Abstract
Background
Human epidermal growth factor receptor 2 (HER2) and claudin 18.2 (CLDN18.2) are key therapeutic targets in metastatic gastric and gastroesophageal junction cancer (mGC/GEJC). Trastuzumab deruxtecan (T-DXd) is standard care for previously treated HER2-positive mGC/GEJC, but the prognostic impact of CLDN18.2 remains unclear.
Patients and methods
We retrospectively reviewed patients with HER2-positive mGC/GEJC treated with T-DXd at National Cancer Center Hospital East until February 2025. T-DXd outcomes were compared between CLDN18.2-positive (≥2+ in ≥75% of tumor cells) and -negative patients in two cohorts: those with HER2 positivity before first-line therapy (overall cohort) and those maintaining HER2 positivity immediately before T-DXd (remaining HER2 cohort). CLDN18.2 expression was evaluated at any time before T-DXd administration.
Results
Eighty-seven patients were included in the overall cohort and 30 in the remaining HER2 cohort. In the overall cohort, progression-free survival (PFS) was shorter in CLDN18.2-positive patients [3.9 versus 5.3 months; hazard ratio (HR) 1.87, 95% confidence interval (CI) 1.04-3.34, P = 0.036], with an adjusted HR of 1.82 (95% CI 1.02-3.28, P = 0.047). Overall survival (OS) showed a shorter trend (8.6 versus 11.2 months, HR 1.36, 95% CI 0.76-2.45, P = 0.30). In the remaining HER2 cohort, CLDN18.2-positive patients had shorter PFS (3.2 versus 8.0 months, HR 3.40, 95% CI 1.38-8.40, P = 0.008) and OS (7.1 versus 12.9 months, HR 2.44, 95% CI 1.04-5.74, P = 0.041).
Conclusions
CLDN18.2 positivity may attenuate the efficacy of T-DXd in HER2-positive mGC/GEJC, supporting the rationale for dual blockade of CLDN18.2 and HER2.
Key words: gastric cancer, trastuzumab deruxtecan, claudin 18.2, HER2
Highlights
-
•
CLDN18.2 positivity was observed in 18% of HER2-positive metastatic gastric/GEJ cancer patients treated with T-DXd.
-
•
CLDN18.2-positive tumors had shorter PFS and OS versus CLDN18.2-negative tumors, especially in patients with maintained HER2.
-
•
Lower cut-off (40%. 25%) partially reduced the prognostic gap, suggesting high CLDN18.2 attenuates the efficacy of T-DXd.
-
•
These findings support the rationale for dual HER2-CLDN18.2 targeting in double-positive gastric/GEJ cancer patients.
Introduction
Human epidermal growth factor receptor 2 (HER2) is overexpressed or amplified in ∼15%-20% of gastric and gastroesophageal junction cancer (GC/GEJC).1, 2, 3 HER2 positivity, defined as 3+ or 2+/in situ hybridization (ISH)-positive, has been independently established as a therapeutic category. Trastuzumab combined with chemotherapy with or without pembrolizumab has been the standard first-line therapy for HER2-positive metastatic GC/GEJC (mGC/GEJC).1,4 Trastuzumab deruxtecan (T-DXd) has shown survival benefit in previously treated HER2-positive mGC/GEJC.5, 6, 7
Tight junction molecule claudin 18.2 (CLDN18.2) has emerged as a novel therapeutic target, with high expression maintained in ∼30%-40% of mGC/GEJC cases.8 Randomized phase III trials, SPOTLIGHT and GLOW demonstrated the survival benefit of adding zolbetuximab, first-in-human monoclonal antibody against CLDN18.2, for treatment-naive patients with HER2-negative and CLDN18.2-positive mGC/GEJC.9,10 Although patients with HER2-positive mGC/GEJC were excluded from these pivotal trials, cases with dual positivity were occasionally identified.
According to the previous studies, CLDN18.2 positivity does not affect the prognosis of patients with mGC/GEJC who received cytotoxic chemotherapy with or without immunotherapy.8,11,12 The impact on treatment outcomes after HER2-directed therapy remains unknown.
In this study, we retrospectively compared the clinicopathological characteristics and treatment outcomes between CLDN18.2-positive and -negative patients among those with HER2-positive mGC/GEJC. The primary objective was to assess the impact of CLDN18.2 status on the efficacy of T-DXd, with first-line trastuzumab-based therapy analyzed as a reference. These findings provide insights into potential combination strategies targeting both HER2 and CLDN18.2 in double-positive cases.
Patients and methods
Study design and patient selection
This was a single-institution, retrospective cohort study conducted at the National Cancer Center Hospital East, Kashiwa, Japan. We reviewed the medical records of consecutive patients with HER2-positive mGC/GEJC who received T-DXd until February 2025. Inclusion criteria were as follows: (i) histologically or cytologically confirmed adenocarcinoma, (ii) unresectable, metastatic advanced or recurrent disease, (iii) HER2 status of immunohistochemistry (IHC) 3+ or IHC 2+ with ISH-positive, (iv) previous treatment with a trastuzumab-based therapy before T-DXd administration, (v) availability of CLDN18.2 expression status assessed by IHC before T-DXd initiation, and (vi) provided written informed consent to the treatment and prospective biomarker study (approval ID: 2017-120, approval date: 26 September 2017). Patients were excluded if the T-DXd dose was <5.4 mg/kg. T-DXd was administered on the same schedule as in the pivotal clinical trials.5, 6, 7
Cohort definition
For this analysis, patients were divided into two cohorts based on the timing of HER2 status assessment. The overall cohort included patients with HER2-positive mGC/GEJC identified before initiation of first-line therapy, whereas the remaining HER2 cohort comprised patients with maintained HER2 positivity confirmed immediately before T-DXd administration. In both cohorts, CLDN18.2 status assessed before initiation of trastuzumab-based therapy was preferentially used; if CLDN18.2 status was unavailable at the initiation of first-line therapy, samples obtained during or after administration of trastuzumab-based therapy were used. All CLDN18.2 assessments were carried out before T-DXd initiation. Furthermore, the impact of CLDN18.2 status on clinical outcomes of first-line trastuzumab-based therapy in enrolled patients was also assessed.
Biomarker assessment
HER2 status was locally assessed by two board-certified pathologists using a monoclonal anti-HER2 antibody [PATHWAY HER2 (4B5), Ventana, Tucson, AZ] and fluorescence in situ hybridization with the PathVysion HER-2 Probe Kit (Abbott Laboratories, Abbott Park, IL). CLDN18.2 expression in primary or metastatic lesions was assessed by IHC using a CLDN18 antibody (clone 43-14A, Roche Ventana). Staining intensity was graded as weak (1+), moderate (2+), or strong (3+), and the proportion of positive tumor cells was recorded. CLDN18.2 positivity was defined as membranous staining of ≥2+ intensity in ≥75% of tumor cells, in accordance with the criteria employed in clinical trials of zolbetuximab.9,10 Additional cut-off analyses were carried out using lower thresholds of 40% and 25% of the tumor area. All patients provided written informed consent for biomarker analysis (UMIN000019129).
Statistical analysis
Progression-free survival (PFS) was defined as the time from the initiation of T-DXd treatment to radiographic or clinical disease progression or death from any cause, whichever occurred first. Overall survival (OS) was defined as the time from the initiation of T-DXd treatment to death from any cause. Patients without an event were censored at the date of last follow-up.
Objective response rate (ORR) and disease control rate (DCR) were assessed according to the RECIST version 1.1.13 Survival curves were estimated using the Kaplan–Meier method and compared with the log-rank test. ORR and DCR with 95% confidence intervals (CIs) were calculated. Fisher’s exact test was used for categorical variables, and the Mann–Whitney U test was applied for continuous variables. Cox proportional hazards model was applied to evaluate prognostic factors. In addition to univariate analysis, multivariate Cox regression was carried out to adjust HRs by prognostic factors identified in our previous study during T-DXd treatment.14 Specifically, the presence of a primary tumor and peritoneal metastasis were included for PFS, whereas disease status (metastatic versus recurrent), the presence of a primary tumor, and peritoneal metastasis were included for OS.
A two-sided P value < 0.05 was considered statistically significant. All statistical analyses were carried out using R software (R Foundation for Statistical Computing, Vienna, Austria) with RStudio as the integrated development environment.
Ethics
All procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and its later versions. Written informed consent to be included in the study or the equivalent was obtained from all patients. This study was approved by the ethics committee of our institution, National Cancer Center Hospital East Certified Review Board (Approval ID: 2017-120, Approval date: 26 September 2017).
Results
Patient flow
A total of 124 consecutive patients with mGC/GEJC received T-DXd at our institution until February 2025. A total of 37 patients were excluded: unavailable CLDN18.2 status (n = 35), starting dose <5.4 mg/kg (n = 1), and no previous use of trastuzumab before T-DXd (n = 1). Ultimately, 87 consecutive patients were eligible for this study and grouped into the overall cohort. Among 87 patients, HER2 status immediately before T-DXd initiation was available for 45 patients, with 30 remaining positive and 15 losing positivity; the former comprised the remaining HER2 cohort. In the overall cohort, HER2 expression was assessed in all 87 patients using tumor samples obtained before trastuzumab initiation. In contrast, CLDN18.2 expression was assessed using tumor samples obtained before trastuzumab in 56 patients and during or after trastuzumab treatment in 31 patients. In the remaining HER2 cohort, HER2 expression was assessed in all 30 patients using tumor samples obtained immediately before T-DXd treatment. CLDN18.2 expression was assessed using tumor samples obtained before trastuzumab in 15 patients and during or after trastuzumab treatment in the remaining 15 patients (Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2025.100300).
Patient characteristics
Baseline characteristics of the overall cohort (n = 87) are summarized in Table 1. The median age was 65 years (range, 26-86 years), and 66 (75%) patients were male. Eastern Cooperative Oncology Group (ECOG) performance status (PS) was 0 in 49 (56%) patients. HER2 status was IHC 3+ in 77 (88%) patients and IHC 2+/ISH-positive in 10 (12%) patients. CLDN18.2 positivity was observed in 16 (18%) patients. Although there were no statistically significant differences in baseline characteristics between CLDN18.2-positive and -negative patients, more patients with PS ≥1 (62% versus 40%), diffuse-type histology (37% versus 21%), and multiple metastatic sites (88% versus 63%) were included in the CLDN18.2-positive group.
Table1.
Baseline patient characteristics of overall cohort
| Characteristics | All (n = 87) | CLDN18.2 ≥75% (n = 16) | CLDN18.2 <75% (n = 71) | P value |
|---|---|---|---|---|
| Age, years, n (%) | ||||
| Median (range) | 65 (26-86) | 67 (26-86) | 65 (30-85) | 0.40 |
| ≥65 | 55 (63) | 12 (75) | 43 (60) | 0.39 |
| ≥75 | 24 (27) | 6 (38) | 18 (25) | 0.37 |
| Sex, n (%) | ||||
| Male | 66 (75) | 12 (75) | 54 (76) | 1.00 |
| Female | 21 (25) | 4 (25) | 17 (24) | — |
| ECPG PS, n (%) | ||||
| 1-2 | 38 (44) | 10 (62) | 28 (40) | 0.15 |
| 0 | 49 (56) | 6 (38) | 43 (60) | — |
| HER2 IHC, n (%) | ||||
| 3+ | 77 (88) | 15 (94) | 62 (87) | 0.68 |
| 2+ | 10 (12) | 1 (6) | 9 (13) | — |
| Lauren classification, n (%) | ||||
| Diffuse | 21 (24) | 6 (37) | 15 (21) | 0.29 |
| Intestinal | 66 (76) | 10 (63) | 56 (79) | — |
| Histology, n (%) | ||||
| Sig | 4 (5) | 2 (12) | 2 (3) | 0.16 |
| Non-sig | 83 (95) | 14 (88) | 69 (97) | — |
| Location, n (%) | ||||
| GEJ | 21 (24) | 2 (12) | 19 (27) | 0.34 |
| Gastric | 66 (76) | 14 (88) | 52 (73) | — |
| Borrmann type 4, n (%) | ||||
| Yes | 7 (8) | 1 (6) | 6 (8) | 1.00 |
| No | 80 (92) | 15 (94) | 65 (92) | — |
| Disease status, n (%) | ||||
| Metastatic | 78 (90) | 16 (100) | 62 (88) | 0.20 |
| Recurrent | 9 (10) | 0 (0) | 9 (12) | — |
| Primary tumor, n (%) | ||||
| Present | 72 (83) | 15 (94) | 57 (80) | 0.29 |
| Absent | 15 (17) | 1 (6) | 14 (20) | — |
| Metastatic site, n (%) | ||||
| Lymph node | 68 (78) | 15 (94) | 53 (75) | 0.18 |
| Liver | 50 (56) | 10 (63) | 40 (55) | 0.78 |
| Peritoneal | 34 (39) | 8 (50) | 26 (36) | 0.40 |
| Lung | 16 (18) | 4 (25) | 12 (17) | 0.48 |
| No of metastatic site, n (%) | ||||
| ≥2 | 59 (68) | 14 (88) | 45 (63) | 0.08 |
CLDN18.2, claudin 18.2; ECOG PS, Eastern Cooperative Oncology Group performance status; GEJ, gastroesophageal junction; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; NA, not available.
Baseline characteristics of the remaining HER2 cohort (n = 30) are shown in Table 2. The median age was 65 years (range, 26-86 years), and 20 (66%) patients were male. ECOG PS of 0 was observed in 15 (50%) patients. All patients were classified as HER2 IHC 3+. CLDN18.2 positivity was observed in 8 (27%) patients. Similar to the overall cohort, PS ≥1 (75% versus 41%), diffuse-type histology (50% versus 18%), and multiple metastatic sites (88% versus 73%) were more frequent in the CLDN18.2-positive group, although the differences did not reach statistical significance.
Table 2.
Baseline patient characteristics of remaining HER2 cohort
| Characteristics | All (n = 30) | CLDN18.2 ≥75% (n = 8) | CLDN18.2 <75% (n = 22) | P value |
|---|---|---|---|---|
| Age, years, n (%) | ||||
| Median (range) | 65 (26-86) | 63 (26-85) | 66 (37-83) | 0.91 |
| ≥65 | 19 (63) | 5 (63) | 14 (64) | 1.00 |
| ≥75 | 10 (33) | 3 (38) | 7 (32) | 1.00 |
| Sex, n (%) | ||||
| Male | 20 (66) | 5 (63) | 15 (68) | 1.00 |
| Female | 10 (34) | 3 (75) | 7 (34) | — |
| ECPG PS, n (%) | ||||
| 1-2 | 15 (50) | 6 (75) | 9 (41) | 0.21 |
| 0 | 15 (50) | 2 (25) | 13 (59) | — |
| HER2 IHC, n (%) | ||||
| 3+ | 30 (100) | 8 (100) | 22 (100) | 1.00 |
| 2+ | 0 (0) | 0 (0) | 0 (0) | — |
| Lauren classification, n (%) | ||||
| Diffuse | 8 (27) | 4 (50) | 4 (18) | 0.16 |
| Intestinal | 22 (73) | 4 (50) | 18 (82) | — |
| Histology, n (%) | ||||
| Sig | 1 (3) | 1 (13) | 0 (0) | 0.29 |
| Non-sig | 26 (87) | 6 (74) | 20 (91) | — |
| Location, n (%) | ||||
| GEJ | 4 (13) | 0 (0) | 4 (18) | 0.55 |
| Gastric | 26 (87) | 8 (100) | 18 (82) | — |
| Borrmann type 4, n (%) | ||||
| Yes | 5 (17) | 0 (0) | 5 (23) | 0.29 |
| No | 25 (83) | 8 (100) | 17 (77) | — |
| Disease status, n (%) | ||||
| Metastatic | 27 (90) | 8 (100) | 19 (86) | 0.54 |
| Recurrent | 3 (10) | 0 (0) | 3 (14) | — |
| Primary tumor, n (%) | ||||
| Present | 26 (87) | 8 (100) | 18 (82) | 0.55 |
| Absent | 4 (13) | 0 (0) | 4 (18) | — |
| Metastatic site, n (%) | ||||
| Lymph node | 28 (93) | 8 (100) | 20 (91) | 1.00 |
| Liver | 14 (47) | 3 (38) | 11 (50) | 0.69 |
| Peritoneal | 13 (43) | 5 (63) | 8 (36) | 0.24 |
| Lung | 5 (17) | 2 (25) | 3 (14) | 0.59 |
| No of metastatic site, n (%) | ||||
| ≥2 | 23 (77) | 7 (88) | 16 (73) | 0.64 |
CLDN18.2, claudin 18.2; ECOG PS, Eastern Cooperative Oncology Group performance status; GEJ, gastroesophageal junction; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; NA, not available.
Treatment outcomes in overall cohort
For the 79 patients with measurable lesions, the ORR and DCR were 53% (95% CI 42.2%-63.8%) and 85% (95% CI 76.8%-92%), respectively. There was no significant difference in ORR (40% versus 57%, P = 0.26) and DCR (80% versus 87%, P = 0.43) between CLDN18.2-positive (n = 15) and CLDN18.2-negative (n = 64).
During a minimum follow-up of 2.7 months, 83 (95%) patients experienced disease progression, and 73 (84%) had died. The median PFS was 5.0 months (95% CI 4.2-7.3 months), and the median OS was 10.9 months (95% CI 8.1-13.7 months). There was a significant difference in PFS between the CLDN18.2-positive and -negative groups (median PFS 3.9 versus 5.3 months, HR 1.87, 95% CI 1.04-3.34, P = 0.036) (Figure 1A). After adjustment, CLDN18.2 positivity remained significantly associated with shorter PFS (adjusted HR 1.82, 95% CI 1.02-3.28, P = 0.047). In contrast, there was no significant difference in OS between the CLDN18.2-positive and -negative groups (median OS 8.6 versus 11.2 months, HR 1.36, 95% CI 0.76-2.45 months, P = 0.30) (Figure 1B), and CLDN18.2 positivity was not an independent prognostic factor for OS after adjustment (adjusted HR 1.43, 95% CI 0.78-2.60, P = 0.25). We evaluated ORR, PFS, and OS according to different CLDN18.2 cut-off levels (75%, 40%, and 25%) (Figure 3A).
Figure 1.
(A) Progression-free survival (PFS) according to CLDN18.2 expression (75%cut-off)in the overall cohort. Kaplan–Meier curves for PFS in the overall cohort stratified by CLDN18.2 expression. Among 87 patients, 16 were CLDN18.2-positive (blue line) and 71 were CLDN18.2-negative (red line). Median PFS was 3.9 months (95% CI 2.5-9.2 months) in CLDN18.2-positive patients and 5.3 months (95% CI 4.3-8.3 months) in CLDN18.2-negative patients. The hazard ratio was 1.87 (95% CI 1.04-3.34, P = 0.036). The number of patients at risk over time is shown below the x-axis. (B) Overall survival (OS) according to CLDN18.2 expression (75%cut-off)in the overall cohort. Kaplan–Meier curves for OS in the overall cohort stratified by CLDN18.2 expression. Among 87 patients, 16 were CLDN18.2-positive (blue line) and 71 were CLDN18.2-negative (red line). Median OS was 8.6 months (95% CI 6.3-23.8 months) in CLDN18.2-positive patients and 11.2 months (95% CI 8.1-17.7 months) in CLDN18.2-negative patients. The hazard ratio was 1.36 (95% CI 0.76-2.45, P = 0.30 months). The number of patients at risk over time is shown below the x-axis.
Figure 3.
(A) Objective response rate (ORR),progression-freesurvival (PFS), and overall survival (OS) according to CLDN18.2 expression (75%, 40%, and 25%cut-off)in the overall cohort. Forest plots of odds ratios (ORs) for ORR and hazard ratios (HRs) for PFS and OS in the overall cohort stratified by CLDN18.2 expression at cut-off levels of ≥75%, ≥40%, and ≥25%. For ORR, the response rates were 40% versus 57% (OR 0.50, 95% CI 0.16-1.57, P = 0.262) at a cut-off of ≥75%, 39% versus 59% (OR 0.40, 95% CI 0.15-1.02, P = 0.062) at ≥40%, and 44% versus 58% (OR 0.50, 95% CI 0.20-1.25, P = 0.169) at ≥25%. For PFS, the median PFS was 3.9 versus 5.3 months (HR 1.87, 95% CI 1.04-3.34, P = 0.036) at ≥75%, 4.3 versus 5.5 months (HR 1.73, 95% CI 1.08-2.78, P = 0.022) at ≥40%, and 4.7 versus 5.0 months (HR 1.43, 95% CI 0.91-2.25, P = 0.123) at ≥25%. For OS, the median OS was 8.6 versus 11.2 months (HR 1.36, 95% CI 0.76-2.45, P = 0.301) at ≥75%, 8.0 versus 12.0 months (HR 1.45, 95% CI 0.89-2.36, P = 0.133) at ≥40%, and 8.6 versus 11.3 months (HR 1.30, 95% CI 0.81-2.09, P = 0.276) at ≥25%. The vertical line indicates an OR or HR of 1.0. (B)Objective response rate (ORR),progression-freesurvival (PFS), and overall survival (OS) according to CLDN18.2 expression (75%, 40%, and 25%cut-off)in the remaining HER2 cohort. Forest plots of odds ratios (ORs) for ORR and hazard ratios (HRs) for PFS and OS in the remaining HER2 cohort stratified by CLDN18.2 expression at cut-off levels of ≥75%, ≥40%, and ≥25%. For ORR, the response rates were 43% versus 72% (OR 0.15, 95% CI 0.02-1.07, P = 0.075) at a cut-off of ≥75%, 50% versus 67% (OR 0.50, 95% CI 0.10-2.58, P = 0.442) at ≥40%, and 54% versus 66% (OR 0.58, 95% CI 0.12-2.95, P = 0.688) at ≥25%. For PFS, the median PFS was 3.2 versus 8.0 months (HR 3.40, 95% CI 1.38-8.40, P = 0.008) at ≥75%, 3.9 versus 7.5 months (HR 2.25, 95% CI 1.00-5.07, P = 0.051) at ≥40%, and 5.3 versus 7.3 months (HR 1.74, 95% CI 0.79-3.85, P = 0.170) at ≥25%. For OS, the median OS was 7.1 versus 12.9 months (HR 2.44, 95% CI 1.04-5.74, P = 0.041) at ≥75%, 9.4 versus 14.6 months (HR 2.52, 95% CI 1.12-5.63, P = 0.020) at ≥40%, and 10.3 versus 12.1 months (HR 1.94, 95% CI 0.89-4.25,P = 0.091) at ≥25%. The vertical line indicates an OR or HR of 1.0.
Treatment outcomes in remaining HER2 cohort
Among 25 patients with measurable lesions, the ORR was 60% (95% CI 40.7%-76.6%) and the DCR was 84% (95% CI 65.3%-93.6%), respectively. The ORR was numerically lower in CLDN18.2-positive group (n = 7) compared with that of CLDN18.2-negative (n = 18), without statistical significance (43% versus 72%, P = 0.075). The DCR was similar in both groups (86% versus 94%, P = 0.49).
The median PFS was 5.5 months (95% CI 3.5-9.8 months), and the median OS was 11.1 months (95% CI 8.6-18.0 months). Patients with CLDN18.2-positive mGC/GEJC (n = 8) had significantly shorter PFS compared with those with CLDN18.2-negative disease (median PFS 3.2 versus 8.0 months, HR 3.40, 95% CI 1.38-8.40, P = 0.008) (Figure 2A). OS was significantly shorter in patients with CLDN18.2-positive disease than in those with CLDN18.2-negative disease (median OS 7.1 versus 12.9 months, HR 2.44, 95% CI 1.04-5.74 months, P = 0.041) (Figure 2B). Similar to the overall cohort, we evaluated ORR, PFS, and OS according to different CLDN18.2 cut-off levels (75%, 40%, and 25%) (Figure 3B).
Figure 2.
(A)Progression-freesurvival (PFS) according to CLDN18.2 expression (75%cut-off)in the remaining HER2 cohort. Kaplan–Meier curves for PFS in the remaining HER2 cohort stratified by CLDN18.2 expression. Among 30 patients, 8 were CLDN18.2-positive (blue line) and 22 were CLDN18.2-negative (red line). Median PFS was 3.2 months (95% CI 2.5 months-not estimable) in CLDN18.2-positive patients and 8.0 months (95% CI 4.4-11.1 months) in CLDN18.2-negative patients. The hazard ratio was 3.40 (95% CI 1.38-8.40, P = 0.008). The number of patients at risk over time is shown below the x-axis. (B) Overall survival (OS) according to CLDN18.2 expression (75%cut-off)in the remaining HER2 cohort. Kaplan–Meier curves for OS in the remaining HER2 cohort stratified by CLDN18.2 expression. Among 30 patients, 8 were CLDN18.2-positive (blue line) and 22 were CLDN18.2-negative (red line). Median OS was 7.1 months (95% CI 3.6 months-not estimable) in CLDN18.2-positive patients and 12.9 months (95% CI 10.3-20.5 months) in CLDN18.2-negative patients. The hazard ratio was 2.44 (95% CI 1.04-5.74, P = 0.041). The number of patients at risk over time is shown below the x-axis.
Treatment outcomes of first-line anti-HER2 therapy
We evaluated the impact of CLDN18.2 status on clinical outcomes of first-line trastuzumab-based therapy. Among 87 patients, a total of 5 patients were excluded: 3 who commenced trastuzumab-based therapy in the second-line or later, and 2 with an unknown duration of first-line therapy. Baseline characteristics were generally comparable between CLDN18.2-positive and -negative patients (Supplementary Table S2, available at https://doi.org/10.1016/j.esmogo.2025.100300). CLDN18.2 positivity was observed in 15 (18%) patients. CLDN18.2 status was assessed using samples obtained before initiation of trastuzumab-based therapy in 53 patients and during or after trastuzumab-based therapy in 29 patients.
Among 79 patients with measurable lesions, the ORR was 71% (95% CI 60.0%-80.0%) and the DCR was 91% (95% CI 82.6%-96.3%). The ORR was numerically higher in CLDN18.2-positive group (n = 15) compared with that of CLDN18.2-negative group (n = 64), without statistical significance (93% versus 67%, P = 0.098). The DCR was similar in both groups (93% versus 92%, P = 1.00).
The median PFS was 6.2 months (95% CI 4.9-7.1 months), and the median OS was 22.5 months (95% CI 20.3-29.9 months). There was no significant difference in PFS between CLDN18.2-positive and -negative groups (median PFS 5.5 versus 6.2 months, HR 1.45, 95% CI 0.82-2.56, P = 0.20) (Figure 4A). Similarly, OS did not differ significantly between the two groups (median OS 19.5 versus 27.1 months, HR 1.56, 95% CI 0.85-2.87 months, P = 0.16) (Figure 4B). Among the 53 patients whose CLDN18.2 status was assessed before first-line trastuzumab-based therapy, no significant differences were observed in ORR, PFS, or OS between the CLDN18.2-positive and -negative groups.
Figure 4.
(A)Progression-freesurvival (PFS) according to CLDN18.2 expression (75%cut-off)in thefirst-linetrastuzumab-based therapy setting. Kaplan–Meier curves for PFS in the first-line trastuzumab-based therapy setting stratified by CLDN18.2 expression. Among 82 patients, 15 were CLDN18.2-positive (blue line) and 67 were CLDN18.2-negative (red line). Median PFS was 5.5 months (95% CI 4.7-7.1 months) in CLDN18.2-positive patients and 6.2 months (95% CI 5.3-8.5 months) in CLDN18.2-negative patients. The hazard ratio was 1.45 (95% CI 0.82-2.56, P = 0.20). The number of patients at risk over time is shown below the x-axis. (B) Overall survival (OS) according to CLDN18.2 expression (75%cut-off)in thefirst-linetrastuzumab-based therapy setting. Kaplan–Meier curves for OS in the first-line trastuzumab-based therapy setting stratified by CLDN18.2 expression. Among 82 patients, 15 were CLDN18.2-positive (blue line) and 67 were CLDN18.2-negative (red line). Median OS was 19.5 months (95% CI 16.2 months-not estimable) in CLDN18.2-positive patients and 27.1 months (95% CI 21.2-34.5 months) in CLDN18.2-negative patients. The hazard ratio was 1.56 (95% CI 0.85-2.87, P = 0.16). The number of patients at risk over time is shown below the x-axis.
Discussion
To the best of our knowledge, this is the first report to evaluate the impact of CLDN18.2 status on the clinical outcomes of T-DXd in patients with HER2-positive mGC/GEJC. In this retrospective analysis, 18% of patients with HER2-positive mGC/GEJC exhibited CLDN18.2 positivity, which is consistent with previous reports.11,15 CLDN18.2 positivity was associated with reduced efficacy of T-DXd, showing a significant difference in PFS among patients with HER2-positive mGC/GEJC in the overall cohort, which remained statistically significant after adjustment by other factors. Moreover, after enriching patients with confirmed HER2-positive status immediately before T-DXd treatment, significant differences in both PFS and OS were observed between CLDN18.2-positive and -negative groups.
GC exhibits intertumoral and intratumoral heterogeneity, which may partly account for the limited success of many targeted therapy trials.16 HER2-positive GC exhibits a broader spectrum of HER2 expressions, with heterogeneity that exceeds that observed in HER2-positive breast cancer.2 Heterogeneous HER2 expressions have been shown to attenuate the efficacy of first-line trastuzumab-containing therapy.17 Clinically, HER2-positive GC is typically associated with a higher incidence among males, a GEJ origin, and intestinal-type histology, resulting in a lower incidence of peritoneal dissemination.18,19 Previous studies have demonstrated a gradual decrease in intestinal-type histology and a concomitant increase in peritoneal metastasis with decreasing levels of HER2 expression.20 Conversely, CLDN18.2-positive GC tends to exhibit a higher proportion of diffuse-type histology and type 4 tumors, which are closely associated with peritoneal dissemination. These findings support the lower rate of CLDN18.2 positivity in patients with HER2-positive GC compared with those with HER2-negative.21 In the present study, HER2-positive/CLDN18.2-positive GC tended to exhibit a phenotype maintaining original CLDN18.2-positive GC, despite HER2 positivity. This observation suggests that coexpression of CLDN18.2 may reflect higher HER2 heterogeneity, which in turn could reduce tumor dependency on the HER2 signaling pathway, resulting in attenuated efficacy of T-DXd monotherapy. In addition, recent spatial transcriptomic analyses have revealed increased CLDN18.2 expression during trastuzumab therapy, suggesting that CLDN18.2 expression may serve as a marker of a tumor phenotype with reduced HER2 dependency.22 This hypothesis is, however, based on preliminary findings with a limited sample size, and a causal relationship between resistance to anti-HER2 therapy and CLDN18.2 up-regulation has yet to be elucidated.
Originally, CLDN18.2 was considered a lineage-specific target that was not directly involved in signal transduction, unlike HER2 or other receptor tyrosine kinases. Several studies have demonstrated that CLDN18.2 status itself does not influence the prognosis of patients treated with systemic chemotherapy with or without immune-checkpoint inhibitors.11,12,23 Our findings among patients who received first-line trastuzumab plus chemotherapy are consistent with these reports. The combination with cytotoxic agents may have diminished the prognostic impact of CLDN18.2 in this setting. The negative prognostic impact of CLDN18.2 is suggested to be apparent in the context of HER2-directed therapy among patients with HER2-positive GC.
Adding zolbetuximab to chemotherapy demonstrated superior efficacies in patients with HER2-negative and CLDN18.2-positive mGC/GEJC.9,10 Recent late-line trials of CLDN18.2-directed ADCs did not exclude patients with HER2-positive mGC/GEJC. A phase I study of CMG0901 (AZD0901) included ∼10% of HER2-positive cases, suggesting the potential antitumor activity of CMG0901 irrespective of HER2 status.24 Provided that no additive toxicity is observed, a combination strategy using ADCs against each target could represent a promising therapeutic approach. Investigation for HER2/CLDN18.2 double-positive tumor is under way by using bispecific antibodies. A novel bispecific antibody targeting both HER2 and CLDN18.2 (HC-2G4S) exerted potent therapeutic efficacy, demonstrating superiority over the combination of individual targeted agents in preclinical studies.25 A prospective evaluation of dual HER2- and CLDN18.2-targeted strategies may further refine the therapeutic landscape for patients with mGC/GEJC. This study provided the rationale for dual targeted therapy against HER2- and CLDN18.2.
This study has several limitations. Firstly, it was a retrospective, single-institution analysis with a relatively small sample size, which may limit the generalizability of the findings. Secondly, the statistical power was limited, which may have reduced the ability to detect modest differences, particularly for OS. Thirdly, the timing of CLDN18.2 assessment was heterogeneous; while pretreatment samples were prioritized, some patients were evaluated using specimens obtained during or after anti-HER2 therapy but before T-DXd. This variability may have introduced bias in the interpretation of CLDN18.2 expression. Although assessment immediately before T-DXd would be ideal, CLDN18.2 expression is known to decrease following prior chemotherapy.26,27 In the late-line setting, therefore, the number of patients with CLDN18.2-positive tumors just before T-DXd is expected to be small, making it challenging to secure sufficient cases for analysis. Additionally, due to its retrospective nature, all reported P values should be interpreted as nominal and not definitive. Finally, potential confounding factors, such as differences in prior treatments and baseline patient characteristics, could not be completely controlled. These limitations warrant cautious interpretation of the findings.
Conclusions
In this retrospective analysis of patients with mGC/GEJC treated with T-DXd, those with HER2 and CLDN18.2 positivity demonstrated poorer clinical outcomes compared with patients who were HER2-positive but CLDN18.2-negative. These findings suggest that CLDN18.2 expression may attenuate the clinical benefit of T-DXd and highlight the potential need for combined HER2- and CLDN18.2-directed strategies, which warrant evaluation in prospective clinical trials.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work, the authors used ChatGPT (https://chat.openai.com) in order to assist with language refinement during the drafting process. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Acknowledgments
Funding
None declared.
Disclosure
IN has received research funding from Astellas, Ono Pharmaceutical Co., Ltd., Merck Sharp & Dohme (MSD), Daiichi Sankyo, Boehringer Ingelheim, and Chugai Pharmaceutical Co., Ltd. Honoraria from Astellas, Ono Pharmaceutical Co., Ltd., Bristol Myers Squibb, Merck Sharp & Dohme (MSD), Daiichi Sankyo, and Chugai Pharmaceutical Co., Ltd outside the submitted work. YM has received honoraria from MSD and Chugai Pharmaceutical Co., Ltd., and Bristol-Myers Squibb outside the submitted work. S. Mishima has received honoraria from Taiho Pharmaceutical Co., Ltd., Chugai Pharmaceutical Co., Ltd., and Eli Lilly Co., Ltd outside the submitted work. DK has received honoraria from Takeda, Chugai, Lilly, MSD, Ono, Taiho, Bristol-Myers Squibb, Daiichi Sankyo, Pfizer, Eisai, and Merck Biopharma and research funding from Ono, MSD, Novartis, Servier, Janssen, IQVIA, Syneos Health, Cimic, and Cimicshiftzero outside the submitted work. AK has received honoraria from Daiichi Sankyo, Lilly, Ono, Taiho, Bristol-Myers Squibb, and Merck Serono Biopharma and research funding from Ono, MSD, Taiho, Bayer, Sumitomo Dainippon, and AstraZeneca outside the submitted work. TH has received honoraria from CytoGen, Inc. and Takata Pharmaceutical; consulting and advisory roles from Guardant Health outside the submitted work. YK has received personal fees for advisory roles from Taiho, Takeda, and Boehringer Ingelheim; honoraria from Taiho, Ono, Bayer, and Sanofi; and institutional grants or contracts with Taiho, Takeda, Ono, AbbVie, AstraZeneca, Boehringer Ingelheim, Incyte, Amgen, Chugai, GlaxoSmithKline, Genmab, Astellas, and Daiichi Sankyo outside the submitted work. HB has received research funding from Ono Pharmaceutical and honoraria from Ono Pharmaceutical, Taiho Pharmaceutical, and Eli Lilly Japan outside the submitted work. TK has received honoraria as an invited speaker and for manuscript writing from Bristol-Myers Squibb, Ono Pharmaceutical Co., Ltd., Covidien Japan, Inc., MSD K.K., TAIHO PHARMACEUTICAL CO., LTD., and Oncolys BioPharma Inc.; research funding from Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb, BeiGene Ltd., EPS Corporation, MSD K.K., Amgen Inc., SHIONOGI & Co., Ltd., and CHUGAI PHARMACEUTICAL CO., LTD outside the submitted work. TY has received honoraria from Chugai Pharmaceutical, Takeda Pharmaceutical, Merck Biopharma, Ono Pharmaceutical, and MSD K.K; consulting fee from Sumitomo Corp., and Indivumed; and research funding from Amgen, Bristol-Myers Squibb, Chugai Pharmaceutical, Daiichi Sankyo, Eisai, Exact Sciences, FALCO Biosystems, Medical & Biological Laboratories, Merus N.V., Miyarisan Pharmaceutical, Molecular Health GmbH, MSD, Nippon Boehringer Ingelheim, Ono Pharmaceutical, Pfizer Japan, Roche Diagnostics, Sanofi, Sysmex, Taiho Pharmaceutical, and Takeda Pharmaceutical outside the submitted work. TK received honoraria from MSD, Daiichi Sankyo, Astellas Pharma, Roche Diagnostics, FALCO biosystems, Agilent, Bristol-Myers Squibb Japan, Amgen, Myriad Genetics and Guardant Health. KS has received advisory fees from Bristol Myers Squibb, Takeda, Ono Pharmaceutical, Novartis, Daiichi Sankyo, Amgen, Boehringer Ingelheim, Merck Pharmaceutical, Astellas, Guardant Health Japan, Janssen, AstraZeneca, Zymeworks Biopharmaceuticals, ALX Oncology Inc., Bayer, GlaxoSmithKline K.K., HEALIOS K.K., Moderna Inc., and Arcus Biosciences Inc.; honoraria from Bristol-Myers Squibb, Ono Pharmaceutical, Janssen, Eli Lilly, Astellas, and AstraZeneca; and research funding from Astellas, Ono Pharmaceutical, Daiichi Sankyo, Taiho Pharmaceutical, Chugai, Merck Pharmaceutical, Amgen, Eisai, PRA Health Sciences, Syneos Health, AstraZeneca, PPD-SNBL K.K., and TORAY outside the submitted work. All other authors have declared no conflicts of interests.
Supplementary data
References
- 1.Bang Y.-J., Van Cutsem E., Feyereislova A., et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376(9742):687–697. doi: 10.1016/S0140-6736(10)61121-X. [DOI] [PubMed] [Google Scholar]
- 2.Sundar R., Nakayama I., Markar S.R., et al. Gastric cancer. Lancet. 2025;405(10494):2087–2102. doi: 10.1016/S0140-6736(25)00052-2. [DOI] [PubMed] [Google Scholar]
- 3.Aoki Y., Nakayama I., Shitara K. Human epidermal growth factor receptor 2 positive advanced gastric or esophagogastric adenocarcinoma: reflecting on the past to gain new insights. Curr Oncol Rep. 2025;27(1):15–29. doi: 10.1007/s11912-024-01626-2. [DOI] [PubMed] [Google Scholar]
- 4.Janjigian Y.Y., Kawazoe A., Bai Y., et al. Pembrolizumab plus trastuzumab and chemotherapy for HER2-positive gastric or gastro-oesophageal junction adenocarcinoma: interim analyses from the phase 3 KEYNOTE-811 randomised placebo-controlled trial. Lancet. 2023;402(10418):2197–2208. doi: 10.1016/S0140-6736(23)02033-0. [DOI] [PubMed] [Google Scholar]
- 5.Shitara K., Bang Y.-J., Iwasa S., et al. Trastuzumab deruxtecan in previously treated HER2-positive gastric cancer. N Engl J Med. 2020;382(25):2419–2430. doi: 10.1056/NEJMoa2004413. [DOI] [PubMed] [Google Scholar]
- 6.Van Cutsem E., di Bartolomeo M., Smyth E., et al. Trastuzumab deruxtecan in patients in the USA and Europe with HER2-positive advanced gastric or gastroesophageal junction cancer with disease progression on or after a trastuzumab-containing regimen (DESTINY-Gastric02): primary and updated analyses from a single-arm, phase 2 study. Lancet Oncol. 2023;24(7):744–756. doi: 10.1016/S1470-2045(23)00215-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shitara K., Van Cutsem E., Gümüş M., et al. Trastuzumab deruxtecan or ramucirumab plus paclitaxel in gastric cancer. N Engl J Med. 2025;393(4):336–348. doi: 10.1056/NEJMoa2503119. [DOI] [PubMed] [Google Scholar]
- 8.Nakayama I., Qi C., Chen Y., Nakamura Y., Shen L., Shitara K. Claudin 18.2 as a novel therapeutic target. Nat Rev Clin Oncol. 2024;21(5):354–369. doi: 10.1038/s41571-024-00874-2. [DOI] [PubMed] [Google Scholar]
- 9.Shitara K., Lordick F., Bang Y.-J., et al. Zolbetuximab plus mFOLFOX6 in patients with CLDN18.2-positive, HER2-negative, untreated, locally advanced unresectable or metastatic gastric or gastro-oesophageal junction adenocarcinoma (SPOTLIGHT): a multicentre, randomised, double-blind, phase 3 trial. Lancet. 2023;401(10389):1655–1668. doi: 10.1016/S0140-6736(23)00620-7. [DOI] [PubMed] [Google Scholar]
- 10.Shah M.A., Shitara K., Ajani J.A., et al. Zolbetuximab plus CAPOX in CLDN18.2-positive gastric or gastroesophageal junction adenocarcinoma: the randomized, phase 3 GLOW trial. Nat Med. 2023;29(8):2133–2141. doi: 10.1038/s41591-023-02465-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kubota Y., Kawazoe A., Mishima S., et al. Comprehensive clinical and molecular characterization of claudin 18.2 expression in advanced gastric or gastroesophageal junction cancer. ESMO Open. 2023;8(1) doi: 10.1016/j.esmoop.2022.100762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhang F., Nakayama I., Sakamoto N., et al. Impact of Claudin 18.2 expression on treatment outcomes of first-line immunochemotherapy in patients with HER2-negative, proficient MMR, PD-L1 CPS ≥ 1 metastatic gastric/gastroesophageal junction cancer. Ther Adv Med Oncol. 2025;17 doi: 10.1177/17588359251369042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Eisenhauer E.A., Therasse P., Bogaerts J., et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1) Eur J Cancer. 2009;45(2):228–247. doi: 10.1016/j.ejca.2008.10.026. [DOI] [PubMed] [Google Scholar]
- 14.Jubashi A., Nakayama I., Koganemaru S., et al. Prognostic and predictive factors for the efficacy and safety of trastuzumab deruxtecan in HER2-positive gastric or gastroesophageal junction cancer. Gastric Cancer. 2025;28(1):63–73. doi: 10.1007/s10120-024-01560-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kim T.-Y., Kwak Y., Nam S.K., et al. Clinicopathological analysis of claudin 18.2 focusing on intratumoral heterogeneity and survival in patients with metastatic or unresectable gastric cancer. ESMO Open. 2024;9(12) doi: 10.1016/j.esmoop.2024.104000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Nakamura Y., Kawazoe A., Lordick F., Janjigian Y., Shitara K. Biomarker-targeted therapies for advanced-stage gastric and gastro-oesophageal junction cancers: an emerging paradigm. Nat Rev Clin Oncol. 2021;18(8):473–487. doi: 10.1038/s41571-021-00492-2. [DOI] [PubMed] [Google Scholar]
- 17.Yagi S., Wakatsuki T., Yamamoto N., et al. Clinical significance of intratumoral HER2 heterogeneity on trastuzumab efficacy using endoscopic biopsy specimens in patients with advanced HER2-positive gastric cancer. Gastric Cancer. 2019;22(3):518–525. doi: 10.1007/s10120-018-0887-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim W.-H., Gomez-Izquierdo L., Vilardell F., et al. HER2 status in gastric and gastroesophageal junction cancer: results of the large, multinational HER-EAGLE study. Appl Immunohistochem Mol Morphol. 2018;26(4):239–245. doi: 10.1097/PAI.0000000000000423. [DOI] [PubMed] [Google Scholar]
- 19.Van Cutsem E., Bang Y.-J., Feng-Yi F., et al. HER2 screening data from ToGA: targeting HER2 in gastric and gastroesophageal junction cancer. Gastric Cancer. 2015;18(3):476–484. doi: 10.1007/s10120-014-0402-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nakayama I., Takahari D., Chin K., et al. Incidence, clinicopathological features, and clinical outcomes of low HER2-expressed, inoperable, advanced, or recurrent gastric/gastroesophageal junction adenocarcinoma. ESMO Open. 2023;8(4) doi: 10.1016/j.esmoop.2023.101582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shitara K., Xu R.-H., Ajani J.A., et al. Global prevalence of claudin 18 isoform 2 in tumors of patients with locally advanced unresectable or metastatic gastric or gastroesophageal junction adenocarcinoma. Gastric Cancer. 2024;27(5):1058–1068. doi: 10.1007/s10120-024-01518-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sheng T., Sundar R., Srivastava S., et al. Spatial profiling of patient-matched HER2-positive gastric cancer reveals resistance mechanisms to targeted therapy. Gut. 2025 doi: 10.1136/gutjnl-2024-334667. gutjnl-2024-334667. [DOI] [PubMed] [Google Scholar]
- 23.Kim H.-D., Shin J., Hyung J., et al. Survival outcomes of patients with gastric cancer treated with first-line nivolumab plus chemotherapy based on claudin 18.2 expression. Gastric Cancer. 2025;28(1):74–82. doi: 10.1007/s10120-024-01566-7. [DOI] [PubMed] [Google Scholar]
- 24.Ruan D.-Y., Liu F.-R., Wei X.-L., et al. Claudin 18.2-targeting antibody-drug conjugate CMG901 in patients with advanced gastric or gastro-oesophageal junction cancer (KYM901): a multicentre, open-label, single-arm, phase 1 trial. Lancet Oncol. 2025;26(2):227–238. doi: 10.1016/S1470-2045(24)00636-3. [DOI] [PubMed] [Google Scholar]
- 25.Yue J., Shao S., Zhou J., et al. A bispecific antibody targeting HER2 and CLDN18.2 eliminates gastric cancer cells expressing dual antigens by enhancing the immune effector function. Invest New Drugs. 2024;42(1):106–115. doi: 10.1007/s10637-024-01417-3. [DOI] [PubMed] [Google Scholar]
- 26.Yamamoto K., Nakayama I., Sakamoto N., et al. Temporal dynamics of CLDN18.2 expression following zolbetuximab treatment in advanced gastric cancer. ESMO Gastrointest Oncol. 2025;9 doi: 10.1016/j.esmogo.2025.100206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jubashi A., Nakayama I., Sakamoto N., et al. Prognostic indicators of preoperative fluorouracil, leucovorin, oxaliplatin, and docetaxel efficacy in locally advanced gastroesophageal and gastric cancer: integrating biomarker analysis and clinicopathological factors. JCO Precis Oncol. 2025;9 doi: 10.1200/PO-24-00925. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.