Managing zolbetuximab-induced nausea and vomiting: a proposal for a pragmatic approach in clinical practice

Abstract

Background

The global phase III SPOTLIGHT and GLOW trials confirmed the superiority of zolbetuximab plus chemotherapy over chemotherapy alone for claudin 18.2-positive advanced gastric cancer (AGC). However, severe nausea/vomiting during zolbetuximab infusion remains a significant safety concern. Our study aimed to evaluate the safety management strategy’s applicability in our institution.

Patients and methods

This registry-based observational study assessed the combination of fosnetupitant, dexamethasone, and 5-HT₃-receptor antagonist in patients receiving zolbetuximab plus chemotherapy. The initial zolbetuximab infusion rate was set at 75 ml/h for the first 60 min, and then 250 ml/h thereafter. The primary endpoint was the feasibility of our management and vomiting prevention during cycle 1.

Results

Among 21 patients (median age 61 years; male 48%; diffuse-type histology 81%; prior gastrectomy 38%; peritoneal metastases 57%), the incidence of nausea/vomiting during zolbetuximab initiation in cycle 1 was 62%/9.5%, with a 62% infusion interruption rate. The median time to first nausea was 93 min (range 77-127 min); median interruption time was 40 min (16-68 min); and median total infusion time of zolbetuximab was 257 min (154-343 min). All patients successfully completed zolbetuximab administration in cycle 1, with an 81% vomiting prevention rate. At cycle 2 day 1, 3 of 14 patients (21%) experienced nausea, with no vomiting.

Conclusion

First-line zolbetuximab plus chemotherapy was safely introduced to real-world patients with claudin 18.2-positive AGC, with effective management of nausea and vomiting, applicable in clinical practice.

Highlights

• •The management of zolbetuximab-induced gastrointestinal toxicity needs to be considered.
• •Our protocol includes antiemetic agents, practical dosing method, and patient instruction, with multidisciplinary discussion.
• •This observational study reported the results of 21 patients receiving zolbetuximab plus chemotherapy.
• •Our protocol is practical for daily use and beneficial in reducing zolbetuximab-induced vomiting during cycle 1.

Introduction

Advancements in treating advanced gastric or gastroesophageal junctional cancers (AGC) benefit not only patients with AGC overexpressing or amplifying human epidermal growth factor receptor 2 (HER2), but also those with HER2-negative AGC, including claudin 18 isoform 2 (CLDN18.2)-positive tumors.1, 2, 3 Zolbetuximab, a chimeric immunoglobulin G1 monoclonal antibody, targets CLDN18.2, overexpressed in 38%-44% of gastric cancer cells (defined as ≥75% of tumor cells showing moderate-to-strong membranous CLDN18 staining),^4,5 and exhibits antitumor efficacy by inducing antibody-dependent and complement-dependent cytotoxicity (ADCC and CDC).6, 7, 8

In the global phase III SPOTLIGHT and GLOW trials, zolbetuximab combined with mFOLFOX6 or CAPOX significantly prolonged overall survival (OS) [SPOTLIGHT: hazard ratio (HR) 0.75, 95% confidence interval (CI) 0.60-0.94, P = 0.005; GLOW: HR 0.77, 95% CI 0.62-0.96, P = 0.012] and progression-free survival in patients with untreated, metastatic, or recurrent CLDN18.2-positive AGC.^9,10 Zolbetuximab plus chemotherapy is expected to become the new standard-of-care first-line treatment for CLDN18.2-positive AGC. However, gastrointestinal toxicity, particularly nausea (any grade 81.0%/grade ≥3 16.1%) and vomiting (64.5%/16.1%), was more frequent in the zolbetuximab group than in the placebo group in both trials. Additionally, integrated analysis showed nausea and vomiting were more common in cycle 1 (nausea 62%; vomiting 52%) but decreased in later cycles.¹¹

CLDN18.2 is highly expressed in normal gastric mucosa.^6,7,12,13 One possible mechanism of zolbetuximab-induced nausea and vomiting was explored in an animal model.¹⁴ In ferrets, zolbetuximab administration immediately caused gastric mucosal distention and epithelial cell loss, suggesting on-target off-tumor toxicity. Notably, a computed tomography scan revealed submucosal stomach edema in a patient experiencing abdominal pain and nausea during zolbetuximab infusion (Figure 1).

Figure 1.

Representative CT scan images of a case with severe vomiting during zolbetuximab infusion in cycle 1. Panels A and B show a Borrmann type 3 gastric adenocarcinoma located at the lesser curvature of the stomach body. Panels C and D depict multiple liver metastases at initial diagnosis, before treatment. Panels E and F reveal submucosal edema of the entire stomach when zolbetuximab was infused, during which the patient experienced severe nausea and stomach pain in cycle 2. Interestingly, shrinkage of the multiple liver metastases was detected at cycle 2.

CT, computed tomography.

In the two trials, an 800 mg/m² loading dose of zolbetuximab was diluted with sterile 0.9% sodium chloride to a final concentration of 2 mg/ml and administered as a 2-h intravenous infusion. Following the recommended infusion rate, 96 (34%) and 52 (21%) patients in the SPOTLIGHT and GLOW trials, respectively, required infusion rate modification or zolbetuximab interruption due to nausea and vomiting during cycle 1. In the combined analysis of the SPOTLIGHT and GLOW trials, dose interruption was needed for 53.7% of the 533 patients receiving zolbetuximab plus chemotherapy.¹¹ Among those who experienced vomiting, the median initial dose rate was 176 ml/h in SPOTLIGHT and 147 ml/h in GLOW, compared with 100 ml/h and 90 ml/h, respectively, in patients without interruption. These findings suggest that nausea and vomiting may be linked to a faster zolbetuximab infusion rate. The product package insert for zolbetuximab in Japan reflects the protocol of these trials (Table 1). However, the stepwise adjustments and dosing method based on mg/m²/h result in varying infusion rates, complicating clinical application.

Table 1.

Product package insert recommendations for zolbetuximab infusion in Japan

			Infusion rate (ml/h)
			Initial 30-60 min	Afterward
BSA (m2)	Zolbetuximab (800 mg/m2)	Total amount (ml) (2.0 mg/ml)	100 mg/m2/h	200-400 mg/m2/h	Total infusion time (min)
1.0	800	400	50	100-200	165-270
1.1	880	440	55	110-220
1.2	960	480	60	120-240
1.3	1040	520	65	130-260
1.4	1120	560	70	140-280
1.5	1200	600	75	150-300
1.6	1280	640	80	160-320
1.7	1360	680	85	170-340
1.8	1440	720	90	180-360
1.9	1520	760	95	190-380
2.0	1600	800	100	200-400

BSA, body surface area.

Prophylactic antiemetics, including a 5-HT₃ receptor blocker, a neurokinin-1 (NK1) receptor blocker, and a corticosteroid, were recommended in both trials and administered according to institutional standards or published guidelines. However, corticosteroids were advised to be avoided, if possible, during the first dose of zolbetuximab/placebo to avoid reducing ADCC by corticosteroids.

Approximately 20% of patients discontinued treatment due to zolbetuximab-related adverse events (AEs) in the SPOTLIGHT and GLOW trials.^9,10 In clinical practice, preventing severe gastrointestinal toxicity, especially during cycle 1, is crucial for maximizing zolbetuximab efficacy while maintaining tolerability. The changes in certain quality-of-life measures appeared similar between patients treated with zolbetuximab and those receiving a placebo in the FAST trial and combined analysis of the SPOTLIGHT and GLOW trials.^11,15 However, management strategies, including the use of a multidisciplinary approach, require further improvement. Based on our experience from the SPOTLIGHT and GLOW trials, we developed a practical dosing method, including a two-stage fixed infusion rate for the loading dose and a management protocol. Our study aimed to explore whether our protocol for zolbetuximab administration is practical for daily use and assess its utility in reducing gastrointestinal toxicity during cycle 1.

Patients and methods

Study design

This single-institution, registry-based study was conducted at the Cancer Institute Hospital of the Japanese Foundation for Cancer Research. The inclusion criteria were (i) patients with unresectable or metastatic AGC, (ii) histologically confirmed adenocarcinoma, (iii) confirmed CLDN18.2-positive status [defined as ≥75% of tumor cells showing moderate-to-strong membranous CLDN18 staining using the VENTANA CLDN18 (43-14A) RxDx Assay], and (iv) patients who received zolbetuximab plus CAPOX or mFOLFOX6 as first-line treatment. The exclusion criteria were (i) patients receiving other chemotherapies and (ii) those with other advanced cancers.

Baseline characteristics and treatment-related AEs of any grade, according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0, were collected from medical records. This study was approved by the Institutional Review Board of the Cancer Institute Hospital of the Japanese Foundation for Cancer Research (Tokyo, Japan; approval date: August 2024; registry number: 2024-GB-095) and conducted in accordance with the Declaration of Helsinki 1964 and its later versions. Informed consent was waived, with an opportunity to opt out of the research.

Treatment plan

A 5-HT3 receptor inhibitor (palonosetron hydrochloride 0.75 mg) was administered intravenously, along with an NK1 receptor inhibitor (fosnetupitant 235 mg) and dexamethasone 9.9 mg intravenously. Oral dexamethasone (4 mg) was also given on days 2 and 3. Olanzapine 5 mg was optionally prescribed by the attending physician for 5 days, starting the evening of the initial zolbetuximab administration. H₁ (polaramine 5 mg) and H₂ (famotidine 20 mg) blockers were mandatory as premedication with the NK1 and 5-HT3 receptor inhibitors (Supplementary Figure S1, available at https://doi.org/10.1016/j.esmogo.2024.100128). Patients were hospitalized for at least 5 days to monitor AEs.

Backbone chemotherapy of CAPOX (capecitabine 1000 mg/m² twice daily on days 1-14 and oxaliplatin 130 mg/m² on day 1, every 3 weeks) or mFOLFOX6 (leucovorin 400 mg/m² on day 1, fluorouracil 400 mg/m² on day 1, 2400 mg/m² in a 46-h infusion, and oxaliplatin 85 mg/m² on day 1, every 2 weeks) was determined by the attending physician (Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2024.100128). The loading dose of zolbetuximab (800 mg/m²) was administered intravenously at an infusion rate of 75 ml/h for the first 60 min, then increased to 250 ml/h until fully infused (Figure 2A and B). To minimize infusion-related incidents, the rate was fixed regardless of patients’ body surface area (BSA), based on the product package insert for patients with a BSA of 1.5 m². If patients experienced intolerable nausea or vomiting, they were instructed to notify health care providers. Zolbetuximab infusion was then interrupted for 30 min, and additional antiemetic agents were permitted (Figure 2C). After symptoms improved, zolbetuximab was restarted at 50% of the previous rate (125 ml/h). If nausea did not recur, the infusion rate was gradually increased to 75% (188 ml/h) and then 100% (250 ml/h) every 30 min.

Figure 2.

Protocol for initiating zolbetuximab at our institution. Panel A shows the standard infusion schedule for each drug. Panel B illustrates the planned infusion schedule for zolbetuximab loading at cycle 1 (800 mg/m²) and subsequent doses (400 or 600 mg/m²). Panel C presents the protocol for interruption and resumption of zolbetuximab infusion in cycle 1. 5-FU, 5-fluorouracil; 5-HT3, 5-HT3 receptor inhibitor; DEX, dexamethasone; H2RA, histamine type 2 receptor antagonist; NK-1, neurokinin-1 receptor inhibitor. ^aThe total infusion time varies per patient at cycle 1. ^bThe infusion rate after 30 min is calculated per patient at subsequent cycles. ^cIf emesis is tolerable at cycle 1 and 2, the infusion rate will be modified to administer within 2 h.

From the second cycle onward, zolbetuximab is generally administered on an outpatient basis. Subsequent doses of 600 mg/m² every 3 weeks (with CAPOX) or 400 mg/m² every 2 weeks (with mFOLFOX6) were planned to be administered as follows: 40 or 60 ml/h for the first 30 min, followed by an increased rate over a 2-h infusion, provided no toxicity occurred (Supplementary Figure S2, available at https://doi.org/10.1016/j.esmogo.2024.100128). If the first and second administrations of zolbetuximab were well tolerated, subsequent cycles would be administered over a total of 120 min for convenience.

Assessment of nausea and vomiting during cycle 1

The primary endpoints of the study were the feasibility of our protocol and the rate of vomiting prevention during cycle 1, defined as the absence of emetic events with or without the use of rescue medication during the first cycle of zolbetuximab plus chemotherapy. The frequency of vomiting, the presence and severity of nausea, and appetite loss were recorded daily using a patient questionnaire and evaluated according to CTCAE ver. 5.0. Information on AEs during zolbetuximab infusion on day 1 was collected separately.

Statistical analyses

To evaluate patient characteristics, summary statistics were generated using frequencies and proportions for categorical variables and medians and ranges for continuous variables. A univariate analysis was conducted to explore risk factors for nausea and vomiting. These analyses were descriptive, as no formal hypotheses were tested. All statistical analyses were carried out using JMP version 17.0.0 software (SAS Institute, Cary, NC).

Results

Patients’ characteristics

In total, 21 consecutive patients with HER2-negative and CLDN18.2-positive AGC received zolbetuximab plus chemotherapy as first-line treatment. Patient characteristics are detailed in Table 2. Overall, 48% of the patients were male, with a median age of 61 years (range 30-81 years) at treatment initiation. Primary tumors were located at the gastroesophageal junction in 29% of patients, and diffuse-type histology was common (81%). Additionally, 38% of patients had undergone gastrectomy before chemotherapy (total gastrectomy, n = 4; distal gastrectomy, n = 2; proximal gastrectomy, n = 1; subtotal esophagectomy, n = 1). The most common sites of metastasis were the peritoneum (57%), lymph nodes (29%), and liver (10%). CAPOX was the most frequently used backbone regimen (62%). Overall, 12 (57%) patients received 5 mg of olanzapine daily for 5 consecutive days starting from day.

Table 2.

Patient characteristics

Characteristics (N = 21)	n (%)
Age, median (range)	61 (30-81)
Sex, male	10 (48%)
Disease status
Recurrent	7 (33%)
Metastatic	14 (67%)
Prior gastrectomy
Yes	8 (38%)
Total gastrectomy	4 (19%)
Distal gastrectomy	2 (10%)
Othersa	2 (10%)
No	13 (62%)
Primary tumor site
EGJ	6 (29%)
Stomach	15 (71%)
Borrmann type
Type 4	11 (52%)
Others	10 (48%)
Histological type
Intestinal	4 (19%)
Diffuse	17 (81%)
ECOG PS
0	11 (52%)
1	10 (48%)
Adequate oral intake
Yes	19 (90%)
No	2 (10%)
Metastatic site
Peritoneum	12 (57%)
Liver	2 (10%)
Lymph node	6 (29%)
No. of Metastases
0-1	18 (86%)
≥2	3 (14%)
PD-L1 CPS
<1	5 (24%)
1≤ CPS <5	12 (57%)
≥5	3 (14%)
Not evaluated	1 (5%)
PD-L1 TPS
<1	17 (81%)
≥1	2 (10%)
Not evaluated	2 (10%)
Backbone chemotherapy
CAPOX	13 (62%)
mFOLFOX6	8 (38%)
BSA, median (range) (m2)	1.53 (1.28-1.91)
Dosage of zolbetuximab, median (range) (mg)	1200 (694-1512)
Dose reduction of chemotherapy at cycle 1
Yes	10 (48%)
No	11 (52%)

BSA, body surface area; CAPOX, capecitabine plus oxaliplatin regimen; CPS, combined positive score; ECOG, Eastern Cooperative Oncology Group; EGJ, esophagogastric junction; mFOLFOX6, modified folinic acid plus fluorouracil plus oxaliplatin regimen; PS, performance status; TPS, tumor proportion score.

^aOthers include subtotal esophagectomy (n = 1) and proximal gastrectomy (n = 1).

Safety outcomes

Nausea and/or vomiting during zolbetuximab infusion at cycle 1 day 1

The actual treatments administered during zolbetuximab infusion for the 21 patients are shown in Figure 3. Primary dose reduction of chemotherapy in the first cycle was required for 11 (52%) patients, though all 21 patients received the full 800 mg/m² loading dose of zolbetuximab. No patients experienced nausea or vomiting during the initial 60 min when the infusion rate was 75 ml/h. However, 13 (62%) patients experienced intolerable nausea during the infusion at a rate of 250 ml/h, leading to interruptions. Additional medications were administered to 12 (57%) patients, and subsequently 11 of the 13 patients (85%) were able to resume the infusion at the full rate without further interruptions. Notably, none of the four patients who had undergone prior total gastrectomy experienced nausea or vomiting during the zolbetuximab infusion. Prior gastrectomy (P = 0.0006) and disease status (recurrent versus metastatic) (P = 0.02) were significantly associated with the occurrence of nausea, while no baseline characteristics were statistically associated with the occurrence of vomiting.

Figure 3.

Overview of treatment for 21 patients during the first initiation of zolbetuximab.

BSA, body surface area; DG, distal gastrectomy; F, female; M, male; TG, total gastrectomy.

Table 3 summarizes the time to each outcome related to zolbetuximab infusion. The median time to the first occurrence of nausea was 93 min (range 77-127 min), the median interruption time was 40 min (range 16-68 min), the median total infusion time was 257 min (range 154-343 min), and the median actual infusion time was 224 min (range 154-275 min).

Table 3.

Zolbetuximab infusion-related outcomes during cycle 1

Time	Time (min)
Time of first occurrence of nausea, median (min-max)	93 (77-127)
Time of interruption of zolbetuximab (min-max)	40 (16-68)
Total infusion time, median (min-max)	257 (154-343)
Actual infusion time, median (min-max)	224 (154-275)

The overall rate of vomiting prevention during cycle 1 was 81% (95% CI 59.9% to 92.3%). The most common all-grade AEs during cycle 1 were nausea (81%), decreased appetite (57%), and vomiting (19%). Grade ≥3 AEs occurred in 52% of patients, including decreased appetite (n = 4), nausea (n = 3), hypoalbuminemia (n = 2), catheter-related infection (n = 1), and decreased neutrophil count (n = 1). The use of olanzapine was insignificantly associated with the incidence of decreased appetite of any grade (56% versus 58%, P = 0.89).

Nausea and/or vomiting during zolbetuximab infusion at cycle 2 day 1

At data cut-off (10 October 2024), a total of 14 patients had received cycle 2 treatment. Two patients experienced grade 3 hypoalbuminemia and discontinued zolbetuximab for cycle 2. Overall, gastrointestinal toxicities were moderate, with three patients (21%) requiring infusion interruptions, but no patients experienced vomiting.

Discussion

To our knowledge, this is the first study to evaluate the safety of introducing zolbetuximab plus chemotherapy for real-world patients with CLDN18.2-positive AGC in a practical setting. Managing zolbetuximab plus chemotherapy is crucial, as early discontinuation may reduce its clinical benefits for these patients. In our pilot study, management of vomiting during zolbetuximab infusion was favorable, with vomiting incidences of 9.5% during infusion and 19% during cycle 1. Although direct comparisons require caution, vomiting during initial zolbetuximab or placebo infusions occurred in 51.6% and 17.5% of patients in the SPOTLIGHT/GLOW trials,¹¹ suggesting that our protocol effectively reduced vomiting to a level comparable to chemotherapy alone. The median total infusion time of zolbetuximab was 257 min, which tended to be shorter than that in the two trials. While 62% of patients required interruptions due to intolerable nausea, all successfully completed the initial zolbetuximab infusion, and nausea incidence was moderate, especially among patients with prior gastrectomy, especially total gastrectomy. In cycle 2, zolbetuximab infusion was well tolerated, with a 0% vomiting rate, compared with 26% in the SPOTLIGHT/GLOW trials.¹¹

Our management strategy is practical and applicable for daily practice. Firstly, we standardized the infusion rate for the initial administration of zolbetuximab based on a BSA of 1.5 m², which is the average BSA for Japanese patients with AGC. This rate generally falls within the range recommended for patients with a BSA of 1.3-1.8 m², as outlined in the product package insert. Secondly, the protocol for dose escalation is simplified, even after interruptions, which may help prevent medical errors. Most importantly, our protocol allows early interruption, regardless of nausea severity. The incidence of zolbetuximab interruption was 62%, compared with 53.7% in the combined analysis of the SPOTLIGHT and GLOW trials.¹¹ Because nausea is subjective, its severity may differ between patient self-reports and health care provider evaluations.¹⁶ Zolbetuximab-induced nausea and vomiting often occur abruptly, making the prompt assessment and management of these symptoms crucial. Therefore, patients were instructed to notify health care providers if they experienced moderate nausea. Early, temporary interruptions may have alleviated symptoms and prevented emesis, thereby helping maintain or improve the quality of life in patients undergoing gastric cancer treatment.¹⁷ Additionally, our protocol regulates infusion rate escalation to 100% after a zolbetuximab interruption. In 85% of cases requiring interruption, patients successfully resumed zolbetuximab at the full infusion rate without further interruptions due to intolerable nausea or vomiting. The reasons why nausea and vomiting did not recur after a 30-min interruption should be explored in future studies. Antiemetic prophylaxis was planned according to guidelines for high emetic risk, as recommended by the Japan Society of Clinical Oncology.¹⁸ We used NK-1 and 5-HT3 receptor inhibitors, and optionally olanzapine.19, 20, 21 H₁ and H₂ blockers were mandatory to prevent oxaliplatin-related allergies, zolbetuximab infusion reactions, and gastrointestinal toxicity. Dexamethasone was routinely given, as corticosteroids may effectively prevent nausea and vomiting without reducing clinical benefits.²² Collectively, these management strategies likely contributed to the low incidence of zolbetuximab-induced vomiting.

Our management strategy also highlighted some issues that need resolution. Firstly, other AEs like peripheral edema and decreased appetite should be considered, as these symptoms negatively impact the quality of life for AGC patients.¹⁷ In fact, two patients experienced grade 3 hypoalbuminemia and grade 2 peripheral edema, requiring temporary discontinuation of zolbetuximab during cycle 2. The decreased appetite may result from the on-target off-tumor toxicity of zolbetuximab; however, oxaliplatin and fluoropyrimidine could also contribute to gastrointestinal toxicity. Although the effect of olanzapine for nausea and vomiting prevention was limited in ferret studies,¹⁴ its use might improve appetite in patients receiving zolbetuximab.²³ In this study, the use of olanzapine for 5 days was not statistically associated with the incidence of decreased appetite. Therefore, further studies are needed to evaluate the efficacy of olanzapine in alleviating gastrointestinal toxicities, including decreased appetite, during treatment with zolbetuximab plus chemotherapy. Additionally, it is important to inform patients before treatment that severe nausea and vomiting related to zolbetuximab are typically confined to the initial administration period, and continuing zolbetuximab may enhance its clinical benefit. From a pharmacist’s perspective, coring of rubber stoppers is a key issue, and the difficulty of dissolving zolbetuximab should be noted by physicians. Most importantly, our management plan should be tailored to each hospital’s circumstances, taking into account health care worker resources. This study emphasizes the importance of a comprehensive approach to introducing zolbetuximab that is practical for daily use, with collaboration among pharmacists, nurses, and doctors.

This study has some limitations. Firstly, the small sample size prevents us from drawing definitive conclusions; however, our findings suggest that appropriate management of zolbetuximab infusion may reduce the risk of nausea and/or vomiting. Secondly, key efficacy parameters, such as OS, were not evaluated by the data cut-off and will be reported in future studies. Thirdly, practical approaches should be tailored to each institution’s circumstances, so our management strategy may not be suitable for all hospitals. Despite these limitations, we believe this study will help clinicians introduce zolbetuximab plus chemotherapy safely and effectively, maximizing its benefit for patients with CLDN18.2-positive AGC.

In conclusion, the practical management of zolbetuximab and chemotherapy was safely applied to real-world patients with CLDN18.2-positive AGC using a multidisciplinary approach. Early zolbetuximab interruption may help prevent vomiting during initial treatment with zolbetuximab plus chemotherapy.

Funding

None declared.