as a first-line treatment of patients with HER2-positive advanced gastro-oesophageal adenocarcinoma (for patients with centrally confirmed HER2-positive tumours: objective response rate of 84%; median duration of response of 20·4 months; median progression-free survival of 15·2 months; and median overall survival of 36·5 months). The safety profile of zanidatamab plus chemotherapy was generally manageable with antidiarrhoeal prophylaxis, and no treatment-related deaths reported. Discontinuations from zanidatamab due to intolerability were infrequent.
For over a decade, trastuzumab plus chemotherapy has been the first-line standard of care for patients with HER2-positive advanced gastro-oesophageal adenocarcinoma. 20 In the phase 3 JACOB trial, adding pertuzumab to first-line trastuzumab plus chemotherapy did not significantly improve the median overall survival versus trastuzumab plus chemotherapy alone (17·5 months vs 14·2 months; p=0·057), with modest improvement in median progression-free survival (8·5 months vs 7·0 months) and objective response rate (57% vs 48%). 21 To date, improved survival has only been observed with the addition of an immune checkpoint inhibitor (eg, a PD-1 inhibitor) to trastuzumab plus chemotherapy for patients whose gastric or gastro-oesophageal junction adenocarcinomas also expressed PD-L1, in KEYNOTE-811. 11,12
Despite inherent limitations of cross-trial comparisons, the response rate, median progression-free survival, and median overall survival with zanidatamab plus chemotherapy in our trial compares favourably with current first-line standard of care. 6,11 These observations are supported by preclinical findings, where zanidatamab induced unique HER2 clustering on the cell surface upon binding and complement-dependent cytotoxicity that was not observed with trastuzumab, pertuzumab, or both in HER2-expressing tumour cells. 13 Additionally, zanidatamab demonstrated greater antitumour activity than trastuzumab or trastuzumab plus pertuzumab in HER2-expressing xenograft gastric cancer models. 13
Safety outcomes observed in this study are generally consistent with the known safety profiles of standard chemotherapy regimens used for treatment of HER2-positive gastro-oesophageal adenocarcinoma. Although diarrhoea is a common adverse event of HER2-targeted therapies 22 and certain chemotherapy components (eg, fluoropyrimidines), the incidence of grade 3 diarrhoea at the initiation of this trial was higher than expected based on the results of a phase 1 trial in previously treated (including with trastuzumab) patients with HER2-expressing gastro-oesophageal adenocarcinoma. 15 In that earlier trial, grade 3 diarrhoea occurred in 3% of patients receiving zanidatamab monotherapy or 7% receiving zanidatamab plus paclitaxel or capecitabine, without the use of antidiarrhoeal prophylaxis.
To mitigate the risk of grade 3 diarrhoea, the protocol was amended to include mandatory antidiarrhoeal prophylaxis for all patients for the first 7 days of treatment and to remove the 5-FU bolus from mFOLFOX6–1. Recent real-world studies showed that omitting 5-FU bolus from the mFOLFOX6 regimen improved tolerability without affecting survival in patients with advanced gastrointestinal cancers. 23 Patients enrolled post-implementation of antidiarrhoeal prophylaxis had a lower incidence of grade 3 diarrhoea, a longer treatment exposure, and a higher antitumour response versus patients enrolled pre-implementation. Although five (24%) of 21 patients had grade 3 diarrhoea post-implementation of antidiarrhoeal prophylaxis, none discontinued zanidatamab because of these events. The higher antitumour response in patients receiving antidiarrhoeal prophylaxis might possibly be attributable in part to the longer treatment duration and fewer treatment delays, dose reductions, and discontinuations due to diarrhoea.
Results observed in this trial broadly align with those from a concurrently run phase 1b/2 trial of zanidatamab plus CAPOX and tislelizumab (an anti–PD-1 antibody), which was conducted at sites in China and South Korea, with mandatory antidiarrhoeal prophylaxis for cycle 1 being added as an amendment. 24 In that trial, among 33 patients with HER2-positive gastric or gastro-oesophageal junction adenocarcinoma receiving the combination regimen, confirmed objective response rate was 75·8% with a median duration of response of 22·8 months and a median progression-free survival of 16·7 months. Grade 3 or 4 treatment-related adverse events occurred in 67% of patients with diarrhoea (27%) and decreased appetite (6%) being most common. Immune-mediated adverse events with tislelizumab occurred in 27% of patients.
Limitations of our study include the small sample size (particularly only two patients treated in the FP cohort), the lack of a comparator control group, the omission of patient-reported outcomes, and the lack of blinded independent central review for the primary endpoint. The extended study follow-up (median >4 years) may have contributed to the higher proportion of patients who received subsequent therapy versus those of previous reports. 11,25 A potential bias is that subsequent anticancer therapy might affect overall survival analyses. Continued trastuzumab treatment after progression on first-line trastuzumab-chemotherapy regimen did not improve survival outcomes. 26 Additionally, there were no meaningful differences in overall survival among the six patients who went on to receive trastuzumab deruxtecan versus other patients in this study. Subsequent anticancer therapy might have a small effect on the overall survival observed in the present study, and first-line treatment with zanidatamab could prime responses to subsequent HER2-targeted therapies for patients. Nonetheless, the randomised, comparator-controlled trial, HERIZON-GEA-01 ( NCT05152147), might shed more light on contributing factors of overall survival.
In conclusion, first-line treatment with zanidatamab plus chemotherapy demonstrated rapid and durable antitumour activity with promising survival outcomes in patients with HER2-positive advanced gastro-oesophageal adenocarcinoma. Zanidatamab plus chemotherapy in the first-line setting was well tolerated with a manageable safety profile when incorporating antidiarrhoeal prophylaxis in the first 7 days of treatment. If these results are confirmed in a large-scale, randomised, phase 3 trial, zanidatamab could represent a substantial advancement in the treatment of HER2-positive advanced gastro-oesophageal adenocarcinoma. The clinical development is ongoing with the global, randomised, phase 3 trial (HERIZON-GEA-01) of zanidatamab plus chemotherapy with or without tislelizumab versus trastuzumab plus chemotherapy for first-line treatment of patients with centrally confirmed HER2-positive advanced gastro-oesophageal adenocarcinoma. Building on findings from this phase 2 study, antidiarrhoeal prophylaxis has been incorporated in HERIZON-GEA-01, and patient-reported outcomes are being assessed to advance our understanding of zanidatamab plus chemotherapy.
ContributorsEE was involved in the conceptualisation and design of the study. EE, JA, HB, CSD, SI, Y-KK, JHK, K-WL, BL, RM, D-YO, SYR, YMS, and GK collected the data. LY, MOA, and PMG analysed the data. All authors accessed and verified the data, interpreted the data, and contributed to the preparation and review of the manuscript for publication. The corresponding author had full access to all data in this study and holds final responsibility for the decision to submit the manuscript for publication.
Data sharingAll relevant data are provided within the manuscript and supporting files. Jazz Pharmaceuticals has established a process to review requests from qualified external researchers for data from Jazz Pharmaceuticals-sponsored clinical trials in a responsible manner that includes protecting patient privacy, assurance of data security and integrity and furthering scientific and medical innovation. External researchers may submit requests for data generated from Jazz Pharmaceuticals-sponsored clinical trials. Requests must be for data owned by Jazz Pharmaceuticals from completed trials where the product or indication is approved in the USA or EU. Requests will be reviewed based on factors including, but not limited to, Jazz Pharmaceuticals's ability to share the requested data, qualifications of the researcher, legitimacy of the research purpose and scientific merit. Additional details on Jazz Pharmaceuticals data sharing criteria and process for requesting access can be found at https://www.jazzpharma.com/science/clinical-trial-data-sharing/.
Declaration of interestsEE has served as a consultant for AbbVie, Adaptimmune, Astellas Pharma, BeiGene, Bristol Myers Squibb, Daiichi-Sankyo, Natera, Roche, Viracta Therapeutics, and Zymeworks; has served as a member of the steering committee for AstraZeneca and Jazz Pharmaceuticals; has received grant or research support from Amgen, Arcus Biosciences, AstraZeneca, Bold Therapeutics, Bristol Myers Squibb, Jazz Pharmaceuticals, and Zymeworks; and has a family member who is employed by Merck. JA has received honoraria from Acrotech Biopharma, Aduro Biotech, Amgen, Astellas Pharma, AstraZeneca, BeiGene, Boehringer Ingelheim, Bristol Myers Squibb, Daiichi-Sankyo, DAVA Pharmaceuticals, Eli Lilly, Fresenius Kabi, Gilead Sciences, GRAIL, Merck, Novartis, Oncotherics, Servier, and Zymeworks; has participated in consulting or advisory roles for the American Cancer Society, Amgen, Arcus Biosciences, Astellas Pharma, BeiGene, Bristol Myers Squibb, Geneos, Gilead Sciences, Insys Therapeutics, Merck, Novartis, Servier, and Vaccinogen; received research funding from Amgen, Bristol Myers Squibb, Daiichi-Sankyo, Delta-Fly Pharma, Gilead Sciences, Lilly/ImClone, MedImmune, Merck, Novartis, ProLynx, Roche/Genentech, Taiho Pharmaceutical, Takeda, and Zymeworks; and institutional funding from Astellas Pharma. HB is an employee of and owns stock or stock options in HCA Healthcare; has received consulting or advisory fees from AstraZeneca, GRAIL, Incyte, Roche, and Vincerx Pharma; has received research funding from AbbVie, Agios, Arch Oncology, ARMO BioSciences, Array BioPharma, Arvinas, AstraZeneca, Bayer, BeiGene, BioAtla, BioMed Valley Discoveries, BioTheryX, Boehringer Ingelheim, Bristol Myers Squibb, CALGB, Celgene, CicloMed, Coordination Pharmaceuticals, CytomX Therapeutics, eFFECTOR Therapeutics, Eli Lilly, EMD Serono, Foundation Medicine, Genentech/Roche, Gilead Sciences, GlaxoSmithKline, Gossamer Bio, Harpoon Therapeutics, Hengrui Therapeutics, Incyte, Janssen, Jounce Therapeutics, Kymab, Macrogenics, MedImmune, Merck, Millennium/Takeda, Moderna Therapeutics, NGM Biopharmaceuticals, Novartis, Pfizer, Revolution Medicines, Ryvu Therapeutics, Seagen, Tesaro, TG Therapeutics, Verastem, Vertex Pharmaceuticals, XBiotech, and Zymeworks; and has uncompensated relationships with Bristol Myers Squibb and Novartis. CSD is an employee of the National Comprehensive Cancer Network; has received institutional research funding from 2seventy bio, AbbVie, Bristol Myers Squibb, Eli Lilly, Genentech Foundation, GlaxoSmithKline, Merck Foundation, Pfizer, and Sanofi; and has participated in a data safety monitoring board for Zymeworks. SI reports consulting fees from Astellas Pharma, AstraZeneca, BeiGene, Exelixis, and Merck; speakers fees from Astellas Pharma; and research funding from Amgen, ASLAN Pharmaceuticals, Astellas Pharma, AstraZeneca, Bayer, BeiGene, Bristol Myers Squibb, Daiichi-Sankyo, Eli Lilly, Indivumed, MSD Oncology, Roche/Genentech, Sillajen, and Zymeworks. Y-KK reports consulting fees from ALX Oncology, Amgen, Blueprint, Bristol Myers Squibb, Daehwa, Macrogenics, MSD, Novartis, Surface Oncology, Roche, and Zymeworks. K-WL received grants for the present manuscript from Zymeworks to his institution for conducting clinical trials; and has also received grants from ABLBIO, ALX Oncology, Amgen, Astellas, AstraZeneca, BeiGene, Bolt Therapeutics, Daiichi-Sankyo, Elevar Therapeutics, Exelixis, Genome & Company, Green Cross Corp, Idience, InventisBio, Leap Therapeutics, Macrogenics, MedPacto, Merck KGaA, Metafines, MSD, Oncologie, Ono Pharmaceutical, Pharmacyclics, Roche, Seagen, Taiho Pharmaceutical, Trishula Therapeutics, and Y-BIOLOGICS to his institution for conducting clinical trials outside the submitted work; received honoraria from Astellas, Boryung, Daiichi-Sankyo, Ono Pharmaceutical, and Sanofi-Aventis; and has participated on a data safety monitoring board or advisory board for ALX Oncology and Metafines. BL has participated in an advisory board for Seagen and received clinical research institutional support from Exelixis, Genentech/Roche, Incyte, Jazz Pharmaceuticals, Merck, Pfizer/Cardiff Oncology, Relay Therapeutics, and Tvardi Therapeutics. RM has participated in advisory boards for Astellas, AstraZeneca, Bristol Myers Squibb, Bostongene, Eli Lilly, GlaxoSmithKline, Guardant Health, Merck, Novartis, and Seagen; has served as a consultant for Eli Lilly; and serves on the data and safety monitoring board for Arcus Biosciences. D-YO has participated in advisory boards for AbbVie, ASLAN, Arcus Biosciences, Astellas, AstraZeneca, Bayer, Basilea, BeiGene, Bristol Myers Squibb/Celgene, Eutilex, Genentech/Roche, Halozyme, IQVIA, J Pharma, LG Chem, Merck Serono, Mirati Therapeutics, MSD, Novartis, Taiho Pharmaceutical, Turning Point, Yuhan, and Zymeworks; and has received research grants from Array, AstraZeneca, BeiGene, Eli Lilly, Handok, MSD, Novartis, and Servier. SYR reports consulting fees from Aadi, Amgen, AstraZeneca, Daiichi-Sankyo, Eisai, Eutilex, Indivumed, LG Chem, MSD Oncology, Ono Pharmaceutical, and Toray Industries; speaker fees from Amgen, AstraZeneca, Bristol Myers Squibb/Ono Pharmaceuticals, Daiichi-Sankyo/UCB Japan, Eisai, and MSD Oncology; and research funding from Amgen, ASLAN Pharmaceuticals, Astellas Pharma, AstraZeneca, Bayer, BeiGene, Bristol Myers Squibb, Daiichi-Sankyo, Eisai, Eli Lilly, Indivumed, MSD Oncology, Roche/Genentech, Sillajen, and Zymeworks. LY, MAO, and PMG are former employees of, and owned stock or stock options in Zymeworks; and are current employees of and own stock or stock options of Jazz Pharmaceuticals. GK has received research funding to his institution from AstraZeneca, Bristol Myers Squibb, CARsgen, Daiichi-Sankyo, I-MAB, Jazz Pharmaceuticals, Merck, Oncolys BioPharma, Pieris Pharmaceuticals, Triumvira, and Zymeworks; has received consulting fees from Astellas, AstraZeneca, Bayer, Bristol Myers Squibb, Daiichi-Sankyo, I-MAB, Jazz Pharmaceuticals, Merck, Pieris Pharmaceuticals, and Zymeworks; and has received travel support from Dava Oncology and I-MAB. YMS and JHK have no conflicts to declare.
AcknowledgmentsThis study was supported by Jazz Pharmaceuticals and Zymeworks. We sincerely thank all patients and their families, all the investigators, clinical trial researchers, personnel, and staff who contributed to the trial. We also thank Jonathan Grim for his critical review of the manuscript and Chengzhi Xie for his work analysing the data. Medical writing support, under the direction of the authors, was provided by Can Huang and editorial support was provided by Claire Cartledge, both of CMC Connect, a division of IPG Health Medical Communications, with funding from Jazz Pharmaceuticals, in accordance with Good Publication Practice (GPP 2022) guidelines.
Supplementary MaterialSupplementary appendix
| | ||
| Age in years, median (IQR) | 58 (55–63) | |
| Sex * | ||
| Male | 39 (85%) | |
| Female | 7 (15%) | |
| Race * | ||
| White | 28 (61%) | |
| Asian | 17 (37%) | |
| Unknown | 1 (2%) | |
| Ethnicity * | ||
| Hispanic or Latino | 3 (7%) | |
| Not Hispanic or Latino | 43 (93%) | |
| ECOG performance status | ||
| 0 | 26 (57%) | |
| 1 | 20 (43%) | |
| Anatomical subtype | ||
| Gastric | 19 (41%) | |
| Gastro-oesophageal junction | 16 (35%) | |
| Oesophageal | 11 (24%) | |
| Stage at initial diagnosis | ||
| IB | 2 (4%) | |
| II | 2 (4%) | |
| III | 4 (9%) | |
| IV | 38 (83%) | |
| Measurable disease per RECIST v1.1 | 43 (93%) | |
| Previous neoadjuvant/adjuvant chemotherapy | 5 (11%) | |
| HER2 status (central or local assessment) † | ||
| IHC 0 and FISH negative | 1 (2%) | |
| IHC 1+ and FISH negative | 2 (4%) | |
| IHC 2+ and FISH negative | 1 (2%) | |
| IHC 2+ and FISH-positive | 5 (11%) | |
| IHC 3+ | 37 (80%) | |
| Centrally confirmed HER2-positive ‡ | 41 (89%) | |
| IHC 2+ and FISH-positive | 5 (11%) | |
| IHC 3+ | 36 (78%) | |
| | | ||
| | |||
| Confirmed objective response rate | 32 (76·2% [60·5–87·9]) | 31 (83·8% [68·0–93·8]) | |
| | |||
| Confirmed best overall response * | |||
| Complete response | 3 (7%) | 3 (8%) | |
| Partial response | 29 (69%) | 28 (76%) | |
| Stable disease | 5 (12%) | 4 (11%) | |
| Progressive disease | 5 (12%) | 2 (5%) | |
| Disease control rate † | 37 (88·1%; [74·4–96·0]) | 35 (94·6%; [81·8–99·3]) | |
| Clinical benefit rate ‡ | 33 (78·6% [63·2–89·7]) | 32 (86·5% [71·2–95·5]) | |
| Time to first confirmed response in months, median (IQR) | 1·3 (1·3–1·4) | 1·3 (1·3–1·4) | |
| Duration of response in months, § median (95% CI) | 18·7 (10·4–44·1) | 20·4 (8·3–44·1) | |
| Duration of response ≥12 months | 57% (37–72) | 59% (39–74) | |
| Duration of response ≥24 months | 40% (22–58) | 41% (22–59) | |
| | | | |
| Any | 16 (35%) | 25 (54%) | 5 (11%) |
| Diarrhoea | 25 (54%) | 18 (39%) | 0 |
| Nausea | 34 (74%) | 3 (7%) | 0 |
| Peripheral sensory neuropathy | 30 (65%) | 0 | 0 |
| Fatigue | 22 (48%) | 2 (4%) | 0 |
| Decreased appetite | 21 (46%) | 0 | 0 |
| Stomatitis | 13 (28%) | 0 | 0 |
| Vomiting | 12 (26%) | 4 (9%) | 0 |
| Anaemia | 11 (24%) | 1 (2%) | 0 |
| Dysgeusia | 11 (24%) | 0 | 0 |
| Infusion-related reactions | 10 (22%) | 0 | 0 |
| Hypomagnesaemia | 9 (20%) | 1 (2%) | 0 |
| Palmar-plantar erythrodysaesthesia syndrome | 9 (20%) | 1 (2%) | 0 |
| Decreased neutrophil count | 8 (17%) | 1 (2%) | 1 (2%) |
| Paraesthesia | 8 (17%) | 0 | 0 |
| Constipation | 7 (15%) | 0 | 0 |
| Decreased platelet count | 7 (15%) | 0 | 0 |
| Temperature intolerance | 7 (15%) | 0 | 0 |
| Dehydration | 6 (13%) | 0 | 0 |
| Increased aspartate aminotransferase | 6 (13%) | 0 | 0 |
| Pruritus | 6 (13%) | 0 | 0 |
| Rash | 6 (13%) | 0 | 0 |
| Weight decrease | 6 (13%) | 0 | 0 |
| Decreased white blood cell count | 5 (11%) | 1 (2%) | 1 (2%) |
| Increased alanine aminotransferase | 5 (11%) | 1 (2%) | 0 |
| Increased blood alkaline phosphatase | 5 (11%) | 0 | 0 |
| Dry skin | 5 (11%) | 0 | 0 |
| Exfoliative rash | 5 (11%) | 0 | 0 |
| Peripheral motor neuropathy | 5 (11%) | 0 | 0 |
| Hypokalaemia | 4 (9%) | 6 (13%) | 4 (9%) |
| Abdominal pain | 4 (9%) | 2 (4%) | 0 |
| Hypophosphatemia | 3 (7%) | 1 (2%) | 0 |
| Neutropenia | 3 (7%) | 1 (2%) | 0 |
| Tinnitus | 3 (7%) | 1 (2%) | 0 |
| Acute kidney injury | 1 (2%) | 2 (4%) | 0 |
| Deafness | 1 (2%) | 1 (2%) | 0 |
| Hypocalcaemia | 1 (2%) | 1 (2%) | 0 |
| Decreased lymphocyte count | 1 (2%) | 1 (2%) | 0 |
| Syncope | 0 | 1 (2%) | 0 |
| Upper gastrointestinal haemorrhage | 0 | 1 (2%) | 0 |
©2025. Elsevier Ltd