Abstract
Objective
Immunotherapeutic outcomes and clinical characteristics of claudin 18 isoform 2 positive (CLDN18.2-positive) gastric cancer (GC) vary in different clinical studies, making it difficult to optimize anti-CLDN18.2 therapy. We conducted a retrospective analysis to explore the association of CLDN18.2 expression with clinicopathological characteristics and immunotherapeutic outcomes in GC.
Methods
A total of 536 advanced GC patients from 2019 to 2021 in the CT041-CG4006 and CT041-ST-01 clinical trials were included in the analysis. CLDN18.2 expression on ≥40% of tumor cells (2+, 40%) and CLDN18.2 expression on ≥70% of tumor cells (2+, 70%) were considered the two levels of positively expressed GC. The clinicopathological characteristics and immunotherapy outcomes of GC patients were analyzed according to CLDN18.2 expression status.
Results
CLDN18.2 was expressed in 57.6% (cut-off: 2+, 40%) and 48.9% (cut-off: 2+, 70%) of patients. Programmed death-ligand 1 (PD-L1) and CLDN18.2 were co-expressed in 19.8% [combined positive score (CPS)≥1, CLDN18.2 (cut-off: 2+, 40%)] and 17.2% [CPS≥5, CLDN18.2 (cut-off: 2+, 70%)] of patients. CLDN18.2 expression positively correlated with younger age, female sex, non-gastroesophageal junction (non-GEJ), and diffuse phenotype (P<0.001). HER2 and PD-L1 expression were significantly lower in CLDN18.2-positive GC (both P<0.05). Uterine adnexa metastasis (P<0.001) was more frequent and liver metastasis (P<0.001) was less common in CLDN18.2-positive GC. Overall survival and immunotherapy-related progression-free survival (irPFS) were inferior in the CLDN18.2-positive group.
Conclusions
CLDN18.2-positive GC is associated with poor prognosis and worse immunotherapeutic outcomes. The combination of anti-CLDN18.2 therapy, anti-PD-L1/PD-1 therapy, and chemotherapy for GC requires further investigation.
Keywords: Gastric cancer, claudin18.2, CT041, PD-L1, immunotherapy
Introduction
Gastric cancer (GC) is the leading cause of cancer-related deaths in China (1). Despite many studies on drug development, anti-human epidermal growth factor receptor 2 [HER2 (also known as ERBB2)] and immune checkpoint inhibitors (ICIs) are the only molecular targeted regimens validated for the first-line treatment of GC. Claudin18 (CLDN18) is a tight junction protein required for cell-cell adhesion. CLDN18 isoform 2 (CLDN18.2) is confined to differentiated gastric epithelial cells and selectively regulates Na+ and H+ ions (2). CLDN18.2 is embedded in the tight junctions of the normal gastric mucosa and is widely expressed during tumorigenesis (2-4), which make it a promising target for cancer treatment.
Zolbetuximab, a monoclonal chimeric antibody that binds to CLDN18.2, causes antibody-dependent cytotoxicity and complement-dependent cytotoxicity, leading to tumor cell death. Based on the outcome of the FAST study (5), zolbetuximab plus EOX (epirubicin, oxaliplatin, and capecitabine), significantly prolonged progression-free survival (PFS) and overall survival (OS) in CLDN18.2-positive (CLDN18.2 ≥2+, 40%) gastric/gastroesophageal junction (G/GEJ) adenocarcinoma patients compared to those with EOX chemotherapy alone. Subsequent phase III studies, including SPOTLIGHT and GLOW (6), have also reported efficacy of zolbetuximab for first-line therapy in CLDN18.2-positive tumors (CLDN18.2 ≥2+, 75%). In addition, the multicenter phase I CT041-CG4006 trial (NCT03874897) substantiated the outstanding antitumor activity of CLDN18.2-specific chimeric antigen receptor (CAR) cells against gastrointestinal tumors (7). Therefore, CLDN18.2 is currently considered the most promising target, especially for HER2-negative microsatellite-stable GC.
With the rapid development of CLDN18.2-targeted therapies, it is crucial to explore the clinical and molecular characteristics of CLDN18.2-positive GC. The correlation between clinicopathological characteristics, CLDN18.2 expression, and cancer prognosis remains controversial (4,8-10). Previous studies are not consistent with the definition of CLDN18.2-positive expression (CLDN18.2 ≥2+, 40% or CLDN18.2 ≥2+, 75%), populations studies, or types of cancer analyzed. Furthermore, with the advancement of immunotherapy for GC, it is essential to investigate the impact of CLDN18.2 on immunotherapy outcomes. It is crucial to explore the clinical and molecular characteristics, and immunotherapy outcomes of CLDN18.2 in positive GC with the rapid development of CLDN18.2-targeted therapies. Therefore, this study aimed to define positive CLDN18.2 expression at two levels and investigate the association of CLDN18.2 expression with clinicopathological characteristics, cancer prognosis, and immunotherapeutic outcomes in GC.
Materials and methods
Patient selection
A total of 541 patients with advanced GC screened between June 2019 and April 2021 from the CT041-CG4006 and CT041-ST-01 trials (No. NCT03874897 and No. NCT04581473, respectively) at Peking University Cancer Hospital, China, were included in this study. The eligible criteria in this study were as follows: 1) 18−75 years old; 2) unresectable or metastatic GC (including both GEJ and non-GEJ types); 3) histopathologically confirmed adenocarcinoma; 4) an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0−1; and 5) with available pre-treatment primary tumor specimens. Five patients with no tissue specimens before therapy were excluded. Formalin-fixed paraffin-embedded (FFPE) surgical or endoscopic tissue samples were collected from each patient before starting therapy. Informed consent was obtained from all participants. This study was approved by the Institutional Ethics Committee of Peking University Cancer Hospital & Institute (No. 2019KT109) and conducted in accordance with the principles of the Declaration of Helsinki. The clinical and molecular characteristics were retrospectively retrieved. HER2-positive status was considered when a pathological exhibition of either immunohistochemistry (IHC) 3+ or IHC 2+ with ERBB2 amplification was observed. The expression levels of MLH1, MSH2, MSH6, and PMS2 were evaluated by IHC to determine mismatch repair (MMR) status. Epstein-Barr-encoded RNA1 (EBER1) expression was detected using in situ hybridization (ISH) to determine Epstein-Barr virus (EBV) status. Specimens in which EBER nuclear expression was observed in 20% or more of the tumor cells were considered EBER positive. Tumor mutation burden was evaluated in tumor tissues or circulating tumor DNA using next-generation sequencing (NGS). The combined positive score (CPS) of programmed death-ligand 1 (PD-L1) was the percentage of positive cells among all tumor cells.
IHC
FFPE tissue sections (4 µm thickness) were used for immunostaining. Automated immunostaining was performed using a Leica BOND-III biosystem (Leica Microsystems Inc., Deerfield, IL, USA) at Carlsbad Bio-Sample Analysis Laboratory (Shanghai, China). Specimens were stained with anti-CLDN18.2 (monoclonal mouse antibody; clone 14F8; CARsgen; working concentration 2.0 μg/mL), using the Bond Polymer Refine Detection system (Leica, Deerfield, IL, USA) and counterstained with hematoxylin according to the manufacturer’s instructions. The presence of membrane staining in tumor cells indicated CLDN18.2-positive cells (Supplementary Figure S1A). Since the current definition of CLDN18.2-positive GC is still not standardized, therefore, we referred to the FAST study and defined CLDN18.2-positive GC by two levels respectively (≥2+ membrane staining intensity in ≥40% of tumor cells and ≥2+ membrane staining intensity in ≥70% of tumor cells) (5). HER2, PD-L1, EBV and MMR status were detected in Peking University Cancer Hospital and reported by two pathologists independently. HER2 protein expression was detected using anti-HER2/neu-antibody (4B5; Roche, Basel, Switzerland) in a working dilution. HER2 amplification defined as a ratio HER2/CEP17≥2.0 was determined using Ventana HER2 dual-colour ISH assay (DISH BenchMark XT). PD-L1 staining was performed with an anti-human PD-L1 monoclonal antibody (22c3; Dako, Glostrup, Denmark) in a working dilution. EBER was performed by using fluorescein-labeled oligonucleotide probes (INFORM EBER Probe; Ventana). MMR protein detection was carried out using primary antibodies against MLH1 (clone GM002, mouse monoclonal antibody, catalogue numbers: GT230407, working solution, Gene Tech, Shanghai, China), MSH2 (clone RED2, rabbit monoclonal antibody, catalogue numbers: GT231007, working solution, Gene Tech), MSH6 (clone EP49, rabbit monoclonal antibody, catalogue numbers: GT219507, working solution, Gene Tech), and PSM2 (clone EP51, rabbit monoclonal antibody, catalogue numbers: GT215907, working solution, Gene Tech).
Figure S1.
Study design. (A) Interpretation of CLDN18.2 expression in tumor and normal gastric mucosa (×20). CLDN18.2 staining on tumor cell membrane was evaluated as 1+: weak staining; 2+: moderate to strong staining; 3+: strong staining; (B) Study flow chart. IHC, immunohistochemistry; CLDN18.2, claudin 18 isoform 2; irPFS, immunotherapy-related progression-free survival.
Outcomes and follow-up methods
We evaluated clinical outcomes including tumor response, PFS and OS. PFS was defined as the time from the initial treatment to the date that clinical disease progression was diagnosed or censored at the last follow-up. OS was defined as the time from the diagnosis of phase IV disease to the date of death or censored at the last follow-up. Objective response rate (ORR) was defined as the percentage of patients with a confirmed complete response (CR) or partial response (PR). The disease control rate (DCR) was defined as the proportion of patients with the best response to CR, PR, and stable disease (SD). If the patient signed the informed consent and successfully enrolled, we will conduct the visit according to the visit plan specified in the clinical research protocol. If the patient did not enroll, we will conduct survival follow-ups with patients via telephone to record their disease progression status and survival status at a frequency of once every six months, from the beginning of this study to May 2023.
Statistical analysis
Data were statistically analyzed using SPSS software (Version 21.0; IBM Corp., NY, USA) and the R programming language (Version 4.1.1; The R Project for Statistical Computing, www.r-project.org). The characteristics of the CLDN18.2-positive (CLDN18.2-pos) and CLDN18.2-negative (CLDN18.2-neg) groups were compared using the χ2 test. In the PFS/OS curves Kaplan-Meier plotting, cases were split into CLDN18.2-positive, CLDN18.2-negative, and CPS. The log-rank test was used to compare survival curves, exhibiting median survival with 95% confidence intervals (95% CIs). Univariate and multivariate Cox regression analyses were used to explore prognostic factors for OS and calculate hazard ratios (HRs) with 95% CIs. Immunotherapeutic ORR/DCR of the CLDN18.2-positive and CLDN18.2-negative groups was compared using χ2 test or Fisher’s exact test. All tests were two-sided, and a P<0.05 was considered statistically significant.
Results
Clinicopathological significance and prognosis of CLDN18.2 expression in GC
As demonstrated in the overall workflow (Supplementary Figure S1B), 536 patients with advanced GC underwent CLDN18.2 IHC staining. Characteristics of these patients are showed in Supplementary Table S1. Identical features were observed in CLDN18.2-positive GC regardless of the cut-off value (Tables 1,2). CLDN18.2 was positive in 57.6% (cut-off: 2+, 40%) and 48.9% (cut-off: 2+, 70%) of patients, which is higher than that of the FAST study (48.0%) (11). In our research, on univariate and multivariate analyses, CLDN18.2 expression was related to inferior OS regardless of the cut-off value for CLDN18.2 positivity (Figure 1). CLDN18.2-positive GC exhibited similar features to diffuse GC including a predilection for younger females (<60 years old), tumor localization in the non-EGJ (P<0.001), negative HER2 expression (P=0.001) and poor survival (Figure 2A,B; Tables 1−3; Supplementary Table S2), which indicated the great potential of CLDN18.2 as a therapeutic target for diffuse GC. What’s more, unique metastatic features were observed in CLDN18.2-positive patients: metastasis to the uterine adnexa was more common, while metastasis to the liver was rarer (P<0.001). Considering the important role of ICIs in GC, we also analyzed the relationship between CLDN18.2 and PD-L1. As demonstrated in Tables 1,2, PD-L1 positivity was approximately 10% lower in the CLDN18.2-positive group compared to the CLDN18.2-negative group, irrespective of the cut-off values of CPS and CLDN18.2. The co-expression of HER2, PD-L1, and CLDN18.2 was further investigated (Table 4), showing that the expression of CLDN18.2 was exclusive to HER2 and PD-L1.
Table S1. Characteristics of enrolled patients.
| Variables | No. of patients |
| NA, not available; EGJ, gastroesophageal junction; HER2, human epidermal growth factor receptor 2; PD-L1, programmed death-ligand 1; CPS, combined positive score; EBV, Epstein-Barr virus; pMMR, mismatch repair-proficient; dMMR, mismatch repair-deficient; TMB, tumor mutational burden; muts, mutations. | |
| Total | 536 |
| Age (year) | |
| <60 | 364 |
| ≥60 | 170 |
| NA | 2 |
| Sex | |
| Male | 336 |
| Female | 198 |
| NA | 2 |
| Histopathology | |
| Non-EGJ | 411 |
| EGJ | 113 |
| NA | 12 |
| Lauren | |
| Intestinal | 148 |
| Diffuse | 189 |
| Mixed | 100 |
| NA | 99 |
| Differentiation | |
| High | 5 |
| Moderate | 97 |
| Poor | 397 |
| NA | 37 |
| Metastasis site | |
| Liver | 144 |
| Lung | 51 |
| Peritoneum | 321 |
| Uterine adnexa | 90 |
| Lymph node | 28 |
| NA | 36 |
| HER2 status | |
| Positive | 53 |
| Negative | 448 |
| NA | 35 |
| PD-L1 status | |
| CPS≥1 | 203 |
| CPS<1 | 285 |
| NA | 48 |
| EBV status | |
| Negative | 403 |
| Positive | 6 |
| NA | 127 |
| MMR status | |
| pMMR | 461 |
| dMMR | 7 |
| NA | 68 |
| TMB (muts/Mb) | |
| TMB<10 | 96 |
| TMB≥10 | 23 |
| NA | 417 |
Table 1. Clinicopathologic and molecular features of CLDN18.2-positive or negative (cut-off: 2+, 40%) GC patients.
| Characteristics | n (%) | P | |
| CLDN18.2-neg | CLDN18.2-pos | ||
| CLDN18.2, claudin 18 isoform 2; GC, gastric cancer; EGJ, gastroesophageal junction; HER2, epidermal growth factor receptor 2; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; amp, amplification; PD-L1, programmed death-ligand 1; CPS, combined positive score; EBV, Epstein-Barr virus; dMMR, mismatch repair-deficient; pMMR, mismatch repair-proficient; TMB, tumor mutational burden; muts, mutations; pos, positive; neg, negative. | |||
| Total | 227 (42.4) | 309 (57.6) | |
| Age (year) | |||
| <60 | 128 (56.4) | 236 (76.9) | <0.001 |
| ≥60 | 99 (43.6) | 71 (23.1) | |
| Sex | |||
| Female | 58 (25.8) | 140 (45.3) | <0.001 |
| Male | 167 (74.2) | 169 (54.7) | |
| Histopathology | |||
| EGJ | 66 (29.6) | 47 (15.6) | <0.001 |
| Non-EGJ | 157 (70.4) | 254 (84.4) | |
| Lauren | |||
| Diffuse | 63 (36.4) | 126 (47.7) | <0.001 |
| Intestinal | 78 (45.1) | 70 (26.5) | |
| Mixed | 32 (18.5) | 68 (25.8) | |
| Differentiation | |||
| High | 1 (0.5) | 4 (1.4) | 0.067 |
| Moderate | 50 (23.9) | 47 (16.2) | |
| Poor | 158 (75.6) | 239 (82.4) | |
| Metastasis site | |||
| Liver | 78 (30.6) | 66 (17.4) | <0.001 |
| Lung | 26 (10.2) | 25 (6.6) | 0.102 |
| Peritoneum | 124 (48.6) | 197 (52.0) | 0.408 |
| Uterine adnexa | 16 (6.3) | 74 (19.5) | <0.001 |
| Lymph node | 11 (4.3) | 17 (4.5) | 0.918 |
| HER2 status | |||
| Negative | 176 (84.2) | 272 (93.2) | 0.001 |
| Positive | 33 (15.8) | 20 (6.8) | |
| IHC 2+ FISH amp |
10 (30.2) | 7 (35.0) | 0.723 |
| IHC 3+ | 23 (69.7) | 13 (65.0) | |
| PD-L1 status | |||
| CPS<1 | 100 (49.0) | 187 (65.8) | 0.001 |
| CPS≥1 | 104 (51.0) | 97 (34.2) | |
| CPS<5 | 135 (66.2) | 223 (78.5) | 0.002 |
| CPS≥5 | 69 (33.8) | 61 (21.5) | |
| CPS<10 | 155 (76.0) | 246 (86.6) | 0.002 |
| CPS≥10 | 49 (24.0) | 38 (13.4) | |
| EBV status | |||
| Negative | 168 (99.4) | 235 (97.9) | 0.217 |
| Positive | 1 (0.6) | 5 (2.1) | |
| MMR status | |||
| dMMR | 3 (1.6) | 4 (1.4) | 0.921 |
| pMMR | 189 (98.4) | 272 (98.6) | |
| TMB (muts/Mb) | |||
| TMB<10 | 41 (77.4) | 55 (83.3) | 0.412 |
| TMB≥10 | 12 (22.6) | 11 (16.7) | |
Table 2. Clinicopathologic and molecular features of CLDN18.2-positive or negative (cut-off: 2+, 70%) GC patients.
| Characteristics | n (%) | P | |
| CLDN18.2-neg | CLDN18.2-pos | ||
| CLDN18.2, claudin 18 isoform 2; GC, gastric cancer; EGJ, gastroesophageal junction; HER2, epidermal growth factor receptor 2; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; amp, amplification; PD-L1, programmed death-ligand 1; CPS, combined positive score; EBV, Epstein-Barr virus; dMMR, mismatch repair-deficient; pMMR, mismatch repair-proficient; TMB, tumor mutational burden; muts, mutations; pos, positive; neg, negative. | |||
| Total | 274 (51.1) | 262 (48.9) | |
| Age (year) | |||
| <60 | 161 (58.8) | 203 (78.1) | <0.001 |
| ≥60 | 113 (41.2) | 57 (21.9) | |
| Sex | |||
| Female | 73 (26.8) | 125 (47.7) | <0.001 |
| Male | 199 (73.2) | 137 (52.3) | |
| Histopathology | |||
| EGJ | 76 (28.1) | 37 (14.6) | <0.001 |
| Non-EGJ | 194 (71.9) | 217 (85.4) | |
| Lauren | |||
| Diffuse | 74 (34.6) | 115 (51.6) | <0.001 |
| Intestinal | 94 (43.9) | 54 (24.2) | |
| Mixed | 46 (21.5) | 54 (24.2) | |
| Differentiation | |||
| High | 2 (0.8) | 3 (1.2) | 0.023 |
| Moderate | 61 (24.2) | 36 (14.6) | |
| Poor | 189 (75.0) | 208 (84.2) | |
| Metastasis site | |||
| Liver | 91 (29.0) | 53 (16.6) | <0.001 |
| Lung | 33 (10.5) | 18 (5.6) | 0.024 |
| Peritoneum | 150 (47.8) | 171 (53.4) | 0.154 |
| Uterine adnexa | 26 (8.3) | 64 (20.0) | <0.001 |
| Lymph node | 14 (4.5) | 14 (4.4) | 0.959 |
| HER2 status | |||
| Negative | 215 (85.0) | 233 (94.0) | 0.001 |
| Positive | 38 (15.0) | 15 (6.0) | |
| IHC 2+ FISH amp |
12 (31.60) | 5 (33.3) | 0.902 |
| IHC 3+ | 26 (68.40) | 10 (66.7) | |
| PD-L1 status | |||
| CPS<1 | 128 (51.8) | 157 (65.1) | 0.003 |
| CPS≥1 | 119 (48.2) | 84 (34.9) | |
| CPS<5 | 168 (68.0) | 189 (78.4) | 0.009 |
| CPS≥5 | 79 (32.0) | 52 (21.6) | |
| CPS<10 | 191 (77.6) | 210 (87.1) | 0.006 |
| CPS≥10 | 55 (22.4) | 31 (12.9) | |
| EBV status | |||
| Negative | 203 (99.5) | 200 (97.6) | 0.101 |
| Positive | 1 (0.5) | 5 (2.4) | |
| MMR status | |||
| dMMR | 3 (0.9) | 4 (1.7) | 0.420 |
| pMMR | 230 (99.1) | 231 (98.3) | |
| TMB (muts/Mb) | |||
| TMB<10 | 46 (78.0) | 50 (83.3) | 0.458 |
| TMB≥10 | 13 (22.0) | 10 (16.7) | |
Figure 1.
Kaplan-Meier plot of OS according to CLDN18.2 status. (A) CLDN18.2 expression cut-off value was 2+, 40% (P=0.003); (B) CLDN18.2 expression cut-off value was 2+, 70% (P=0.013). OS, overall survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; 95% CI, 95% confidence interval.
Figure 2.
Kaplan-Meier plot of irPFS based on CLDN18.2 expression. Total: CLDN18.2 expression cut-off value was 2+, 40% (P=0.050) (A) and 2+, 70% (P=0.077) (B); First-line: CLDN18.2 expression cut-off value was 2+, 40% (P=0.014) (C) and 2+, 70% (P=0.040) (D). irPFS, immunotherapy-related progression-free survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; 95% CI, 95% confidence interval.
Table S2. Univariate analysis for OS.
| Parameters | HR | 95% CI | P |
| OS, overall survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; EGJ, gastroesophageal junction; PD-L1, programmed death-ligand 1; CPS, combined positive score; HER2, human epidermal growth factor receptor 2; EBV, Epstein-Barr virus; pMMR, mismatch repair-proficient; dMMR, mismatch repair-deficient; TMB, tumor mutational burden; HR, hazard ratio; 95% CI, 95% confidence interval. | |||
| CLDN18.2 cut-off value ≥2+, 40% | |||
| Age ≥60 vs. <60 (year) | 0.951 | 0.764−1.183 | 0.650 |
| Male vs. Female | 0.797 | 0.628−1.012 | 0.062 |
| CLDN18.2-pos vs. -neg | 1.405 | 1.125−1.755 | 0.002 |
| Non-EGJ vs. EGJ | 1.080 | 0.587−1.400 | 0.561 |
| Intestinal vs. Diffuse | 0.648 | 0.496−0.847 | 0.001 |
| PD-L1 CPS≥1 vs. <1 | 0.901 | 0.695−1.169 | 0.433 |
| HER2-pos vs. -neg | 0.630 | 0.443−0.896 | 0.010 |
| EBV-pos vs. -neg | 0.718 | 0.267−1.931 | 0.512 |
| pMMR vs. dMMR | 1.307 | 0.487−3.510 | 0.595 |
| TMB≥10 vs. <10 | 1.352 | 0.766−2.386 | 0.298 |
| CLDN18.2 cut-off value ≥2+, 70% | |||
| Age ≥60 vs. <60 (year) | 0.951 | 0.764−1.183 | 0.650 |
| Male vs. Female | 0.797 | 0.628−1.012 | 0.062 |
| CLDN18.2-pos vs. -neg | 1.314 | 1.062−1.627 | 0.012 |
| Non-EGJ vs. EGJ | 1.080 | 0.587−1.400 | 0.561 |
| Intestinal vs. Diffuse | 0.648 | 0.496−0.847 | 0.001 |
| PD-L1 CPS≥1 vs. <1 | 0.901 | 0.695−1.169 | 0.433 |
| HER2-pos vs. -neg | 0.630 | 0.443−0.896 | 0.010 |
| EBV-pos vs. -neg | 0.718 | 0.267−1.931 | 0.512 |
| pMMR vs. dMMR | 1.307 | 0.487−3.510 | 0.595 |
| TMB≥10 vs. <10 | 1.352 | 0.766−2.386 | 0.298 |
Table 4. Co-expression of CLDN18.2, HER2 and PD-L1.
| Characteristics | n (%) | |
| CLDN18.2 ≥2+, 40% | CLDN18.2 ≥2+, 70% | |
| CLDN18.2, claudin 18 isoform 2; HER2, epidermal growth factor receptor 2; PD-L1, programmed death-ligand 1; pos, positive; CPS, combined positive score. | ||
| HER2-pos | 20 (4.0) | 15 (3.0) |
| PD-L1-pos | ||
| CPS≥1 | 97 (19.8) | 84 (17.2) |
| CPS≥5 | 61 (12.5) | 52 (10.7) |
| CPS≥10 | 38 (7.8) | 31 (6.4) |
Table 3. Multivariate analysis for OS.
| Parameters | HR | 95% CI | P |
| OS, overall survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; EGJ, gastroesophageal junction; PD-L1, programmed death-ligand 1; CPS, combined positive score; HER2, epidermal growth factor receptor 2; EBV, Epstein-Barr virus; dMMR, mismatch repair-deficient; pMMR, mismatch repair-proficient; TMB, tumor mutational burden; HR, hazard ratio; 95% CI, 95% confidence interval. | |||
| CLDN18.2 cut-off ≥2+, 40% | |||
| Age ≥60 vs. <60 (year) | 0.849 | 0.653−1.104 | 0.222 |
| Male vs. Female | 0.773 | 0.606−0.987 | 0.039 |
| CLDN18.2-pos vs. -neg | 1.362 | 1.062−1.747 | 0.015 |
| Non-EGJ vs. EGJ | 0.920 | 0.697−1.214 | 0.553 |
| Intestinal vs. Diffuse | 0.719 | 0.535−0.968 | 0.030 |
| PD-L1 CPS≥1 vs. <1 | 0.927 | 0.702−1.224 | 0.595 |
| HER2-pos vs.-neg | 0.737 | 0.504−1.076 | 0.114 |
| EBV-pos vs. -neg | 0.294 | 0.579−0.209 | 1.605 |
| pMMR vs. dMMR | 1.244 | 0.457−3.385 | 0.669 |
| TMB≥10 vs. <10 | 1.370 | 0.751−2.501 | 0.305 |
| CLDN18.2 cut-off ≥2+, 70% | |||
| Age ≥60 vs. <60 (year) | 0.825 | 0.636−1.071 | 0.148 |
| Male vs. Female | 0.775 | 0.607−0.991 | 0.042 |
| CLDN18.2-pos vs. -neg | 1.278 | 1.007−1.623 | 0.044 |
| Non-EGJ vs. EGJ | 0.928 | 0.703−1.225 | 0.597 |
| Intestinal vs. Diffuse | 0.725 | 0.538−0.979 | 0.036 |
| PD-L1 CPS≥1 vs.<1 | 0.924 | 0.699−1.220 | 0.576 |
| HER2-pos vs. -neg | 0.720 | 0.493−1.051 | 0.089 |
| EBV-pos vs. -neg | 0.579 | 0.208−1.607 | 0.294 |
| pMMR vs. dMMR | 1.206 | 0.443−3.287 | 0.714 |
| TMB≥10 vs. <10 | 1.363 | 0.746−2.491 | 0.313 |
Correlation of CLDN18.2 expression with immunotherapeutic response and prognosis
Of the 536 patients, 83 received immunotherapy. As shown in Supplementary Table S3, 47 of 83 (56.6%) patients received first-line immunotherapy, and 36 of 83 (43.4%) patients received second-line or later-line immunotherapy. Patient characteristics are summarized in Supplementary Table S4. As depicted in Figure 2A,B, CLDN18.2 expression exerted adverse impact on immunotherapy-related PFS (irPFS) (cut-off: 2+, 40%: 6.2 vs. 7.0 months, P=0.05; cut-off: 2+, 70%: 5.9 vs. 7.0 months, P=0.077). When patients were stratified by therapeutic lines, the same conclusion was drawn for those who received first-line immunotherapy (Figure 2C,D). Although the P value was not statistically significant, CLDN18.2-positive patients tended to have inferior second or later-line irPFS (Supplementary Figure S2). When patients were further stratified by PD-L1 CPS and CLDN18.2 (Supplementary Figure S3), no statistically significant differences were found among the groups. Univariate and multivariate analyses indicated that positive CLDN18.2 expression and therapeutic line were adverse factors for irPFS (Supplementary Table S5, Table 5). Accordingly, the ORR/DCR tended to be lower in CLDN18.2-positive patients. These results indicated that CLDN18.2 was an adverse risk factor of immunotherapy in advanced GC (Supplementary Table S6).
Table S3. Therapeutic information of patients receiving ICIs.
| Therapeutic information | n (%) |
| ICI, immune checkpoint inhibitor; HER2, human epidermal growth factor receptor 2; ORR, objective response rate; DCR, disease control rate; irPFS, immunotherapy-related progression-free survival; 95% CI, 95% confidence interval. | |
| Line | |
| 1st | 47 (56.6) |
| ≥2nd | 36 (43.4) |
| Regimen | |
| ICI | 12 (14.5) |
| ICI+ICI | 3 (3.6) |
| ICI + chemotherapy | 44 (53.0) |
| ICI + anti-HER2 + chemotherapy | 11 (13.3) |
| ICI + anti-angiogenesis + chemotherapy | 13 (15.7) |
| ORR (%) | |
| 1st-line | 46.8 |
| ≥2nd-line | 22.2 |
| DCR (%) | |
| 1st-line | 87.2 |
| 2nd-line | 55.6 |
| irPFS [median (95% CI)] (month) | |
| 1st-line | 8.1 (7.0−10.1) |
| ≥2nd-line | 3.6 (2.6−5.9) |
Table S4. Characteristics of patients receiving ICIs.
| Characteristics | n (%) |
| ICI, immune checkpoint inhibitor; CLDN18.2, claudin 18 isoform 2; dMMR, mismatch repair-deficient; pMMR, mismatch repair-proficient; NA, not available; EBV, Epstein-Barr virus; HER2, human epidermal growth factor receptor 2; TMB, tumor mutational burden; CPS, combined positive score; PD-L1, programmed death-ligand 1. | |
| CLDN18.2 status | |
| CLDN18.2 ≥2+, 40% | 52 (62.7) |
| CLDN18.2 <2+, 40% | 31 (37.3) |
| CLDN18.2 ≥2+, 70% | 42 (50.6) |
| CLDN18.2 <2+, 70% | 41 (49.4) |
| MMR status | |
| dMMR | 1 (1.2) |
| pMMR | 77 (92.8) |
| NA | 5 (6.0) |
| EBV status | |
| Positive | 3 (3.6) |
| Negative | 66 (79.5) |
| NA | 14 (16.9) |
| HER2 status | |
| Positive | 12 (14.5) |
| Negative | 70 (84.5) |
| NA | 1 (1.2) |
| TMB status | |
| TMB≥10 | 4 (4.8) |
| TMB<10 | 13 (15.7) |
| NA | 66 (79.5) |
| CPS status | |
| PD-L1 CPS≥1 | 44 (53.0) |
| PD-L1 CPS<1 | 35 (42.2) |
| NA | 4 (4.8) |
| PD-L1 CPS≥5 | 34 (41.0) |
| PD-L1 CPS<5 | 45 (54.2) |
| NA | 4 (4.8) |
| PD-L1 CPS≥10 | 26 (31.3) |
| PD-L1 CPS<10 | 53 (63.9) |
| NA | 4 (4.8) |
Figure S2.
Kaplan-Meier plot of second line irPFS based on CLDN18.2 expression. (A) CLDN18.2 expression cut-off value was 2+, 40% (P=0.114); (B) CLDN18.2 expression cut-off value was 2+, 70% (P=0.247). irPFS, immunotherapy-related progression-free survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; 95% CI, 95% confidence interval.
Figure S3.
Kalan-Meier plot of irPFS based on PD-L1 CPS status and CLDN18.2 expression. (A) CLDN18.2: 2+, 40%; CPS: 1; (B) CLDN18.2: 2+, 40%; CPS: 5; (C) CLDN18.2: 2+, 40%; CPS: 10; (D) CLDN18.2: 2+, 70%; CPS: 1; (E) CLDN18.2: 2+, 70%; CPS: 5; (F) CLDN18.2: 2+, 40%; CPS: 10. irPFS, immunotherapy-related progression-free survival; CLDN18.2, claudin 18 isoform 2; CPS, combined positive score; 95% CI, 95% confidence interval; pos, positive; neg, negative.
Table S5. Univariate analysis for irPFS with ICIs therapy.
| Parameters | HR | 95% CI | P |
| irPFS, immunotherapy-related progression-free survival; ICI, immune checkpoint inhibitor; CLDN18.2, claudin 18 isoform 2; EBV, Epstein-Barr virus; pos, positive; neg, negative; HER2, epidermal growth factor receptor 2; TMB, tumor mutational burden; PD-L1, programmed death-ligand 1; CPS, combined positive score; HR, hazard ratio; 95% CI, 95% confidence interval. | |||
| CLDN18.2 cut-off ≥2+, 40% | |||
| EBV-pos vs. -neg | 1.361 | 0.424−4.372 | 0.605 |
| HER2-pos vs. -neg | 0.601 | 0.309−1.171 | 0.135 |
| TMB≥10 vs. <10 | 1.089 | 0.344−3.452 | 0.884 |
| Therapeutic line (≥2nd vs. 1st) | 3.326 | 1.910−5.792 | <0.001 |
| CLDN18.2-pos vs. -neg | 1.645 | 1.002−2.700 | 0.049 |
| PD-L1 CPS≥1 vs. <1 | 0.889 | 0.552−1.430 | 0.627 |
| CLDN18.2 cut-off ≥2+, 70% | |||
| EBV-pos vs. -neg | 1.361 | 0.424−4.372 | 0.605 |
| HER2-pos vs. -neg | 0.601 | 0.309−1.171 | 0.135 |
| TMB≥10 vs. <10 | 1.089 | 0.344−3.452 | 0.884 |
| Therapeutic line (≥2nd vs. 1st) | 3.326 | 1.910−5.792 | <0.001 |
| CLDN18.2-pos vs. -neg | 1.531 | 0.953−2.458 | 0.078 |
| PD-L1 CPS≥1 vs. <1 | 0.889 | 0.552−1.430 | 0.627 |
Table 5. Multivariate analysis for irPFS with ICIs therapy.
| Parameters | HR | 95% CI | P |
| irPFS, immunotherapy-related progression-free survival; ICI, immune checkpoint inhibitor; CLDN18.2, claudin 18 isoform 2; EBV, Epstein-Barr virus; pos, positive; neg, negative; HER2, epidermal growth factor receptor 2; TMB, tumor mutational burden; PD-L1, programmed death-ligand 1; CPS, combined positive score; HR, hazard ratio; 95% CI, 95% confidence interval. | |||
| CLDN18.2 cut-off ≥2+, 40% | |||
| EBV-pos vs. -neg | 1.094 | 0.280−4.268 | 0.897 |
| HER2-pos vs.-neg | 0.506 | 0.227−1.126 | 0.095 |
| TMB≥10 vs. <10 | 1.243 | 0.361−4.285 | 0.730 |
| Therapeutic line (≥2nd vs. 1st) | 5.640 | 2.772−11.474 | <0.001 |
| CLDN18.2-pos vs. -neg | 1.714 | 0.955−3.076 | 0.071 |
| PD-L1 CPS≥1 vs. <1 | 0.961 | 0.384−2.520 | 0.972 |
| CLDN18.2 cut-off ≥2+, 70% | |||
| EBV-pos vs. -neg | 1.173 | 0.302−4.560 | 0.818 |
| HER2-pos vs.-neg | 0.463 | 0.220−1.045 | 0.054 |
| TMB≥10 vs. <10 | 1.202 | 0.350−4.127 | 0.770 |
| Therapeutic line (≥2nd vs. 1st) | 5.407 | 2.677−10.922 | <0.001 |
| CLDN18.2-pos vs. -neg | 1.449 | 0.846−2.483 | 0.177 |
| PD-L1 CPS≥1 vs. <1 | 0.925 | 0.519−1.650 | 0.792 |
Table S6. ORR/DCR with ICIs according to CLDN18.2 expression.
| Positivity | n (%) |
| ORR, objective response rate; DCR, disease control rate; ICI, immune checkpoint inhibitor; CLDN18.2, claudin 18 isoform 2. | |
| ORR (total) | |
| CLDN18.2 ≥2+, 40% | 16 (30.7) |
| CLDN18.2 <2+, 40% | 14 (45.2) |
| DCR (total) | |
| CLDN18.2 ≥2+, 40% | 38 (73.0) |
| CLDN18.2 <2+, 40% | 23 (74.2) |
| ORR (1st line) | |
| CLDN18.2 ≥2+, 40% | 14 (42.4) |
| CLDN18.2 <2+, 40% | 8 (57.1) |
| DCR (1st line) | |
| CLDN18.2 ≥2+, 40% | 28 (84.8) |
| CLDN18.2 <2+, 40% | 13 (92.9) |
| ORR (≥2nd line) | |
| CLDN18.2 ≥2+, 40% | 2 (10.5) |
| CLDN18.2 <2+, 40% | 6 (35.3) |
| DCR (≥2nd line) | |
| CLDN18.2 ≥2+, 40% | 10 (52.6) |
| CLDN18.2 <2+, 40% | 10 (58.8) |
| ORR (total) | |
| CLDN18.2 ≥2+, 70% | 15 (35.7) |
| CLDN18.2 <2+, 70% | 15 (36.6) |
| DCR (total) | |
| CLDN18.2 ≥2+, 70% | 31 (73.8) |
| CLDN18.2 <2+, 70% | 30 (73.2) |
| ORR (1st line) | |
| CLDN18.2 ≥2+, 70% | 13 (52.0) |
| CLDN18.2 <2+, 70% | 9 (42.9) |
| DCR (1st line) | |
| CLDN18.2 ≥2+, 70% | 22 (88.0) |
| CLDN18.2 <2+, 70% | 19 (90.5) |
| ORR (≥2nd line) | |
| CLDN18.2 ≥2+, 70% | 2 (12.5) |
| CLDN18.2 <2+, 70% | 6 (30.0) |
| DCR (≥2nd line) | |
| CLDN18.2 ≥2+, 70% | 9 (56.3) |
| CLDN18.2 <2+, 70% | 11 (55.0) |
Exploration of a first-line therapy regimen for CLDN18.2-positive GC
The association between CLDN18.2 expression and the standard first-line therapeutic response was analyzed in 227 HER2-negative, mismatch repair-proficient (pMMR) patients. Intriguingly, patients with positive CLDN18.2 expression had superior PFS (cut-off: 2+, 40%: 6.7 vs. 5.4 months, P=0.099; cut-off: 2+, 70%: 6.9 vs. 5.4 months, P=0.021) (Figure 3). There was no significant difference in PFS between chemotherapy plus immunotherapy and chemotherapy alone in CLDN18.2-positive patients (cut-off: 2+, 40%: 6.5 vs. 7.6 months, P=0.540; cut-off: 2+, 70%: 6.9 vs. 7.4 months, P=0.906) (Figure 4). Therefore, for CLDN18.2-positive patients, the addition of immunotherapy may not exert synergistic anti-tumor effect and anti-CLDN18.2 therapy was necessary for those patients.
Figure 3.
Kaplan-Meier plots of PFS in patients who received standard first-line chemotherapy based on CLDN18.2 status. (A) CLDN18.2 expression cut-off value was 2+, 40% (P=0.099); (B) CLDN18.2 expression cut-off value was 2+, 70% (P=0.021). PFS, progression-free survival; CLDN18.2, claudin 18 isoform 2; pos, positive; neg, negative; 95% CI, 95% confidence interval.
Figure 4.
Kaplan-Meier plots of PFS in CLDN18.2-positive patients who received standard first-line chemotherapy or chemotherapy plus ICI therapy. (A) CLDN18.2 expression cut-off value was 2+, 40% (P=0.540); (B) CLDN18.2 expression cut-off value was 2+, 70% (P=0.906). PFS, progression-free survival; CLDN18.2, claudin 18 isoform 2; ICI, immune checkpoint inhibitor; pos, positive; neg, negative; 95% CI, 95% confidence interval.
Discussion
The clinicopathological characteristics and prognosis of CLDN18.2-positive GC have been previously explored. Nevertheless, due to inconsistencies in study populations, antibodies used, and the case definition of CLDN18.2-positive GC, the therapeutic potential of targeting CLDN18.2 in GC has been unclear. In this study, we retrospectively reviewed 536 Chinese patients with unresectable advanced GC using two cut-off values [according to the FAST study, (cut-off value 1): 2+, 40%; (cut-off value 2): 2+, 70%] defining CLDN18.2 positivity. The clinicopathological characteristics, prognosis, and clinical outcomes of standard chemotherapy and immunotherapy for CLDN18.2-positive GC were analyzed.
According to FAST study, Zolbetuximab plus EOX significantly prolonged the survival [median PFS (mPFS): 7.5 vs. 5.3 months, P<0.005; mOS: 13.0 vs. 8.3 months, P<0.001] in CLDN18.2-positive GC patients. The subsequent positive result of SPOTLIGHT [zolbetuximab plus mFOLFOX6 vs. mFOLFOX6 (mPFS): 10.61 vs. 8.67 months, P<0.001] (6) promotes zolbetuximab plus chemotherapy to become first-line treatment of CLDN18.2-positive GC. Besides, the phase III GLOW trial showed that the addition of zolbetuximab with CAPOX (capecitabine and oxaliplatin) improved PFS for patients with CLDN18.2-positive, HER2-negative, locally advanced unresectable or metastatic G/GEJ adenocarcinoma (HR=0.687; 95% CI, 0.544−0.866; P<0.001) (12). The success of the FAST, GLOW and SPOTLIGHT studies made CLDN18.2 the most likely molecular target in patients with HER2-negative GC. Antibodies, small-molecule inhibitors, antibody-drug conjugates, bispecific antibodies (CLDN18.2-4-1BB; CLDN18.2-CD20 etc.), and cell-based therapies targeting CLDN18.2 (CAR-CLDN18.2 T cell therapy) are on the rise. Notably, the CT041-CG4006 trial showed potent antitumor activity of CLDN18.2-specific CAR cells against gastrointestinal tumors (7). Patients with GC who received at least two prior lines of therapy were enrolled in the study, which reported an ORR of 61.1% and a mPFS of 5.6 (95% CI, 2.6−9.2) months. Additionally, with the advent of the immunotherapy era, increasing amount of clinical trial explored the efficacy of combination of ICIs and anti-CLDN18.2. Hence, it was warranted to explore the clinical and prognosis features of CLDN18.2 expression.
CLDN18.2 positivity varied in diverse studies owing to different definitions of positive expression, various antibodies used, and different patient populations. In this study, CLDN18.2 positivity was 57.6% (cut-off value: 2+, 40%) and 48.9% (cut-off value: 2+, 70%), which was higher than previously reported values. Moderate-to-strong expression of CLDN18.2 in at least 40% of tumor cells was observed in 48.0% of patients enrolled in the FAST study (CLAUDETECT™ 18.2 kit) (11). With the same standard of positive staining in the FAST study, 52.0% of Japanese patients were identified as CLDN18.2 positive (CLAUDETECT™ 18.2 kit) (13). In addition, CLDN18.2 positivity was observed in 42.2% Caucasians (Abcam, cut-off: 1+) (14), 29.4% Koreans (Abcam, cut-off: 2+, 50%) (15) and 24.0% Japanese (Roche Ventana, cut-off: 2+, 75%) (16). CLDN18.2 was highly expressed in signet-ring cell carcinoma (SRCC) (64.8%; cut-off: 2+; Abcam) (17) and repressed in alpha-fetoprotein-producing GC (AFPGC) (21.6%, cut-off: 2+, 40%; Ventana) (18). Currently, the definition of CLDN18.2 positivity for appropriate identification of patients remains unclear. Moderate-to-strong expression of CLDN18.2 in at least 40% of tumor cells was used as a standard based on the FAST study. Accordingly, patients with CLDN18.2-positive GC accounted for 50%−60% of the study population, which was much higher than the frequency of HER2-positive patients; this suggests larger potential beneficiaries of CLDN18.2-targeted therapies for GC. Consistent with the FAST study, the clinical trial of CT041 applied moderate-to-strong expression in at least 40% of tumor cells as an inclusion criterion, whereas the GLOW and SPOTLIGHT studies recruited patients with moderate-to-strong expression of CLDN18.2 in at least 75% of tumor cells, considering the remarkable survival improvement in these patients in the FAST study. Therefore, for different drugs, the definition of CLDN18.2 positivity needs to be determined based on the antibodies used for detection, therapeutic response, and prognosis, rather than a unified standard.
The clinicopathological characteristics of CLDN18.2-positive GC are still unclear (10,19), especially in Chinese patients with advanced GC. In addition to younger age and female sex, CLDN18.2 expression was also significantly associated with diffuse GC and poor survival in our research. Compared to CLDN18.2-negative GC, positive CLDN18.2 expression was associated with non-EGJ, diffuse-type, and poor differentiation. Furthermore, the frequencies of GIN2A mutations and KRAS amplification tended to be higher in CLDN18.2-positive GC (16,17). Hence, these outcomes may contribute to the inferior OS of CLDN18.2-positive GC.
CLDN18.2 expression is associated with varying tumor immune microenvironment characterized by altered amounts of neutrophils and CD8+PD-1−, CD8+LAG-3−, and CD8+TIM-3− T cells (20,21). The level of CD68, a marker for tumor-associated macrophages, is also increased in CLDN18.2-positive tumors (16), and such cells can participate in angiogenesis and immune microenvironment regulation (22,23). LAG-3 and TIM-3 are immune checkpoints that are co-expressed and co-regulated with PD-1 on CD8+ T cells (24), and targeting TIM-3 and PD-1 can restore antitumor immunity (25-27). CLDN18.2-positive GC displays higher proportions of CD8+ T cells with negative expression of PD-1, LAG-3, and TIM-3, which was associated with a poor prognosis of GC (28). The present study innovates by showing that CLDN18.2-positive GC has a poor prognosis and that it does not respond well to immunotherapy. While the recent evidence shows that CLDN18.2-negative GC with CD4+ or CD8+ infiltration has relatively good prognosis (21). Even though PD-L1-negative/positive expression does not reliably predict the efficacy of immunotherapy (29), the fact that the immune checkpoints are weakly expressed in CLDN18.2-positive tumors could participate in the response to immunotherapy, as previously hypothesized. Taken together, CLDN18.2 plays an important role in the altered tumor immune microenvironment, which probably further influences the response to immunotherapy as well as the prognosis of patients with GC. Nevertheless, studies are still necessary to investigate the mechanisms involved.
Our results show that CLDN18.2 and HER2 displayed exclusive expression, demonstrating the importance of CLDN18.2 as a target in HER2-negative GC patients. However, for patients co-expressing HER2 and CLDN18.2, it is necessary to explore the appropriate therapeutic regimens. In contrast to previous studies (14), no difference in the proportion of EBV-positive GC was found between the CLDN18.2-positive and -negative groups. However, this may be because of the small sample size. Furthermore, we observed a higher incidence of uterine adnexa metastasis and a lower occurrence of liver metastasis in CLDN18.2-positive GC, suggesting that the microenvironment of different organs dominantly influences CLDN18.2-positive cells. Consistent with a previous study, we also observed a negative correlation between CLDN18 expression and liver metastasis (10). Therefore, for patients with liver metastases as the main recurrent lesion, the evaluation of CLDN18.2 expression is essential before deciding on anti-CLDN18.2 therapy.
Currently, immunotherapy plus chemotherapy is a first-line standard treatment for advanced GC. The efficacy of immunotherapy for CLDN18.2-positive GC remains unclear. Few studies have focused on the association between CLDN18.2 expression and immunotherapeutic outcomes. Kubota et al. reported no significant differences according to CLDN18.2 status (16). Nevertheless, our analysis indicated CLDN18.2-positive patients treated with immunotherapy showed inferior clinical outcomes compared to the CLDN18.2-negative group. Intriguingly, although significant differences were observed in PD-L1 expression between the CLDN18.2-positive and negative groups, PD-L1 expression was not associated with immunotherapy outcome. In our previous work, CLDN18.2-positive GC displayed unique immune microenvironment characteristics. The numbers of CD8+PD-1−, CD8+LAG3−, and CD8+TIM-3− T cells surrounding the tumor cells were higher in CLDN18.2-positive GC. The relatively low PD-1/PD-L1 expression levels in T cells led to worse immunotherapeutic survival in patients who received anti-PD-1/anti-PD-L1 treatment (20). Following the success of the Checkmate649 study (30), combination chemotherapy and immunotherapy is now standard first-line treatment. Given the adverse role of CLDN18.2 in anti-PD-1/anti-PD-L1 therapy, the combination therapy failed to outperform chemotherapy in CLDN18.2-positive patients. Our analysis suggests that screening for CLDN18.2 expression before immunotherapy is warranted, and the addition of anti-CLDN18.2 treatment may optimize the efficacy of ICIs in CLDN18.2-positive patients.
Conclusions
CLDN18.2-positive GC harbored specific clinicopathological characteristics and served as prognostic marker for immunotherapy. CLDN18.2 expression correlated with poor OS. A combination of anti-CLDN18.2 and ICIs in treating CLDN18.2-positive GC patients should be explored further. The limitations of this study are that the patients analyzed were from a single institution, a small number of patients treated with immunotherapy were included in the analysis, which may affect the reliability of the conclusions.
Acknowledgements
This study is supported by Beijing Natural Science Foundation (No. Z20J00105) and the National Natural Science Foundation of China (No. 82272627).
References
- 1.National Health Commission of the People’s Republic of China National guidelines for diagnosis and treatment of gastric cancer 2022 in China (English version) Chin J Cancer Res. 2022;34:207–37. doi: 10.21147/j.issn.1000-9604.2022.03.04. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Niimi T, Nagashima K, Ward JM, et al claudin-18, a novel downstream target gene for the T/EBP/NKX2. 1 homeodomain transcription factor, encodes lung- and stomach-specific isoforms through alternative splicing. Mol Cell Biol. 2001;21:7380–90. doi: 10.1128/mcb.21.21.7380-7390.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Türeci O, Koslowski M, Helftenbein G, et al Claudin-18 gene structure, regulation, and expression is evolutionary conserved in mammals. Gene. 2011;481:83–92. doi: 10.1016/j.gene.2011.04.007. [DOI] [PubMed] [Google Scholar]
- 4.Sahin U, Koslowski M, Dhaene K, et al Claudin-18 splice variant 2 is a pan-cancer target suitable for therapeutic antibody development. Clin Cancer Res. 2008;14:7624–34. doi: 10.1158/1078-0432.Ccr-08-1547. [DOI] [PubMed] [Google Scholar]
- 5.Sahin U, Türeci Ö, Manikhas G, et al FAST: a randomised phase II study of zolbetuximab (IMAB362) plus EOX versus EOX alone for first-line treatment of advanced CLDN18. 2-positive gastric and gastro-oesophageal adenocarcinoma. Ann Oncol. 2021;32:609–19. doi: 10.1016/j.annonc.2021.02.005. [DOI] [PubMed] [Google Scholar]
- 6.Shitara K, Lordick F, Bang YJ, 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;401:1655-68.
- 7.Qi C, Gong J, Li J, et al Claudin18. 2-specific CAR T cells in gastrointestinal cancers: phase 1 trial interim results. Nat Med. 2022;28:1189–98. doi: 10.1038/s41591-022-01800-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Coati I, Lotz G, Fanelli GN, et al Claudin-18 expression in oesophagogastric adenocarcinomas: a tissue microarray study of 523 molecularly profiled cases. Br J Cancer. 2019;121:257–63. doi: 10.1038/s41416-019-0508-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Arnold A, Daum S, von Winterfeld M, et al Prognostic impact of Claudin 18. 2 in gastric and esophageal adenocarcinomas. Clin Transl Oncol. 2020;22:2357–63. doi: 10.1007/s12094-020-02380-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pellino A, Brignola S, Riello E, et al Association of CLDN18 protein expression with clinicopathological features and prognosis in advanced gastric and gastroesophageal junction adenocarcinomas. J Pers Med. 2021;11:1095. doi: 10.3390/jpm11111095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Al-Batran SE, Schuler MH, Zvirbule, Z, et al. FAST: An international, multicenter, randomized, phase II trial of epirubicin, oxaliplatin, and capecitabine (EOX) with or without IMAB362, a first-in-class anti-CLDN18.2 antibody, as first-line therapy in patients with advanced CLDN18.2+ gastric and gastroesophageal junction (GEJ) adenocarcinoma. J Clin Oncol 2016;34(18_suppl):LBA4001.
- 12.Shah MA, Shitara K, Ajani JA, et al Zolbetuximab plus CAPOX in CLDN18. 2-positive gastric or gastroesophageal junction adenocarcinoma: the randomized, phase 3 GLOW trial. Nat Med. 2023;29:2133–41. doi: 10.1038/s41591-023-02465-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rohde C, Yamaguchi R, Mukhina S, et al Comparison of Claudin 18. 2 expression in primary tumors and lymph node metastases in Japanese patients with gastric adenocarcinoma. Jpn J Clin Oncol. 2019;49:870–6. doi: 10.1093/jjco/hyz068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dottermusch M, Krüger S, Behrens HM, et al Expression of the potential therapeutic target claudin-18. 2 is frequently decreased in gastric cancer: results from a large Caucasian cohort study. Virchows Arch. 2019;475:563–71. doi: 10.1007/s00428-019-02624-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Baek JH, Park DJ, Kim GY, et al Clinical implications of claudin18. 2 expression in patients with gastric cancer. Anticancer Res. 2019;39:6973–9. doi: 10.21873/anticanres.13919. [DOI] [PubMed] [Google Scholar]
- 16.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;8:100762. doi: 10.1016/j.esmoop.2022.100762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Xu B, Liu F, Liu Q, et al Highly expressed Claudin18. 2 as a potential therapeutic target in advanced gastric signet-ring cell carcinoma (SRCC). J Gastrointest Oncol. 2020;11:1431–9. doi: 10.21037/jgo-20-344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Weng W, Zhang M, Ni S, et al Decreased expression of claudin-18. 2 in alpha-fetoprotein-producing gastric cancer compared to conventional gastric cancer. J Gastrointest Oncol. 2022;13:1035–45. doi: 10.21037/jgo-22-462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ooki A, Yamaguchi K. The dawn of precision medicine in diffuse-type gastric cancer. Ther Adv Med Oncol 2022;14:17588359221083049.
- 20.Jia K, Chen Y, Sun Y, et al Multiplex immunohistochemistry defines the tumor immune microenvironment and immunotherapeutic outcome in CLDN18. 2-positive gastric cancer. BMC Med. 2022;20:223. doi: 10.1186/s12916-022-02421-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wang C, Wang Y, Chen J, et al CLDN18. 2 expression and its impact on prognosis and the immune microenvironment in gastric cancer. BMC Gastroenterol. 2023;23:283. doi: 10.1186/s12876-023-02924-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zhu Q, Wu X, Tang M, et al Observation of tumor-associated macrophages expression in gastric cancer and its clinical pathological relationship. Medicine (Baltimore) 2020;99:e19839. doi: 10.1097/MD.0000000000019839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gambardella V, Castillo J, Tarazona N, et al The role of tumor-associated macrophages in gastric cancer development and their potential as a therapeutic target. Cancer Treat Rev. 2020;86:102015. doi: 10.1016/j.ctrv.2020.102015. [DOI] [PubMed] [Google Scholar]
- 24.Chihara N, Madi A, Kondo T, et al Induction and transcriptional regulation of the co-inhibitory gene module in T cells. Nature. 2018;558:454–9. doi: 10.1038/s41586-018-0206-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sakuishi K, Apetoh L, Sullivan JM, et al Targeting Tim-3 and PD-1 pathways to reverse T cell exhaustion and restore anti-tumor immunity. J Exp Med. 2010;207:2187–94. doi: 10.1084/jem.20100643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Curigliano G, Gelderblom H, Mach N, et al Phase I/Ib clinical trial of sabatolimab, an anti-TIM-3 antibody, alone and in combination with spartalizumab, an anti-PD-1 antibody, in advanced solid tumors. Clin Cancer Res. 2021;27:3620–9. doi: 10.1158/1078-0432.CCR-20-4746. [DOI] [PubMed] [Google Scholar]
- 27.Hellmann MD, Bivi N, Calderon B, et al Safety and immunogenicity of LY3415244, a bispecific antibody against TIM-3 and PD-L1, in patients with advanced solid tumors. Clin Cancer Res. 2021;27:2773–81. doi: 10.1158/1078-0432.CCR-20-3716. [DOI] [PubMed] [Google Scholar]
- 28.Park Y, Seo AN, Koh J, et al Expression of the immune checkpoint receptors PD-1, LAG3, and TIM3 in the immune context of stage II and III gastric cancer by using single and chromogenic multiplex immunohistochemistry. Oncoimmunology. 2021;10:1954761. doi: 10.1080/2162402X.2021.1954761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ribas A, Hu-Lieskovan S What does PD-L1 positive or negative mean. J Exp Med. 2016;213:2835–40. doi: 10.1084/jem.20161462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Janjigian YY, Shitara K, Moehler M, et al First-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trial. Lancet. 2021;398:27–40. doi: 10.1016/s0140-6736(21)00797-2. [DOI] [PMC free article] [PubMed] [Google Scholar]