TROP2 expression in gastric and GEJ adenocarcinomas: associations with clinicopathological features, biomarkers, and treatment outcomes Authors: Y Tong, I Nakayama, N Sakamoto, Y Miyashita, Y Shiba, U Hirotaka, Y Sugawara, D Okemoto, K Yamamoto, K Seguchi, T Ogura, Y Matsubara, T Hashimoto, S Mishima, D Kotani, A Kawazoe, Y Kuboki, H Bando, T Kojima, T Yoshino, K Shitara Published: 2026 ABSTRACT AbstractBackgroundTrophoblast cell surface antigen 2 (TROP2) is frequently expressed in solid tumors, including gastric cancer (GC) and gastroesophageal junction cancer (GEJC), but its distribution and clinical relevance remain unclear.Patients and methodsWe analyzed patients with GC/GEJC who underwent prospective multibiomarker testing. TROP2 expression was assessed by immunohistochemistry using multiple cut-offs and associations with clinicopathological features, biomarkers, and outcomes were evaluated.ResultsAmong 272 patients (169 with metastatic cancer), TROP2 expression was observed in 87%, 79%, and 62% at ≥25%, ≥50%, and ≥75% cut-offs, respectively. Higher cut-offs were associated with higher proportions of male patients, gastroesophageal junction tumors, intestinal-type histology, and liver metastasis. Among patients with TROP2-high tumors (>75% positivity), positivity rates in the full cohort were 17.8% for human epidermal growth factor receptor 2, 50.9% for claudin 18.2, 4.7% for deficient mismatch repair, and 30.2% for a programmed death-ligand 1 combined positive score >10; in the metastatic cohort, the corresponding rates were 18.6%, 48.0%, 5.9%, and 35.3%, respectively. Overall, 38.8% of TROP2-high patients in the full cohort and 31.3% in the metastatic cohort were negative for all four biomarkers. TROP2 expression was not associated with treatment outcomes for first-line platinum-based doublet chemotherapy or second-line taxane-based chemotherapy.ConclusionsTROP2 is widely expressed in GC/GEJC irrespective of stage. Higher expression thresholds delineate distinct clinicopathological features without prognostic impact. Its broad prevalence extends to patients lacking actionable biomarkers, supporting its potential as a therapeutic target.Key words: biomarkers, clinicopathologic feature, gastric cancer, TROP2 Introduction Trophoblast cell surface antigen 2 (TROP2), also known as tumor-associated calcium signal transducer 2, has emerged as a promising therapeutic target in oncology.1 As a membrane glycoprotein, TROP2 regulates calcium signaling, cell-cycle progression, and cellular proliferation.2 Aberrant overexpression has been reported in multiple malignancies, including lung cancer, breast cancer, and gastric cancer (GC),3, 4, 5, 6 and is associated with aggressive tumor biology and poor prognosis.3, 4, 5, 6 Owing to its biological relevance, therapeutic strategies targeting TROP2—particularly antibody-drug conjugates (ADCs)—have demonstrated substantial antitumor activity in clinical trials.7, 8, 9 Sacituzumab govitecan, an ADC linking a TROP2 antibody to the topoisomerase I inhibitor SN-38, has been approved for hormone receptor–positive, human epidermal growth factor receptor 2 (HER2)–negative metastatic breast cancer, where it substantially improved progression-free survival (PFS) and overall survival (OS).10, 11, 12 This breakthrough highlights the clinical utility of TROP2-directed therapy and heralds a new era in the treatment of solid tumors. In GC and gastroesophageal junction cancer (GEJC), the integration of targeted therapy and immunotherapy into first-line regimens—such as HER2-directed and programmed cell death protein 1–directed treatments—has improved patient outcomes.13, 14, 15, 16, 17 More recently, zolbetuximab, a claudin (CLDN) 18.2–targeting antibody combined with chemotherapy, demonstrated important clinical benefit in patients with HER2-negative and CLDN18.2-positive disease, further expanding therapeutic options.6,18,19 Despite these advances, the prognosis for patients with metastatic GC/GEJC (mGC/GEJC) remains poor, underscoring the need for novel biomarkers to refine patient stratification and treatment strategies. Encouraging results from a phase II trial of TROP2-targeting ADCs have led to the initiation of an international randomized phase III study (NCT06356311) in previously treated patients with mGC/GEJC.20 Nevertheless, the expression profile of TROP2 in GC/GEJC, its associations with clinicopathological and molecular features, and its correlation with established actionable biomarkers and clinical outcomes remain poorly defined. To address this gap, we explored its association with patient survival outcomes and clinicopathological features. Materials and methods Study design and patient selection This single-institution retrospective cohort study was conducted at the National Cancer Center East (NCCE), Japan. Eligible patients were those who (i) had histologically or cytologically confirmed gastric or gastroesophageal junction (GEJ) adenocarcinoma, (ii) had received no prior systemic anticancer treatment, (iii) had received treatment between July 2024 and September 2025 at NCCE, and (iv) had an archival tissue specimen available for biomarker analysis. Patients with nonmetastatic disease received heterogeneous treatments according to disease stage and treatment period, including perioperative Fluorouracil, Leucovorin, Oxaliplatin, and Docetaxel; upfront surgery followed by adjuvant chemotherapy with S-1, Capecitabine and Oxaliplatin/S-1 and Oxaliplatin, or docetaxel (Taxotere, Nippon Kayaku, Tokyo, Japan) plus S-1; and surgery alone. All patients provided written informed consent for treatment and prospective biomarker study (UMIN000019129). The study protocol was approved by the ethics committee of NCCE and conducted in accordance with the Declaration of Helsinki guidelines for biomedical research. Assessments Clinicopathological data on all eligible participants were collected from electronic medical records. TROP2 expression at primary or metastatic sites was evaluated by immunohistochemistry (IHC) using an anti-TROP2 antibody (clone EPR20043; Abcam, Cambridge, UK). Appropriate cut-off levels for TROP2 expression by IHC were explored using thresholds of ≥25%, ≥50%, and ≥75% of tumor cells exhibiting moderate (2+) to strong (3+) staining intensity. HER2 positivity was defined as IHC 3+ or fluorescence in situ hybridization-confirmed IHC 2+. Mismatch repair (MMR) status was determined by IHC for MutL homolog 1, MutS homolog 2, postmeiotic segregation increased 2, and MutS homolog 6. Tumors lacking any MMR protein were classified as deficient mismatch repair (dMMR); others were considered proficient mismatch repair (pMMR). Programmed death-ligand 1 (PD-L1) expression was assessed using an anti-PD-(L)1 antibody and was quantified by a combined positive score (CPS), calculated as the percentage of PD-L1–positive cells (tumor cells, lymphocytes, and macrophages) relative to viable tumor cells. CLDN18.2 positivity was defined as moderate-to-strong membrane staining in ≥75% of tumor cells, based on pivotal clinical trials.18,19 Treatment outcomes were evaluated in patients with mGC/GEJC. Tumor response was assessed according to RECIST version 1.1. The objective response rate (ORR) was defined as the proportion of patients achieving a complete response (CR) or partial response (PR); the disease control rate (DCR) included patients with CR, PR, or sustained stable disease. PFS was measured from treatment initiation to the first documentation of disease progression or death, whichever occurred earlier. OS was defined as the time from treatment initiation to death from any cause or the date of last follow-up. Patients without a death event were censored at last contact. Outcomes and statistical analysis Analyses were conducted for two groups: (i) all patients meeting eligibility criteria with evaluable TROP2 protein expression (‘full cohort’) and (ii) all patients from the full cohort with mGC/GEJC (‘metastatic cohort’). Variables are presented as median (range) for continuous variables and numbers (percentage) for categorical variables. Categorical characteristics were compared using two-sided Fisher’s exact tests. Time-to-event analyses for PFS and OS were carried out using Kaplan–Meier curves. Median follow-up time was estimated using the reverse Kaplan–Meier method. All statistical analyses, conducted in GraphPad Prism 10.1 (GraphPad Software, Inc., San Diego, CA), employed two-sided tests, with significance defined as P < 0.05. Results Patient characteristics A total of 272 consecutive patients with GC/GEJC were eligible for this study and comprised the full cohort, of whom 169 with mGC/GEJC represented the metastatic cohort. In the mGC/GEJC cohort, 137 patients received platinum-based doublet chemotherapy (with or without immunotherapy/targeted therapy) as the first-line treatment, and 65 patients subsequently received paclitaxel-based therapy as the second-line treatment. The baseline characteristics of the full cohort and the metastatic cohort are shown in Supplementary Table S1 (available at https://doi.org/10.1016/j.esmogo.2026.100413). Clinicopathological trends of TROP2 expression The distribution of clinicopathological characteristics according to different TROP2 expression levels (≥25%, ≥50%, and ≥75%) is shown in Figure 1A. Among 272 patients in the full cohort, 62% showed TROP2 expression in ≥75% of tumor cells, whereas only 13% showed TROP2 expression in <25% of tumor cells. In the metastatic cohort (n = 169), the distribution was similar, with 60% showing TROP2 expression in ≥75% of tumor cells and 13% showing TROP2 expression in <25% of tumor cells. In both the full and metastatic cohorts, raising the TROP2 cut-off level from 25% to 75% revealed consistent trends in clinicopathological distribution. The proportions of male patients, GEJ tumors, intestinal-type histology, and liver metastases gradually increased with higher cut-offs, whereas those of signet ring cell carcinoma and peritoneal dissemination decreased (Figure 1B). Overall, membranous TROP2 expression was broadly observed in both the full and metastatic cohorts, with no major differences across subgroups at different cut-offs. However, increasing the cut-off threshold revealed consistent clinicopathological trends, suggesting that a higher threshold better characterizes TROP2-expressing GC/GEJC. Therefore, subsequent analyses were conducted using the most stringent cut-off (≥75%) to define TROP2 positivity. Distribution of trophoblast cell surface antigen 2 (TROP2) expression in gastric cancer/gastroesophageal junction (GEJ) cancer. (A) Vertical bar plots showing the number of patients stratified by TROP2 expression cut-offs (<25%, 25% to <50%, 50% to <75%, and ≥75%) in the full cohort (left, N = 272) and the metastatic cohort (right, n = 169). Case counts are indicated above each bar. (B) Horizontal bar plots showing the proportion (%) of patients with each clinicopathological feature according to TROP2 expression cut-offs (≥25%, ≥50%, and ≥75%) in the full cohort (left, N = 272) and the metastatic cohort (right, n = 169). Bars are color-coded by cut-off levels in ascending order of expression, with percentages labeled on the right side of each bar. LN, lymph node; Mets, metastases; Sig, signet ring cell carcinoma. Association of TROP2 expression with clinicopathological features We compared clinicopathological characteristics between the TROP2-high and TROP2-low groups in both the full (N = 272) and metastatic cohorts (n = 169). No statistically significant associations were observed between TROP2 expression status and age, sex, tumor location (gastric versus GEJ), metastatic sites, or the number of metastatic sites (all P > 0.05) (Tables 1 and 2). In contrast, histological subtype demonstrated statistically significant associations. In the full cohort, TROP2-high tumors were more frequently intestinal type (42.6% versus 26.2%, P = 0.007) and non-signet ring cell carcinoma (84.6% versus 70.9%, P = 0.008). In the metastatic cohort, TROP2-high mGC/GEJC was significantly more likely to be non-signet ring cell carcinoma (84.3% versus 68.7%, P = 0.022). Table 1Clinicopathological characteristics according to TROP2 expression in the full cohort (N = 272) (cut-off level: 75%)CharacteristicsTROP2-low (n = 103)TROP2-high (n = 169)P valueAge (years), median (range)66 (23-91)70 (26-91)0.143Gender, n (%)0.087 Male61 (59.2)118 (69.8) Female42 (40.8)51 (30.2)Tumor location, n (%)0.060 Gastric89 (86.4)130 (76.9) GEJ14 (13.6)39 (23.1)Lauren, n (%)0.012 Intestinal26 (25.2)70 (41.4) Diffuse73 (70.9)96 (56.8)Histology, n (%)0.008 Sig30 (29.1)26 (15.4) Non-Sig69 (67.0)140 (82.8)Metastatic site, n (%) Peritoneum24 (23.3)27 (16.0)0.1507 Liver15 (14.6)28 (16.6)0.7335 LN82 (79.6)133 (78.7)0.8794 Lung4 (3.9)7 (4.1)>0.999 Bone3 (2.9)3 (1.8)0.6762No. of Mets ≥2, n (%)42 (40.8)61 (36.1)0.443GEJ, gastroesophageal junction; LN, lymph node; Mets, metastases; Sig, signet ring cell carcinoma; TROP2, trophoblast cell surface antigen 2. Clinicopathological characteristics according to TROP2 expression in the full cohort (N = 272) (cut-off level: 75%) GEJ, gastroesophageal junction; LN, lymph node; Mets, metastases; Sig, signet ring cell carcinoma; TROP2, trophoblast cell surface antigen 2. Table 2Clinicopathological characteristics according to TROP2 expression in the metastatic cohort (n = 169) (cut-off level: 75%)CharacteristicsTROP2-low (n = 67)TROP2-high (n = 102)P valueAge (years), median (range)64 (23-91)69 (26-91)0.066Gender, n (%)0.068 Male38 (56.7)73 (71.6) Female29 (43.3)29 (28.4)Tumor location, n (%)0.6655 Gastric58 (86.6)85 (83.3) GEJ9 (13.4)17 (16.7)Lauren, n (%)0.6062 Intestinal18 (26.9)33 (32.4) Diffuse47 (70.1)69 (67.6)Histology, n (%)0.0137 Sig21 (31.3)16 (15.7) Non-Sig44 (65.7)86 (84.3)Metastatic site, n (%) Peritoneum24 (35.8)27 (26.5)0.2314 Liver15 (22.4)28 (27.5)0.4775 LN63 (94.0)96 (94.1)>0.999 Lung4 (6.0)7 (6.9)>0.999 Bone3 (4.5)3 (2.9)0.6824No. of Mets ≥2, n (%)42 (62.7)61 (59.8)0.7489GEJ, gastroesophageal junction; LN, lymph node; Mets, metastases; Sig, signet ring cell carcinoma; TROP2, trophoblast cell surface antigen 2. Clinicopathological characteristics according to TROP2 expression in the metastatic cohort (n = 169) (cut-off level: 75%) GEJ, gastroesophageal junction; LN, lymph node; Mets, metastases; Sig, signet ring cell carcinoma; TROP2, trophoblast cell surface antigen 2. Association of TROP2 expression with HER2, MMR, PD-L1, and CLDN18.2 Among the patients with TROP2-high tumors, the proportions of HER2 positivity, CLDN18.2 positivity, dMMR, and PD-L1 CPS ≥10 were 17.8%, 50.9%, 4.7%, and 30.2% in the full cohort and 18.6%, 48.0%, 5.9%, and 35.3% in the metastatic cohort, respectively (Figure 2A). The proportions of dMMR, PD-L1 CPS ≥10, and CLDN18.2 positivity did not differ significantly between the TROP2-high and TROP2-low groups in either the full or metastatic cohort; HER2 positivity was significantly higher in the TROP2-high group (17.8% in the full cohort and 18.6% in the metastatic cohort) than in the TROP2-low group (5.8% in the full cohort and 6.0% in the metastatic cohort) (P = 0.005 and P = 0.022, respectively) (Figure 2A). Association of trophoblast cell surface antigen 2 (TROP2) with other actionable biomarkers. (A) Bar charts comparing the proportions of human epidermal growth factor receptor 2 (HER2)–positive, claudin (CLDN) 18.2-positive, deficient mismatch repair (dMMR), and programmed death-ligand 1 (PD-L1) combined positive score (CPS) ≥10 tumors between the TROP2-high (≥75%) and TROP2-low groups in both the full and metastatic cohorts. The values shown represent the percentage of patients within each subgroup. (B) Venn diagrams showing the distribution and overlap of HER2-positive, CLDN18.2-positive, and TROP2-high tumors in the proficient mismatch repair population of the full cohort. TROP2 positivity was defined using a cut-off level of ≥75%. The size of each circle reflects the relative number of cases, and the numbers indicate patient counts in each subset. Given that dMMR represents a distinct immunogenic subtype in GC/GEJC and other solid tumors, we examined the co-expression of TROP2 with HER2 and CLDN18.2 in pMMR cases. In the full cohort, co-occurrence of HER2 positivity and TROP2-high expression was observed in 15 patients (5.9%), whereas 70 patients (27.3%) showed co-occurrence of CLDN18.2 positivity and TROP2-high expression. Concurrent HER2 positivity, CLDN18.2 positivity, and high TROP2 expression were observed in 15 patients (5.9%). In the metastatic cohort, the corresponding numbers were 11 (7.0%), 40 (25.3%), and 8 (5.1%), respectively. Conversely, in the full cohort (n = 256), 61 patients (23.8%) had high TROP2 expression (≥75%) and were negative for both HER2 and CLDN18.2. In the metastatic cohort (n = 158), the corresponding number was 37 patients (23.4%). Overall, 40 patients (38.8%) with TROP2-high tumors in the full cohort and 21 patients (31.3%) in the metastatic cohort were negative for all four biomarkers (Figure 2B). Treatment outcomes of chemotherapy according to TROP2 status In the metastatic cohort of 169 patients, 137 (81.1%) started first-line platinum-based doublet chemotherapy at NCCE (Supplementary Table S2, available at https://doi.org/10.1016/j.esmogo.2026.100413). The median follow-up time was 9.87 months [95% confidence interval (CI) 8.65-10.97]; median PFS did not differ significantly between the TROP2-low (<75%) and TROP2-high (≥75%) groups [6.90 versus 6.73 months, hazard ratio (HR) 1.04, 95% CI 0.68-1.43, P = 0.846] (Figure 3A). Similarly, OS was comparable between the two groups (median OS not reached, HR 0.81, 95% CI 0.39-1.69, P = 0.565) (Figure 3B). Treatment response rates were comparable irrespective of TROP2 expression, with ORRs of 58.8% and 54.2% (P = 0.713) and DCRs of 88.2% and 83.3% (P = 0.606) in the TROP2-low and TROP2-high groups, respectively (Tables 3 and 4). Following disease progression on first-line therapy, 65 of these patients subsequently received second-line treatment with paclitaxel or nab-paclitaxel with or without ramucirumab. The median PFS was comparable between the TROP2-low (<75%) and TROP2-high (≥75%) groups (4.82 versus 3.55 months, HR 0.89, 95% CI 0.48-1.64, P = 0.703) (Figure 3C). Similarly, OS did not differ significantly between the two groups (median OS not reached, HR 0.45, 95% CI 0.16-1.24, P = 0.135) (Figure 3D). There were no significant differences in ORR (19.2% versus 13.3%, P = 0.719) or DCR (57.5% versus 60.0%, P = 1.000) between the two groups (Tables 3 and 4). These findings suggest that TROP2 expression does not appear to influence the efficacy of first-line platinum-based doublet chemotherapy or second-line taxane-based chemotherapy in patients with mGC/GEJC. Progression-free survival (PFS) and overall survival (OS) according to trophoblast cell surface antigen 2 (TROP2) expression status. (A) Kaplan–Meier curves of PFS for patients receiving first-line platinum-based doublet chemotherapy, stratified by TROP2-low (<75%) and TROP2-high (≥75%) subgroups. Median PFS, hazard ratio (HR) with 95% confidence interval (CI), and P value are shown. (B) Kaplan–Meier curves of OS for the full cohort, stratified by TROP2 expression using the same cut-off. Median OS, HR with 95% CI, and P value are shown. (C) Kaplan–Meier curves of PFS for patients receiving second-line treatment with paclitaxel or paclitaxel plus ramucirumab, stratified by TROP2 expression using the same cut-off. Median PFS, HR with 95% CI, and P value are shown. (D) Kaplan–Meier curves of OS for the metastatic cohort, stratified by TROP2 expression using the same cut-off. Median OS, HR with 95% CI, and P value are shown. NR, not reached. Table 3ORR and DCR according to TROP2 expression in patients receiving first-line platinum-based doublet chemotherapy (n = 137)ResponseTROP2-low (n = 57)TROP2-high (n = 80)P valueMeasurable lesion, n5477CR, n (%)0 (0.0)2 (2.6)PR, n (%)30 (55.6)37 (48.1)SD, n (%)15 (27.8)21 (27.3)PD, n (%)3 (5.6)3 (3.9)NE, nORR (%)DCR (%)655.683.31450.777.90.59870.5094Analysis is limited to patients with measurable lesions.CR, complete response; DCR, disease control rate (CR + PR + SD); NE, not evaluated; ORR, objective response rate (CR + PR); PD, progressive disease; PR, partial response; SD, stable disease; TROP2, trophoblast cell surface antigen 2. ORR and DCR according to TROP2 expression in patients receiving first-line platinum-based doublet chemotherapy (n = 137) Analysis is limited to patients with measurable lesions. CR, complete response; DCR, disease control rate (CR + PR + SD); NE, not evaluated; ORR, objective response rate (CR + PR); PD, progressive disease; PR, partial response; SD, stable disease; TROP2, trophoblast cell surface antigen 2. Table 4ORR and DCR according to TROP2 expression in patients receiving second-line paclitaxel-based therapy (n = 65)ResponseTROP2-low (n = 30)TROP2-high (n = 35)P valueMeasurable lesion, n3034CR, n (%)0 (0.0)0 (0.0)PR, n (%)5 (16.7)4 (11.8)SD, n (%)10 (33.3)14 (41.2)PD, n (%)7 (23.3)9 (26.5)N/E, nORR (%)DCR (%)816.750.0711.852.90.7231>0.999Analysis is limited to patients with measurable lesions.CR, complete response; DCR, disease control rate (CR + PR + SD); NE, not evaluated; ORR, objective response rate (CR + PR); PD, progressive disease; PR, partial response; SD, stable disease; TROP2, trophoblast cell surface antigen 2. ORR and DCR according to TROP2 expression in patients receiving second-line paclitaxel-based therapy (n = 65) Analysis is limited to patients with measurable lesions. CR, complete response; DCR, disease control rate (CR + PR + SD); NE, not evaluated; ORR, objective response rate (CR + PR); PD, progressive disease; PR, partial response; SD, stable disease; TROP2, trophoblast cell surface antigen 2. Discussion To our knowledge, this is the first study to evaluate TROP2 expression in GC/GEJC across multiple thresholds (≥25%, ≥50%, and ≥75%) and to investigate its clinical relevance in relation to other actionable biomarkers as well as treatment outcomes. We demonstrated that higher TROP2 expression is associated with intestinal or non–signet ring cell histology and it occasionally co-occurs with HER2 positivity. Importantly, high TROP2 expression was identified irrespective of dMMR, PD-L1, or CLDN18.2 status. Notably, 31.3% of patients with mGC/GEJC had high TROP2 expression (≥75%) but lacked a current actionable biomarker, suggesting the potential for a novel therapeutic option. Higher TROP2 expression was not associated with outcomes of first-line platinum-based doublet chemotherapy or second-line taxane-based chemotherapy. TROP2 is a transmembrane glycoprotein frequently overexpressed in epithelial malignancies, with limited expression in normal tissues, making it an attractive therapeutic target.21 Several studies have previously investigated TROP2 expression in GC. Analyses of 412 Korean, 104 Austrian, and 704 Japanese patients reported high or positive TROP2 expression rates of 51%, 56%, and 46.9%, respectively.2, 3, 4 In the present study, we evaluated TROP2 expression on the basis of the proportion of tumor cells showing moderate (2+) or strong (3+) membranous staining and explored thresholds of ≥25%, ≥50%, and ≥75%. We adopted this approach because weak (1+) staining is often difficult to distinguish reliably from absent staining. Furthermore, assessing the proportion of tumor cells with moderate-to-strong staining is consistent with the evaluation framework used for CLDN18.2 in the phase III SPOTLIGHT (NCT03504397) and GLOW (NCT03653507) phase 3 trials, which investigated zolbetuximab plus chemotherapy in patients with CLDN18.2-positive gastric or gastroesophageal junction adenocarcinoma, offering a more straightforward and clinically applicable metric. We explored multiple thresholds and selected the ≥75% cut-off on the basis of the observed distribution of TROP2 expression and its ability to define a distinct TROP2-high population in our cohort. In our study, the prevalence of high TROP2 expression was higher, observed at cut-offs of ≥25%, ≥50%, and ≥75% in 87.1%, 79.4%, and 62.1% of patients in the full cohort and in 86.4%, 78.7%, and 60.4% of patients in the metastatic cohort, respectively (Figure 1A). These discrepancies may be attributable to ethnic differences, the use of different antibodies with varying affinities, or inconsistent cut-off definitions across studies. Despite variability in the reported prevalence across studies, TROP2 expression is observed in a substantial proportion of patients with GC/GEJC, supporting its potential as a therapeutic target. Our analysis demonstrated that increased membranous TROP2 expression was associated with distinct clinicopathological changes. The full cohort was analyzed to provide a comprehensive baseline characterization of gastric/GEJ adenocarcinomas, whereas the metastatic cohort was evaluated separately to address the clinically relevant question of whether TROP2 expression differs in advanced-stage disease. Specifically, higher TROP2 expression correlated with an increased proportion of male patients, GEJ tumors, intestinal-type histology, and liver metastasis, alongside a decreased prevalence of peritoneal metastasis. The TROP2-high group was defined using a stringent definition of moderate-to-strong staining intensity on ≥75% of tumor cells, and TROP2-high mGC/GEJC was more frequently associated with intestinal-type histology. Given that TROP2 signaling has been implicated in the enhancement of tight junction formation—an architectural feature characteristic of intestinal-type histology—these findings suggest that increased TROP2 expression may contribute to the stabilization of epithelial cell–cell adhesion and reduced cellular dispersion. These observations align with and extend the findings from recent studies. For instance, Kim et al.3 similarly reported associations among TROP2 expression, intestinal subtype, and HER2 positivity; however, their biomarker analysis was limited to PD-L1 (22C3), leaving the interplay with other emerging targets unexplored. Furthermore, although Awatsu et al.22 recently evaluated TROP2 and CLDN18.2 expression, their investigation focused primarily on the prognostic relevance of these markers in a specific cohort receiving third-line nivolumab monotherapy. In contrast, our study provides a broader characterization of TROP2 expression patterns in a general metastatic population, independent of specific immunotherapy selection. Collectively, increased TROP2 expression appears to reflect a biologically relevant continuum that may delineate a distinct molecular or phenotypic subset. Given the high prevalence of TROP2 expression, a substantial overlap with CLDN18.2 and HER2 was observed in pMMR GC/GEJC. Notably, almost all HER2-positive GCs/GEJCs were included in TROP2-high tumors (≥75%) in both the full and metastatic cohorts. This finding is in line with common clinicopathological features such as intestinal-type histology, GEJ location, and liver metastasis. These observations provide a biological rationale for dual-targeting strategies incorporating both TROP2 and HER2 in selected patient subsets. Despite this overlap, the high prevalence of TROP2 expression encompasses patients with GC/GEJC lacking established actionable biomarkers. This highlights TROP2 as a potentially actionable biomarker addressing an unmet clinical need in GC/GEJC. Currently, TROP2-directed ADCs have been implemented in clinical practice for patients with breast or lung cancer. These agents demonstrate clinical activity irrespective of TROP2 expression. Therefore, TROP2 expression level has not been established as a predictive biomarker for patient selection.23,24 Similarly, exploratory analyses in GC have not shown a clear difference in efficacy according to the level of TROP2 expression.20 Further investigation in ongoing clinical trials is warranted to establish the clinical utility of TROP2 expression in patients with mGC/GEJC. Furthermore, biomarker analyses from the TROPION-Lung study reported that quantitative assessment incorporating cytoplasmic TROP2 expression, using an artificial intelligence–based model [TROP2 quantitative cytoplasmic staining–normalized membrane ratio), correlated with the efficacy of datopotamab deruxtecan, a TROP2-directed ADC.25, 26, 27 These findings suggest that the subcellular localization of ADC targets may affect clinical efficacy. Patients with GC/GEJC displaying different subcellular TROP2 localization patterns may respond differently to TROP2-directed ADC therapies, underscoring an important area for future research. This study has several limitations. It was a single-center retrospective and prospective analysis with a relatively small sample size, which may limit the generalizability of the findings. The median follow-up time for chemotherapy was relatively short, potentially underestimating long-term outcomes. In addition, when assessing the impact of TROP2 expression on the efficacy of first-line therapy, patients were not stratified by treatment regimen (chemotherapy alone, chemotherapy plus targeted therapy, chemotherapy plus immunotherapy, or combination approaches). Furthermore, the cut-off used to define high TROP2 expression was determined on the basis of the distribution within our cohort and is exploratory in nature; its clinical utility requires validation in independent cohorts. These factors should be considered when interpreting the results. Conclusions TROP2 is widely expressed in GC/GEJC and represents a potential therapeutic target irrespective of disease status. Its expression reflects a biologically relevant continuum associated with distinct clinicopathological features but not with prognosis or response to chemotherapy. Despite partial overlap with HER2, TROP2 identifies a subset of patients lacking established actionable biomarkers, supporting its potential for clinical development in patients with mGC/GEJC. Acknowledgements The authors thank all patients and their families for their participation, as well as the clinical and nursing staff for their valuable support throughout the study. Informed consent Written informed consent, or its equivalent, was obtained from every patient before inclusion. Ethical approval All procedures complied with the ethical standards of the institutional and national committees on human experimentation, as well as with the 1964 Declaration of Helsinki and its subsequent amendments. The study protocol was reviewed and approved by the Certified Review Board of the National Cancer Center Hospital East (approval ID: 2017-120; approval date: 26 September 2017). Data availability The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Funding This work was supported by research funding from the Japan-China Sasakawa Medical Fellowship (46th) (no grant number). Disclosure IN reports research funding from Astellas, Ono Pharmaceutical Co., Ltd, Merck Sharp & Dohme (MSD), Daiichi Sankyo, Boehringer Ingelheim, AstraZeneca, and Chugai Pharmaceutical Co., Ltd and honoraria from Astellas, Ono Pharmaceutical Co., Ltd, Bristol Myers Squibb, MSD, Taiho Pharmaceutical Co., Ltd, and Daiichi Sankyo. Y Matsubara reports honoraria from Taiho, Takeda, Eli Lilly Japan, MSD, and Bristol Myers Squibb. SM reports honoraria from Taiho Pharmaceutical Co., Ltd, Chugai Pharmaceutical Co., Ltd, and Eli Lilly Co., Ltd. DK reports 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. AK reports 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. TH reports honoraria from CytoGen, Inc. and Takata Pharmaceutical and consulting and advisory roles at Guardant Health. YK reports advisory roles (personal fees) at Taiho, Takeda, and Boehringer Ingelheim; honoraria from Taiho, Ono, Bayer, and Sanofi; and institutional grants or contracts from Taiho, Takeda, Ono, AbbVie, AstraZeneca, Boehringer Ingelheim, Incyte, Amgen, Chugai, GlaxoSmithKline, Genmab, Astellas, and Daiichi Sankyo. HB reports research funding from Ono Pharmaceutical and honoraria from Ono Pharmaceutical, Taiho Pharmaceutical, and Eli Lilly Japan. TK reports 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. and 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. TY reports honoraria from Chugai, Takeda, Merck, and Ono; consulting fees from Sumitomo Corp. and Indivumed; and research grants from Bristol Myers Squibb, Caris, Chugai, Daiichi Sankyo, Eisai, Exact Sciences, FALCO Biosystems, Medical & Biological Laboratories, Merus N.V., MSD, Miyarisan, Natera, Nippon Boehringer Ingelheim, Ono, Pfizer, Sysmex, Taiho, and Takeda. K Shitara reports receiving personal fees for consulting and advisory roles 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, Arcus Biosciences Inc, AbbVie, CMIC Co., Ltd, Elevation Oncology, Oncolys BioPharma, Sanofi, Suzhou Linghui Network Technology, Gilead Sciences, eChinaHealth, Leap Therapeutics, Phans Therapeutics, Scandion Oncology, and Revolution Medicines; honoraria from Bristol Myers Squibb, Ono Pharmaceutical, Janssen, Eli Lilly, Astellas, and AstraZeneca; and research funding (all to the institution) from Astellas, Ono Pharmaceutical, Daiichi Sankyo, Taiho Pharmaceutical, Chugai, Merck Pharmaceutical, Amgen, Eisai, PRA Health Sciences, Syneos Health, AstraZeneca, PPD-SNBL K.K., TORAY, Insyte Biosciences G.K, and Medpace Japan K.K. All remaining authors have no competing interests to declare. Supplementary data Supplementary Table S1 Supplementary Table S2 References Associated Data Supplementary Materials Supplementary Table S1 Supplementary Table S2 Data Availability Statement The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.