Impact and effective management of nausea/vomiting on patients treated with zolbetuximab + chemotherapy: insights from the phase III SPOTLIGHT and GLOW studies

Abstract

Background

In the phase III SPOTLIGHT (NCT03504397) and GLOW (NCT03653507) trials, nausea and/or vomiting were the most common adverse events in patients receiving zolbetuximab plus chemotherapy. We further characterize the adverse effects of nausea/vomiting on treatment adherence and efficacy, and evaluate strategies for managing nausea/vomiting through the SPOTLIGHT and GLOW combined analysis.

Materials and methods

Ad hoc exploratory analyses included treatment discontinuations, antiemetic use, and progression-free survival (PFS) and overall survival (OS) after censoring patients with inadequate dose exposure (<75% relative exposure intensity) or early discontinuation (within 63 days) of zolbetuximab or placebo initiation.

Results

This combined analysis included 1072 patients. Nausea and/or vomiting (nausea/vomiting) were most common on the first zolbetuximab infusion day [nausea: 46.5% (246/529); vomiting: 41.0% (217/529)]; on the second infusion day, nausea dropped to 23.2% (97/418) and vomiting dropped to 24.4% (102/418). Nausea/vomiting continued to decline on subsequent infusion days. More patients on zolbetuximab versus placebo had nausea/vomiting leading to inadequate dose exposure or early discontinuation [9.9% (53/537) versus 0.6% (3/535)]. More white than Asian patients had nausea/vomiting leading to inadequate dose exposure or early discontinuation with zolbetuximab [15.4% (36/234) versus 4.7% (12/254)] and placebo [0.9% (2/224) versus 0.4% (1/255)]. Before censoring, median PFS (mPFS) (hazard ratio) for zolbetuximab versus placebo was 9.2 versus 8.2 months (0.71), and median OS (mOS) was 16.4 versus 13.7 months (0.77). After censoring, mPFS was 10.4 versus 8.2 months (0.65), and mOS was 17.9 versus 13.7 months (0.69). On the first zolbetuximab infusion day, patients without vomiting had higher utilization of three-drug prophylactic regimens than those with vomiting [75.3% (73/97) versus 24.7% (24/97)]. In patients receiving steroids, including as antiemetic prophylaxis, mPFS was 10.5 months with zolbetuximab versus 8.3 months with placebo (0.66), and mOS was 18.4 versus 13.8 months (0.71).

Conclusion

Optimal antiemetic prophylaxis, with >3 drug classes including steroids, for zolbetuximab-associated nausea/vomiting is important for treatment adherence and maximal clinical benefits.

Graphical abstract

Highlights

• •Nausea/vomiting was most prevalent on day 1 of the first zolbetuximab infusion and decreased on subsequent infusion days.
• •Patients whose nausea/vomiting caused dose interruption or early discontinuation received less drug and had shorter PFS/OS.
• •Patients who received zolbetuximab with steroids had longer PFS and OS than those who did not receive steroids.
• •Using ≥3-drug prophylactic antiemetics, including steroids, and modifying infusion rates may help reduce nausea/vomiting.
• •Managing nausea/vomiting is key for treatment adherence to maximize dose exposure and may lead to improved outcomes.

Introduction

Gastric cancer is a leading cause of cancer deaths worldwide.¹ Claudin 18 isoform 2 (CLDN18.2), a tight junction protein expressed exclusively in gastric mucosa cells in normal tissue that is often retained in gastric and gastroesophageal junction (G/GEJ) adenocarcinomas, is a validated therapeutic target.2, 3, 4, 5, 6, 7 Zolbetuximab is a first-in-class immunoglobulin G1 monoclonal antibody targeting CLDN18.2 that mediates death of CLDN18.2-positive G/GEJ adenocarcinoma cells via antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity.^4,6,7 In the phase III SPOTLIGHT (NCT03504397) and GLOW (NCT03653507) trials, and the recent combined analysis of both trials, first-line zolbetuximab plus chemotherapy improved progression-free survival (PFS) and overall survival (OS) compared with placebo plus chemotherapy in patients with human epidermal growth factor receptor 2 (HER2)-negative, locally advanced unresectable or metastatic G/GEJ adenocarcinoma whose tumors were CLDN18.2-positive.6, 7, 8

Nausea and/or vomiting (nausea/vomiting) were the most common treatment-emergent adverse events (TEAEs) in both SPOTLIGHT and GLOW.6, 7, 8 With zolbetuximab in SPOTLIGHT and GLOW, approximately 82% and 69% of patients experienced any-grade nausea, respectively; 67% and 66% of patients experienced any-grade vomiting, respectively.^6,7 In comparison, with placebo, 61% and 50% of patients experienced any-grade nausea, respectively; 36% and 31% of patients experienced any-grade vomiting, respectively.^6,7 Nausea and vomiting led to higher rates of discontinuations with zolbetuximab than with placebo.^6,7 Further characterization of how nausea/vomiting and its management may affect zolbetuximab efficacy is important to improve clinical outcomes, as well as the patient and health care professional experience.

Here, we analyzed zolbetuximab-associated nausea/vomiting to determine whether nausea/vomiting-related inadequate dose exposure or early discontinuation impacted patient outcomes. We also evaluated strategies to manage nausea/vomiting and optimize clinical benefits for patients.

Materials and methods

Patients and treatment in SPOTLIGHT and GLOW

Full details of SPOTLIGHT, GLOW, and their combined analysis were published previously.6, 7, 8 Briefly, eligible patients were adults with HER2-negative, previously untreated, locally advanced unresectable or metastatic G/GEJ adenocarcinoma whose tumors were CLDN18.2-positive, defined as ≥75% of tumor cells demonstrating moderate-to-strong membranous CLDN18 staining using the VENTANA CLDN18 (43-14A) RxDx Assay (Roche Diagnostics Solutions, Tucson, AZ).6, 7, 8 Patients received zolbetuximab 800 mg/m² [cycle (C) 1, day (D) 1] followed by 600 mg/m² (every 3 weeks) plus chemotherapy [modified folinic acid (or levofolinate), fluorouracil, and oxaliplatin (mFOLFOX6, SPOTLIGHT) or capecitabine and oxaliplatin (CAPOX, GLOW) regimens], or placebo plus chemotherapy (mFOLFOX6, SPOTLIGHT or CAPOX, GLOW regimens).^6,7 In SPOTLIGHT, zolbetuximab was administered on D1 and D22 (every 3 weeks), and mFOLFOX6 was administered on D1, D15, and D29 (every 2 weeks) of each 42-day treatment cycle. In GLOW, zolbetuximab was administered on D1 (every 3 weeks), and CAPOX was administered on D1-D14 of each 21-day treatment cycle.

Nausea/vomiting

Adverse events (AEs), including nausea/vomiting, were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03 and were monitored throughout the trials and for 90 days after treatment discontinuation. Characterization of nausea/vomiting includes additional ad hoc assessments of incidence and timing of nausea/vomiting such as occurrence of nausea/vomiting on the day of zolbetuximab or placebo infusion, time from initiation of zolbetuximab or placebo infusion to the first episode of nausea/vomiting, the effect of infusion rates on nausea/vomiting during the first zolbetuximab infusion, incidence of nausea/vomiting leading to inadequate dose exposure or early discontinuation, nausea/vomiting in white versus Asian subgroups and gastric adenocarcinoma (GC) or GEJ subgroups, and the association of antiemetics with occurrences of nausea/vomiting.

Additional exploratory ad hoc analyses of nausea/vomiting leading to treatment discontinuation stratified by prior gastrectomy are reported in the Supplementary Results and Tables S8, available at https://doi.org/10.1016/j.esmoop.2025.105931.

Inadequate dose exposure or early discontinuation

Inadequate dose exposure was defined as a dose interruption resulting in <75% relative exposure intensity (REI) of zolbetuximab or placebo. REI was calculated as follows:

• •REI = (actual cumulative dose/planned cumulative dose for duration of treatment) × 100%
• •Planned cumulative dose = 800 + 600 × (number of planned doses - 1)
• •Number of planned dosing = duration of zolbetuximab/21-day dosing interval, then rounded up to the nearest whole number

Early discontinuation was defined as the withdrawal of zolbetuximab or placebo within 63 days from D1 of infusion.⁹ Owing to the fact that zolbetuximab was administered every 3 weeks (21 days), with tumor assessment occurring every 9 weeks ± 7 days from cycle 1 day 1 (C1D1), early discontinuation was defined as a period of <63 days, before or by the first scan.

Efficacy

Ad hoc efficacy analyses included PFS and OS after censoring patients with nausea/vomiting leading to inadequate dose exposure or early discontinuation and after stratifying by use of steroids to evaluate the impact of these factors on clinical outcomes. PFS per RECIST version 1.1 by an independent review committee was the primary endpoint in SPOTLIGHT and GLOW. OS was a key secondary endpoint in SPOTLIGHT and GLOW.

Antiemetics in SPOTLIGHT and GLOW

In both SPOTLIGHT and GLOW, the use of neurokinin-1 (NK-1) and serotonin (5-HT3) receptor antagonists and other prophylactic antiemetics was recommended per institutional and clinical practice guidelines. Avoidance or minimization of steroids, particularly for the first zolbetuximab/placebo treatment, was advised; however, steroid use was not prohibited.

Statistical analysis

Statistical methods were published previously.6, 7, 8 Efficacy was assessed in the full analysis set, which comprised all randomly assigned patients. PFS and OS were compared between treatment groups using log-rank test stratified by region (Asia versus non-Asia), number of organs with metastatic sites (0-2 versus ≥3), and prior gastrectomy (yes versus no). Additional ad hoc exploratory analyses were conducted to compare outcomes stratified by race (white versus Asian) and tumor location (GC or GEJ). Safety was assessed in the safety analysis set, which comprised all patients who received at least one dose of study drug. Statistical analysis was descriptive in nature and additional P values that were calculated in subgroup comparisons were also descriptive. All statistical tests were two-sided, and missing values were not imputed. Categorical variables were summarized by frequency distributions (number, percentage of patients, and missing values); continuous variables were summarized by number of patients (n), mean, standard deviation, median, first and third quartiles, minimum, maximum, and missing values. Time-to-event endpoints were summarized using Kaplan–Meier methods.

Results

All results represent the combined analysis of SPOTLIGHT and GLOW unless otherwise specified. The data cut-off date in SPOTLIGHT was 8 September 2023. The data cut-off date in GLOW was 12 January 2024. For patient characteristics and individual SPOTLIGHT and GLOW datasets, see Supplementary Results, available at https://doi.org/10.1016/j.esmoop.2025.105931.

Incidence and timing of nausea/vomiting

Nausea/vomiting was most common on the first infusion day of zolbetuximab and generally decreased on infusion days 2-4 (Figure 1, Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2025.105931). On the first zolbetuximab infusion day (C1D1), out of all nausea/vomiting events, 89.4% (220/246) of nausea events and 81.1% (176/217) of vomiting events were grade 1-2, and no grade 4 nausea or vomiting events occurred. At C1D1, 46.5% (246/529) of patients experienced nausea and 41.0% (217/529) of patients experienced vomiting. On the second zolbetuximab infusion day (SPOTLIGHT: C1D22; GLOW: C2D1), 23.2% (97/418) of patients experienced nausea and 24.4% (102/418) of patients experienced vomiting. On the first infusion day for the placebo group (C1D1), 12.0% (63/527) of patients experienced nausea and 3.2% (17/527) of patients experienced vomiting. On the second infusion day for the placebo group (SPOTLIGHT: C1D15; GLOW: C2D1), 6.9% (31/447) of patients experienced nausea and 2.2% (10/447) of patients experienced vomiting. In a subset of patients with available data, median time to the first occurrence of nausea/vomiting on C1D1 was 50.5 min after starting the first zolbetuximab infusion (range 0-371 min, n = 90) and 170.5 min after starting the first infusion for placebo (range 0-510 min, n = 4; Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2025.105931).

Figure 1.

Occurrences of nausea or vomiting on the day of infusion. Infusions 1-4 in patients who received (A) zolbetuximab or (B) placebo. Data represent the combined analysis of SPOTLIGHT and GLOW.

During the first zolbetuximab infusion, patients who underwent infusion modifications due to AEs had a faster initial infusion rate than patients without infusion modifications (mean/median, SPOTLIGHT: 221/176 versus 167/100 ml/h; GLOW: 194/147 versus 126/90 ml/h; Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2025.105931). In SPOTLIGHT, 85% (82/96) of patients who underwent infusion modifications experienced any-grade nausea/vomiting. In GLOW, 79% (41/52) of patients who underwent infusion modifications experienced any-grade nausea/vomiting. The duration of infusion interruptions and proportion of patients with infusion interruptions were highest during the first zolbetuximab infusion and generally decreased thereafter (Supplementary Table S3, available at https://doi.org/10.1016/j.esmoop.2025.105931). On C1D1, 32% (87/271) of patients in SPOTLIGHT experienced infusion interruptions with a median time of 22 min for interruption, and 27% (67/250) of patients in GLOW experienced infusion interruptions with a median time of 16 min for interruption. Infusion time was longer in early cycles and generally shortened with each subsequent cycle.

Treatment discontinuation due to nausea/vomiting

A higher proportion of patients receiving zolbetuximab had nausea/vomiting leading to inadequate dose exposure or early discontinuation than those receiving placebo [9.9% (53/537) versus 0.6% (3/535)]. A higher proportion of patients receiving zolbetuximab discontinued treatment at any time during the study period due to nausea/vomiting than those receiving placebo [nausea: 3.5% (18/509) versus 0.4% (2/493); vomiting: 3.7% (19/509) versus 0.6% (3/493); Supplementary Table S4, available at https://doi.org/10.1016/j.esmoop.2025.105931].

Correlation of nausea/vomiting with efficacy

In the combined analysis of SPOTLIGHT and GLOW, PFS and OS were longer with zolbetuximab than with placebo. Median PFS (mPFS) was 9.2 months [95% confidence interval (CI) 8.4-10.4 months] with zolbetuximab and 8.2 months (95% CI 7.6-8.4) with placebo [hazard ratio (HR) 0.71, 95% CI 0.61-0.83]. Median OS (mOS) was 16.4 months (95% CI 15.0-17.9 months) with zolbetuximab and 13.7 months (95% CI 12.3-15.3 months) with placebo (HR 0.77, 95% CI 0.67-0.89).⁸

Considering the potential effects of nausea/vomiting leading to inadequate dose exposure or early discontinuation on subsequent risk of suboptimal zolbetuximab exposure, PFS and OS were assessed after censoring patients with nausea/vomiting that led to inadequate dose exposure or early discontinuation with zolbetuximab [9.9% (53/537)] or placebo [0.6% (3/535); Figure 2]. The exploratory ad hoc analysis showed that mPFS after censoring was 10.4 months (95% CI 8.8-12.2 months) with zolbetuximab and 8.2 months (95% CI 7.7-8.4 months) with placebo (HR 0.65, 95% CI 0.56-0.76); mOS after censoring was 17.9 months (95% CI 16.4-19.5 months) with zolbetuximab and 13.7 months (95% CI 12.4-15.3 months) with placebo (HR 0.69, 95% CI 0.60-0.80). With zolbetuximab, mPFS after censoring was 1.2 months longer than before censoring (10.4 versus 9.2 months) and mOS after censoring was 1.5 months longer than before censoring (17.9 versus 16.4 months), suggesting that patients who did not experience inadequate dose exposure or early discontinuation had numerically longer PFS and OS.

Figure 2.

Kaplan–Meier curves after censoring patients with nausea/vomiting. (A) PFS and (B) OS after censoring patients with nausea/vomiting leading to inadequate dose exposure or early discontinuation.^a Data represent the combined analysis of SPOTLIGHT and GLOW.

CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; REI, relative exposure index. ^aNausea/vomiting leading to dose interruption resulting in <75% REI of zolbetuximab or placebo, or nausea/vomiting leading to withdrawal of zolbetuximab or placebo treatment within 63 days.

PFS and OS after censoring patients with nausea/vomiting leading to inadequate dose exposure or early discontinuation are presented in Supplementary Figures S2 and S3, available at https://doi.org/10.1016/j.esmoop.2025.105931, for SPOTLIGHT and GLOW trials, respectively.

Treatment discontinuation due to nausea/vomiting stratified by race (white versus Asian) and tumor location (GC versus GEJ)

With zolbetuximab, the proportion of white patients with nausea/vomiting leading to inadequate dose exposure or early discontinuation was higher than that of the overall population and Asian population [white: 15.4% (36/234); overall: 9.9% (53/537); Asian: 4.7% (12/254)]. With placebo, the proportion of white patients with nausea/vomiting leading to inadequate dose exposure or early discontinuation was similar to the overall population and Asian population [0.9% (2/224); 0.6% (3/535); 0.4% (1/255), respectively].

With zolbetuximab, higher proportions of white patients than Asian patients discontinued treatment at any time during the study period due to nausea/vomiting [nausea: 3.9% (9/232) versus 2.4% (6/253); vomiting: 4.7% (11/232) versus 2.0% (5/253); Supplementary Table S4, available at https://doi.org/10.1016/j.esmoop.2025.105931]. With placebo, similar proportions of white and Asian patients discontinued treatment at any time during the study period due to nausea/vomiting [nausea: 0% (0/222) and 0.8% (2/250), respectively; vomiting: 0.5% (1/222) and 0.8% (2/250), respectively; Supplementary Table S4, available at https://doi.org/10.1016/j.esmoop.2025.105931].

See Supplementary Results and Figure S4, available at https://doi.org/10.1016/j.esmoop.2025.105931, for correlation of nausea/vomiting with efficacy stratified by race (white versus Asian).

As published previously,⁸ patients with GC or GEJ adenocarcinoma were well balanced between treatment groups. With zolbetuximab, 81.6% (438/537) of patients had GC and 18.4% (99/537) of patients had GEJ adenocarcinoma. With placebo, 78.3% (419/535) of patients had GC and 21.7% (116/535) of patients had GEJ adenocarcinoma.⁸ However, within those who received zolbetuximab, a lower proportion of white patients had GC compared with Asian patients with GC, whereas a higher proportion of white patients had GEJ adenocarcinoma compared with Asian patients with GEJ adenocarcinoma. Of those with GC who received zolbetuximab, 40.6% (178/438) were white and 51.8% (227/438) were Asian. Of those with GEJ adenocarcinoma who received zolbetuximab, 56.6% (56/99) were white and 27.3% (27/99) were Asian. Supplementary Table S4, available at https://doi.org/10.1016/j.esmoop.2025.105931, shows that a higher proportion of patients receiving zolbetuximab compared with placebo (regardless of race or tumor location) discontinued treatment at any time during the study period due to nausea/vomiting. The ad hoc analysis stratified by race and tumor location (Supplementary Table S5, available at https://doi.org/10.1016/j.esmoop.2025.105931) showed that, among patients receiving zolbetuximab, a higher proportion of white patients with GC discontinued treatment due to nausea/vomiting compared with Asian patients with GC, and a higher proportion of white patients with either GC or GEJ adenocarcinoma had treatment interruptions due to nausea/vomiting compared with Asian patients with GC or GEJ adenocarcinoma. Comparing drug exposure by race and tumor location (Supplementary Table S6, available at https://doi.org/10.1016/j.esmoop.2025.105931), among white patients with GEJ adenocarcinoma, those on zolbetuximab had reduced duration and dose exposure to study drug (and fewer days of exposure to oxaliplatin, which is part of the chemotherapy backbone) compared with Asian patients with either GC or GEJ adenocarcinoma.

The impact on the efficacy of treatment interruption, discontinuation, and inadequate exposure in these subgroups is presented in the efficacy analyses by race and tumor type (Supplementary Table S7, available at https://doi.org/10.1016/j.esmoop.2025.105931). In white patients with GC, mPFS for zolbetuximab versus placebo was 8.5 versus 8.3 months (HR 0.77, 95% CI 0.58-1.00), and mOS was 15.9 versus 13.6 months (HR 0.82, 95% CI 0.64-1.06). In Asian patients with GC, mPFS was 10.5 versus 6.6 months (HR 0.54, 95% CI 0.43-0.70), and mOS was 18.7 versus 13.0 months (HR 0.67, 95% CI 0.54-0.84).

In white patients with GEJ adenocarcinoma, mPFS was 8.3 versus 10.4 months (HR 1.15, 95% CI 0.72-1.83), and mOS was 14.4 versus 16.0 months (HR 1.30, 95% CI 0.85-1.98). In Asian patients with GEJ adenocarcinoma, mPFS was 8.1 versus 8.0 months (HR 0.85, 95% CI 0.39-1.85), and mOS was 17.7 versus 16.5 months (HR 0.86, 95% CI 0.43-1.74).

For treatment discontinuation due to nausea/vomiting by prior gastrectomy, please see Supplementary Results and Table S8, available at https://doi.org/10.1016/j.esmoop.2025.105931.

Usage of antiemetics

Given that the incidence of nausea/vomiting was highest on the day of the first infusion and decreased with infusions 2-4, we explored the antiemetic regimens used as prophylaxis for nausea/vomiting on C1D1 and effects of the antiemetic regimens on the incidence of nausea/vomiting. Combinations of antiemetics were given prophylactically before the first infusion of zolbetuximab (Table 1). Overall, 97 patients received a three-drug prophylactic antiemetic regimen; among these, 75.3% (73/97) of patients did not experience vomiting, and 60.8% (59/97) of patients did not experience nausea on C1D1. Of 280 patients who received a two-drug prophylactic antiemetic regimen, 58.9% (165/280) did not experience vomiting and 51.4% (144/280) did not experience nausea on C1D1. See Supplementary Table S9, available at https://doi.org/10.1016/j.esmoop.2025.105931, for combinations of prophylactic antiemetics given before the first infusion for the placebo group.

Table 1.

Use of prophylactic antiemetics and occurrence of nausea or vomiting on the day of the first zolbetuximab infusion

Use of prophylactic antiemetics, n (%)	n	With nausea	Without nausea	With vomiting	Without vomiting
Any antiemetics	521	242	279	213	308
Yes	8	4	4	4	4
No
Standard antiemetic regimensa	377	174 (46.2)	203 (53.8)	139 (36.9)	238 (63.1)
Two-drug antiemetic regimens
NK-1 receptor antagonist + 5-HT3 receptor antagonist	187	94 (50.3)	93 (49.7)	70 (37.4)	117 (62.6)
5-HT3 receptor antagonist + steroids	93	42 (45.2)	51 (54.8)	45 (48.4)	48 (51.6)
Three-drug antiemetic regimens
NK-1 receptor antagonist + 5-HT3 receptor antagonist + steroids	97	38 (39.2)	59 (60.8)	24 (24.7)	73 (75.3)
Other antiemetic regimensb	144	68 (47.2)	76 (52.8)	74 (51.4)	70 (48.6)

Data represent the combined analysis of SPOTLIGHT and GLOW.

5-HT3, serotonin (5-hydroxytryptamine 3); NK-1, neurokinin-1.

^aGiven that sample sizes beyond three-drug antiemetic regimens were small, various combinations of antiemetics not commonly used may not have been captured. Antiemetic regimens were excluded if not accurately documented as prophylactic or concomitant medication.

^bIncludes patients with less common combinations of antiemetics as prophylactic or concomitant medications.

Before the first zolbetuximab infusion, 44.4% (124/279) of patients in SPOTLIGHT and 33.6% (85/253) of patients in GLOW received steroids as prophylactic medication. Among 190 patients who received steroids as part of an antiemetic combination regimen on C1D1, 57.9% (110/190) did not experience nausea and 63.7% (121/190) did not experience vomiting. The impact of steroid use on treatment efficacy at any time during the study was assessed in an exploratory ad hoc analysis. Patients who received zolbetuximab or placebo with steroids at any time during the study period had a mean duration of steroid exposure of 28.4 days (n = 198) or 27.6 days (n = 146), respectively. The exploratory ad hoc analysis showed that in patients who received steroids, mPFS was 10.5 months (95% CI 8.9-12.5 months) with zolbetuximab and 8.3 months (95% CI 8.1-9.2 months) with placebo (HR 0.66, 95% CI 0.54-0.81); mOS was 18.4 months (95% CI 16.4-19.7 months) with zolbetuximab and 13.8 months (95% CI 12.3-16.0 months) with placebo (HR 0.71, 95% CI 0.59-0.86; Figure 3). In patients who did not receive steroids, mPFS was 8.3 months (95% CI 6.4-8.7 months) with zolbetuximab and 7.1 months (95% CI 6.2-8.3 months) with placebo (HR 0.87, 95% CI 0.67-1.13); mOS was 12.0 months (95% CI 10.7-14.5 months) with zolbetuximab and 13.2 months (95% CI 11.3-14.9 months) with placebo (HR 0.95, 95% CI 0.75-1.20; Figure 3).

Figure 3.

Kaplan–Meier curves in patients who did or did not receive steroids with zolbetuximab or placebo. (A) PFS and (B) OS in patients who received steroids with zolbetuximab or placebo and (C) PFS and (D) OS in patients who did not receive steroids with zolbetuximab or placebo. Data represent the combined analysis of SPOTLIGHT and GLOW.

CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; REI, relative exposure index.

Patients receiving zolbetuximab with steroids had longer mPFS of 2.2 months and longer mOS of 6.4 months than those who did not receive steroids (mPFS: 10.5 versus 8.3 months, HR 0.65, 95% CI 0.51-0.84; mOS: 18.4 versus 12.0 months, HR 0.70, 95% CI 0.56-0.87; Figure 4). For efficacy in patients who received steroids in individual trials of SPOTLIGHT and GLOW, see Supplementary Results and Figures S5-S8, available at https://doi.org/10.1016/j.esmoop.2025.105931.

Figure 4.

Kaplan–Meier curves stratified byuse of steroids in the zolbetuximab group. (A) PFS and (B) OS in patients who received steroids with zolbetuximab and in patients who did not receive steroids with zolbetuximab. Data represent the combined analysis of SPOTLIGHT and GLOW.

CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival.

In the zolbetuximab group, 53.6% (285/532) and 28.0% (149/532) of patients received a proton pump inhibitor (PPI) and a histamine-2 receptor (H2) blocker, respectively, at any time during the study period. In the placebo group, 48.4% (255/527) and 19.5% (103/527) of patients received a PPI and a H2 blocker, respectively, at any time during the study period.

Discussion

The management of G/GEJ adenocarcinoma is increasingly guided by biomarker-selected approaches, with several therapies, including zolbetuximab, having associated biological targets.6, 7, 8^,10, 11, 12, 13, 14 Zolbetuximab plus chemotherapy is a global option for first-line treatment of patients with HER2-negative, locally advanced unresectable or metastatic G/GEJ adenocarcinoma whose tumors are CLDN18.2-positive.6, 7, 8 The focused analyses presented here underscore the importance of managing nausea and vomiting associated with zolbetuximab. We show that patients with inadequate dose exposure or early discontinuation due to nausea/vomiting generally had numerically shorter PFS and OS in the phase III SPOTLIGHT and GLOW trials.

Nausea and vomiting were the most common TEAEs in patients treated with zolbetuximab plus chemotherapy, occurring in 46.5% and 41.0%, respectively, of patients on C1D1. Considering the effect of chemotherapy-induced nausea and vomiting (CINV),¹⁵ this analysis focused on nausea or vomiting occurring on the actual day of zolbetuximab or placebo infusion to determine the time of onset and symptoms likely attributable to zolbetuximab. Given the rapid median time to initiation of nausea/vomiting after starting the infusion of zolbetuximab, it is likely that nausea/vomiting on C1D1 was attributable to zolbetuximab and should be considered acute. After C1D1, it becomes challenging to determine causality given the gastrointestinal effects (e.g. nausea and vomiting) associated with CINV. However, we demonstrated decreased incidence of nausea/vomiting events with each subsequent zolbetuximab infusion day. Of note, delayed onset of nausea/vomiting possibly related to zolbetuximab could not be evaluated in the current analysis.

During the first zolbetuximab infusion, patients who underwent infusion modifications due to AEs had a faster initial infusion than patients without infusion modifications, suggesting that a faster initial infusion may have contributed to AEs observed during the first infusion. Slowing the rate or temporarily interrupting zolbetuximab infusions may be helpful in mitigating nausea/vomiting.¹⁶ Of note, local prescribing information may suggest slower infusion rates than the infusion rates recommended in SPOTLIGHT and GLOW.¹⁷ These mitigation strategies leveraged in the real world may result in reduced occurrences of AEs during the first infusion than what was observed in clinical trials.

Nausea/vomiting leading to inadequate dose exposure and early discontinuation in some patients who received zolbetuximab may have resulted in suboptimal zolbetuximab exposure. Exploratory ad hoc analyses showed that, after censoring patients with inadequate dose exposure or early discontinuation due to these AEs, survival outcomes improved, reinforcing the clinical relevance of managing these symptoms early. Further analyses submitted to regulatory bodies showed even better OS HR (0.66) after censoring patients with inadequate dose exposure or discontinuation before 180 days due to nausea or vomiting.¹⁸ By minimizing early discontinuations, optimal treatment exposure may be reached to maximize efficacy.

Among patients receiving zolbetuximab, a greater proportion of white patients than Asian patients had nausea/vomiting leading to inadequate dose exposure or early discontinuation. In both white and Asian patients, OS was prolonged after censoring for inadequate dose exposure or early discontinuation; however, the magnitude of benefit was greater for white patients, as a higher proportion of white patients than Asian patients had nausea/vomiting leading to inadequate dose exposure and early discontinuation. These findings support that inadequate dose exposure or early discontinuation may have affected the observed PFS and OS in white patients who received zolbetuximab and may have contributed to the numerically impaired PFS and OS per the observed HR for white versus Asian patients who received zolbetuximab in SPOTLIGHT and GLOW.

The observed difference in efficacy between subgroups (i.e. efficacy appearing less favorable in the white and GEJ subgroup compared with the overall population) can likely be explained by lower therapy exposure caused by treatment interruption and early discontinuation due to nausea/vomiting. This phenomenon was evident in the higher rate of treatment interruption and discontinuation due to nausea/vomiting in white patients with GC compared with Asian patients with GC and the higher rate of treatment interruption due to nausea/vomiting in white patients with GEJ adenocarcinoma compared with Asian patients with GEJ adenocarcinoma. This subsequently lower study therapy exposure in white patients compared with Asian patients has important clinical relevance.

Zolbetuximab plus chemotherapy is considered highly emetogenic.^19,20 International guidelines recommend a minimum of a three-drug antiemetic prophylaxis (e.g. olanzapine, NK-1 receptor antagonist, 5-HT3 receptor antagonist, dexamethasone) for such regimens.19, 20, 21 Steroids are standard antiemetics and are not contraindicated per zolbetuximab prescribing information.¹⁷ A preclinical study utilizing ferret models demonstrated that the administration of steroids can mitigate gastric mucosal injury associated with zolbetuximab-induced emesis.²² The use of steroids was not recommended during SPOTLIGHT and GLOW (particularly during the first zolbetuximab infusion) out of concerns of diminishing the antibody-dependent cellular cytotoxicity of zolbetuximab; however, they were not prohibited. Patients who received steroids demonstrated a maintenance of efficacy with zolbetuximab treatment versus those who did not, with numerical improvements in mPFS and mOS. Patients also had lower rates of nausea and vomiting when steroids were used as part of a two- or three-drug antiemetic regimen. Patients who received zolbetuximab with steroids at any time during the study had numerically longer PFS and OS than those who received zolbetuximab without steroids.

This analysis showed that when patients on zolbetuximab received a two-drug antiemetic regimen, 51.4% and 58.9% of patients did not experience nausea or vomiting, respectively. When patients received a three-drug antiemetic regimen, a meaningful increase was observed in the proportion of patients who did not experience nausea (60.8%) or vomiting (75.3%). When patients received steroids as part of a two- or three-drug antiemetic regimen, 57.9% of patients did not experience nausea and 63.7% did not experience vomiting. These data suggest that the use of multi-class antiemetic prophylaxis (three or more classes that include the use of steroids) may be helpful in mitigating nausea/vomiting, therefore leading to maximizing potential efficacy by maintaining treatment adherence and exposure.

A recent Delphi panel developed consensus-based guidance to help prevent and manage nausea/vomiting associated with zolbetuximab.¹⁶ The Delphi panel endorsed the prophylactic use of any regimens for high emetic risk as recommended by the National Comprehensive Cancer Network® (NCCN®), such as NK-1 receptor antagonist + 5-HT3 receptor antagonist + dexamethasone + olanzapine; NK-1 receptor antagonist + 5-HT3 receptor antagonist + dexamethasone; or 5-HT3 receptor antagonist + dexamethasone + olanzapine.^16,19 The Delphi panel also suggested, depending on patient symptoms of nausea/vomiting, to consider modifying the zolbetuximab infusion rate, interrupting zolbetuximab infusions temporarily for 30-60 min, concomitantly administering antiemetic medications not used for prophylaxis, and/or providing intravenous hydration.¹⁶ For maximal mucosal protection, providers may consider using antiulcer medications (e.g. PPI or H2 blocker) a few days to 1 week before zolbetuximab treatment.¹⁶ Education of patients, caregivers, and clinic staff was also emphasized to prepare for and manage these AEs effectively.¹⁶ Real-world experience at a single institution in Japan showed that 61.9% (13/21) of patients had nausea at the first zolbetuximab infusion. However, with proper management such as infusion interruptions and rescue antiemetics, only 9.5% (2/21) of patients had vomiting, and all patients successfully completed the first zolbetuximab infusion.²³

This analysis has limitations, and the findings should be interpreted with caution. Further real-world studies are essential to support and validate these findings as new evidence becomes available. Median time to the first occurrence of nausea/vomiting was only evaluated in a subset of patients who experienced nausea/vomiting, due to limited time stamp data availability with nausea/vomiting AE reporting. The small sample size in some subgroups—particularly the GEJ subgroup—limits interpretation. Analyses such as the use of prophylactic PPIs or H2 blockers and their impact on nausea/vomiting could not be fully assessed due to incomplete data. Additionally, the analysis focused on the most common antiemetic combinations, potentially excluding less common premedication combinations. Given that some datasets required manual enumeration, it is possible that errors in counting may have occurred.

Despite these limitations, this work highlights the need to proactively manage zolbetuximab-associated nausea/vomiting. These AEs not only affect patient and health care professional experience but can lead to inadequate dose exposure or early discontinuation that may compromise efficacy.

The experience with zolbetuximab also underscores a broader consideration in oncology: the clinical consequences of early toxicities are often underexplored in pivotal trials. A deep understanding of the impact of AEs on outcomes is important for optimizing patient care. Emerging agents like zolbetuximab and the anti-fibroblast growth factor receptor 2 antibody bemarituzumab have distinct AE profiles,²⁴ and familiarity with toxicity management is key to striking a balance between efficacy and toxicity in patients receiving these treatments. Understanding and addressing toxicities early, especially during the critical first cycles, may support adherence and optimize clinical benefits.

Conclusions

Nausea and vomiting are the most common AEs associated with zolbetuximab. Their early identification and optimal management through antiemetic prophylaxis, infusion modifications, and patient education are not only essential to improve patient comfort but also to maintain treatment exposure and maximize clinical benefits. These strategies should be adopted into routine clinical practice and evaluated in real-world studies. As supportive care practices evolve, real-world outcomes with zolbetuximab may further improve beyond those observed in the pivotal trials.