Claudin-18 status and its correlation with HER2 and PD-L1 expression in gastric cancer with peritoneal dissemination

Abstract

Background: Gastric cancer with peritoneal dissemination (PD) has a dismal prognosis, and current treatments have shown little efficacy. CLDN18.2-targeted therapies have shown promising efficacy against gastric cancers that express high levels of CLDN18. Because of the limited information regarding CLDN18.2 status in PD, we analyzed PD-positive gastric cancers for CLDN18 status in both primary and PD, along with HER2 and PD-L1 combined positive score (CPS).

Methods: Immunohistochemical analyses were performed on 84 gastric cancer cases using paired primary and PD tissue samples.

Results: At 40% cut-off, CLDN18 was positive in 57% (48/84) primary tumors and in 44% (37/84) PDs. At 75% cut-off, 28.6% (24/84) primary tumors and 20.2% (17/84) PDs were CLDN18-positive. The concordance rate between primary tumors and PD was 79.8% at 40% cut-off and 75% at 75% cut-off. When comparing biopsy and surgical specimens, the concordance rates were 87.5% at 40% cut-off and 81.3% at 75% cut-off. Within a tumor, the superficial area tended to have a higher CLDN18-positive rate than the invasive front (P = 0.001). Although HER2 -positivity was only 11.9% in this cohort, CLDN18 positivity in HER2-negative tumors (n = 74) was relatively high: 60.8% at 40% cut-off and 28.4% at 75% cut-off. Among double-negative (HER2 - and PD-L1 CPS < 1) tumors, CLDN18 positivity was 67.6% at 40% cut-off and 26.5% at 75% cut-off.

Conclusions: CLDN18 expression is generally maintained in PD and is relatively high even in double-negative tumors, making it a promising therapeutic target for PD-positive gastric cancer.

Keywords: Claudin-18; Gastric cancer; Immunohistochemistry; Molecular targeted therapy; Peritoneal dissemination.

Fig. 1

Summary of specimens used in each analysis. A total of 84 cases with gastric cancer with PD were included in the study, and five main analyses were performed. The numbers in parentheses indicate the number of cases studied

Fig. 2

Representative images of CLDN18 immunohistochemistry. Pair images (H&E staining and CLDN18 immunohistochemistry) of four categories in CLDN18 intensity: 0, no expression (A and B); 1 + , weak expression (C and D); 2 + , moderate expression (E and F); and 3 + , strong expression (G and H). CLDN18 immunostainings of a representative case with CLDN18-positive in both the primary (I) and peritoneal dissemination (J). CLDN18 immunostainings of a case with discordant results that showed CLDN18-positive in the primary tumor (K) but negative in the peritoneal dissemination (L)

Fig. 3

Comparison of CLDN18 expression between primary tumors and peritoneal dissemination. The percentage of CLDN18-positive tumor cells in the primary tumor is shown on the X-axis, and the percentage of peritoneal dissemination is shown on the Y-axis. The line represents a linear regression curve

Fig. 4

Intratumoral CLDN18 heterogeneity. a A representative case exhibiting heterogeneous CLDN18 expression. Loupe view of H&E and CLDN18 immunostaining (left). At higher power magnification, on the right, diffuse CLDN18 expression in the superficial area (upper pictures in the red box), partial positive staining in the center (middle pictures in the blue box), and almost negative staining in the invasive front (lower pictures in the green box) were noted. b The Percentage of CLDN18-positive neoplastic cells in the superficial, central, and invasive front areas. CLDN18 expression tended to decrease from the superficial to deep areas

Fig. 5

Correlation between CLDN18 status and HER2, PD-L1 CPS, and Lauren histological classification. Although CLDN18 status was not significantly correlated with HER2 and PD-L1, a significant proportion of HER2-negative or PD-L1 CPS < 1 cases were CLDN18-positive