Composition for predicting immune checkpoint inhibitor treatment response in male gastric cancer patients, including PD-L1 ex-pression level measuring agent

Abstract

The present invention relates to a composition for predicting the therapeutic responsiveness of immune checkpoint inhibitors in male gastric cancer patients, and a composition for treating gastric cancer in male gastric cancer patients comprising immune checkpoint inhibitors. In one aspect, the present invention confirmed that PD-L1 expression and the therapeutic effect of ICI differ according to gender in gastric cancer. In men, EBV infection and gastric tumor location were independent risk factors, while in women, MSI-H status was a significant risk factor. These results highlight the importance of gender-specific immunobiological characteristics in customizing gastric cancer treatment strategies. Thus, the composition for predicting ICI therapeutic responsiveness according to one aspect of the present invention has an excellent effect of easily predicting the therapeutic responsiveness of ICI in male gastric cancer patients by immunohistochemically measuring PD-L1 protein expression levels in EBV-positive and / or tumor-located in the gastric antrum to calculate a Composite Positive Score (CPS) and detecting the presence of Epstein-Barr virus (EBV) infection.

Description

Technology Field

[0001] The present specification discloses a composition for predicting the therapeutic responsiveness of an immune checkpoint inhibitor in male gastric cancer patients, and a composition for treating gastric cancer in male gastric cancer patients comprising an immune checkpoint inhibitor. Background Technology

[0002] Gastric cancer (GC) is the sixth most common malignant tumor and the fourth most fatal tumor worldwide. Its incidence is particularly high in East Asia, specifically in Korea, Japan, and China. In Korea, while men historically had the highest incidence and mortality rates for gastric cancer, both figures have steadily decreased, and it now ranks fourth in terms of incidence (76.3 per 100,000 men and 38.2 per 100,000 women). This trend appears to be attributed to factors such as a decrease in Helicobacter pylori infection rates and changes in dietary habits.

[0003] Programmed cell death ligand 1 (PD-L1) is a member of the B7 superfamily and is a transmembrane glycoprotein expressed in immune cells, dendritic cells, epithelial cells, and some tumor cells. Members of the B7 family interact with Programmed Cell Death 1 (PD-1) to inhibit T cell activation [1]. PD-1 / PD-L1 interactions are important regulatory checks for excessive adaptive immune responses to antigens and autoimmunities [2]. In cancer, PD-L1 acts as a negative regulator of T cell receptor activation through PD-1, which is highly expressed in tumor-specific T cells, and inhibits both innate and adaptive immune responses, such as anti-tumor CD8+ cytotoxic T cell function [3].

[0004] Immune checkpoint inhibitors (ICIs) targeting PD-1 / PD-L1 have shown positive results in gastric cancer and other cancers. However, clinical outcomes in gastric cancer patients vary due to the complex heterogeneity of tumor cells. Furthermore, in addition to intrinsic factors such as genetic mutations, morphological features, and pTNM staging, differences in extrinsic factors such as the tumor immune microenvironment also contribute to variability in clinical outcomes. Helicobacter pylori ( H. pylori Infections by pathogens such as Epstein-Barr virus (EBV) and microsatellite instability (MSI) can induce extrinsic inflammatory changes, leading to precancerous lesions. Additionally, tumor cells evade the immune system by reprogramming immune cells within the tumor microenvironment. Tumor-infiltrating lymphocytes (TILs), particularly CD8+ cytotoxic T cells that support tumor cell death, lose their function upon prolonged exposure to cancer cells. Therefore, the immune microenvironment plays a significant role in the development of gastric cancer (GC) [4]. Consequently, it is important to interpret the tumor microenvironment associated with tumor-infiltrating immune cells to predict the response to immunotherapy. PD-L1 is not only a target for immunotherapy but is also used as an indicator of tumor-infiltrating lymphocyte (TIL) density [5].

[0005] Gender is a significant factor influencing the development and function of the immune system. On average, women exhibit stronger innate and adaptive immune responses than men [6]. Differences in T-cell polarization have been reported between men and women, with men tending to exhibit Th1 responses and women tending to exhibit Th2 responses. Additionally, while a Th17 bias is predominantly observed in men, women have been shown to exhibit stronger B-cell responses to various antigens, including various vaccines [7]. Furthermore, men and women differ in immune evasion mechanisms. For example, in non-small cell lung cancer (NSCLC), the tumor microenvironment (TME) in women contains significantly more innate and adaptive immune cell types, including specific T-cell subgroups, compared to men. Conversely, the TME in men exhibits a phenotype that primarily excludes T-cells [8]. Hormonally, sex hormones regulate the expression and function of PD-1 and PD-L1, and hormonal influences on the PD-1 pathway are important factors mediating autoimmunity [9]. Testosterone and estradiol directly influence anticancer responses by regulating innate and adaptive immune cells through specific receptors. Androgen receptors (AR) induce an exhausted phenotype in CD8+ T cells and reduce the expression of tumor necrosis factor (TNF), interferon (IFN)-γ, and granzyme B, leading to dysfunction of the antitumor immune response

[10] . Meanwhile, 17β-estradiol (E2) has been shown to induce estrogen receptor beta (ERβ) expression in women and inhibit the induction of inflammatory and antioxidant enzymes

[11] . In contrast, estrogen receptor alpha (ERα) polarizes tumor-associated macrophages (TAMs) into an immunosuppressive M2 phenotype at the expense of the antitumor M1 phenotype, thereby promoting CD8+ T cell dysfunction and ICI resistance to antitumor effects

[12] .

[0006] Therefore, while inflammatory responses are strong in men, inflammation tends to be suppressed in women, leading to increased PD-1 expression. Interestingly, despite obesity being associated with an increased risk of cancer due to chronic inflammation, a sex-specific analysis of several cohorts of patients with advanced melanoma revealed a positive correlation between higher body mass index (BMI) and better outcomes from immune checkpoint inhibitor (ICI) treatment only in male patients

[13] . Based on these findings, the inventors hypothesized that PD-L1 expression may exhibit different patterns due to sex differences in the immune system, and that the efficacy of ICIs for gastric cancer may also vary by sex. Many prior studies have investigated PD-L1 expression as a prognostic factor in gastric and other cancers

[14] , but these findings are controversial. Studies investigating sex-related differences, including the efficacy of ICI treatment in gastric cancer, are very rare

[15] . Therefore, this study aimed to investigate the factors affecting PD-L1 expression in Korean gastric cancer tissues, focusing on clinicopathological variables, particularly gender and their relationship with prognosis evaluation, and as a result of the research, the present invention was completed. The problem to be solved

[0008] In one aspect, the object of the present invention is to provide a biomarker for predicting the responsiveness to immune checkpoint inhibitor treatment in male gastric cancer patients.

[0009] In another aspect, the object of the present invention is to provide a composition for treating gastric cancer in male gastric cancer patients. means of solving the problem

[0010] In one aspect, the present invention provides a composition for predicting the responsiveness to an immune checkpoint inhibitor (ICI) treatment in male gastric cancer patients, wherein the composition comprises an anti-PD-L1 antibody for calculating a Combined Positive Score (CPS) by immunohistochemically measuring the PD-L1 (Programmed death-ligand 1) protein expression level, and wherein the male gastric cancer patient is EBV positive and / or the tumor location is the gastric antrum.

[0011] In another aspect, the present invention provides a composition for treating gastric cancer in a male gastric cancer patient, wherein the composition comprises an immune checkpoint inhibitor (ICI), said male gastric cancer patient is Epstein-Barr virus (EBV) positive and / or the tumor location is the gastric antrum, and said male gastric cancer patient has a Combined Positive Score (CPS) of PD-L1 protein expression calculated by Formula 1 below of 5 or higher:

[0012] [Equation 1]

[0013] . Effects of the invention

[0014] In one aspect, the present invention confirmed that PD-L1 expression and the therapeutic effect of ICI differ according to gender in gastric cancer. In men, EBV infection and gastric tumor location were independent risk factors, while in women, MSI-H status was a significant risk factor. These results highlight the importance of gender-specific immunobiological characteristics in customizing gastric cancer treatment strategies. Thus, the composition for predicting ICI treatment responsiveness according to one aspect of the present invention has an excellent effect of easily predicting ICI treatment responsiveness in male gastric cancer patients by immunohistochemically measuring the PD-L1 (Programmed death-ligand 1) protein expression level in male patients who are EBV-positive and / or whose tumor is located in the gastric antrum to calculate a Combined Positive Score (CPS) and detecting the presence of Epstein-Barr virus (EBV) infection. Brief explanation of the drawing

[0016] FIG. 1 is a diagram for recruiting research participants according to one embodiment of the present invention. Figure 2 is a graph showing the results of comparing cumulative survival rates according to the PD-L1 expression status according to gender, confirmed according to one embodiment of the present invention. Figure 3 is a graph showing the results of comparing the cumulative survival rate according to whether ICI was administered according to gender, as confirmed according to one embodiment of the present invention. Figure 4 is a graph showing the results of comparing cumulative survival rates according to ICI treatment and gender according to one embodiment of the present invention. Figure 5 is a graph showing the results of comparing the cumulative survival rate according to the first ICI treatment and the second treatment according to one embodiment of the present invention. Figure 6 is a graph showing the results of comparing the cumulative survival rates of ICI monotherapy and ICI-chemotherapy combination therapy according to one embodiment of the present invention. Specific details for implementing the invention

[0017] The present invention will be described in detail below.

[0019] In one aspect, the present invention provides a composition for predicting the responsiveness to an immune checkpoint inhibitor (ICI) treatment in male gastric cancer patients, wherein the composition comprises an anti-PD-L1 antibody for calculating a Combined Positive Score (CPS) by immunohistochemically measuring the level of PD-L1 (Programmed death-ligand 1) protein expression, and wherein the male gastric cancer patient is EBV positive and / or the tumor location is the gastric antrum.

[0020] A composition according to one aspect of the present invention may include an anti-PD-L1 antibody for calculating a Combined Positive Score (CPS) by immunohistochemically measuring the expression level of PD-L1 (Programmed death-ligand 1) protein.

[0021] An anti-PD-L1 antibody according to one aspect of the present invention may be an antibody that specifically binds to a human PD-L1 protein, and specifically may include a clinically validated anti-PD-L1 antibody clone.

[0022] An antibody according to one aspect of the present invention may be one or more selected from the group consisting of polyclonal antibodies, monoclonal antibodies, recombinant antibodies, and combinations thereof. More specifically, the antibody may include a polyclonal antibody, a monoclonal antibody, a recombinant antibody, and a complete form having two full-length light chains and two full-length heavy chains, as well as functional fragments of the antibody molecule, e.g., Fab, F(ab'), F(ab')2, and Fv. Antibody production can be easily performed using techniques widely known in the field to which the present invention belongs, and antibodies that are manufactured and commercially available may be used.

[0023] In one aspect of the present invention, therapeutic responsiveness to an immune checkpoint inhibitor can be predicted by measuring the level of PD-L1 expression in a subject's sample using the anti-PD-L1 antibody. The expression level can be measured according to conventional immunoassay methods, and can be measured through radioimmunoassay, radioimmunoprecipitation, immunoprecipitation, immunohistochemical staining, ELISA (enzyme-linked immunosorbent assay), capture-ELISA, inhibition or competition analysis, sandwich analysis, flow cytometry, immunofluorescence staining, and immunoaffinity purification using the antibody against PD-L1, and specifically, can be measured through immunohistochemical staining.

[0024] A composition according to one aspect of the present invention may further include a label that enables quantitative or qualitative measurement of the formation of an antigen-antibody complex, a conventional tool used in immunological analysis, a reagent, etc.

[0025] In one aspect of the present invention, labels that enable qualitative or quantitative measurement of the formation of the antigen-antibody complex include, but are not limited to, enzymes, fluorescent agents, ligands, luminescent agents, microparticles, redox molecules, and radioisotopes. Enzymes available for use as detection labels include, but are not limited to, β-glucuronidase, β-D-glucosidase, β-D-galactosidase, urease, peroxidase, alkaline phosphatase, acetylcholinesterase, glucose oxidase, hexokinase and GDPase, RNase, glucose oxidase and luciferase, phosphofructokinase, phosphoenolpyruvate carboxylase, aspartate aminotransferase, phosphphenolpyruvate decarboxylase, β-latamase, etc. Fluorescent materials include, but are not limited to, fluorescein, isothiocyanates, rhodamine, phycoerytherin, phycocyanin, allophycocyanin, o-phthaldehydrides, fluorescarmine, etc. Ligands include, but are not limited to, biotin derivatives, etc. Luminescent materials include, but are not limited to, acridinium esters, luciferin, luciferase, etc. Microparticles include, but are not limited to, colloidal gold, colored latex, etc. Redox molecules include ferrocene, ruthenium complexes, viologen, quinone, Ti ion, Cs ion, diimide, 1,4-benzoquinone, hydroquinone, K4W(CN) 8 , [Os(bpy)3] 2+ , [RU(bpy)3] 2+ , [MO(CN)8] 4- These include, but are not limited to. Radioisotopes include 3 H, 14 C, 32 P, 35 S, 36 Cl, 51 Cr, 57 Co, 58 Co, 59 Fe, 90 Y, 125 I, 131 I, 186There are Re, etc., and are not limited to this.

[0026] An anti-PD-L1 antibody according to one aspect of the present invention may be used to calculate a Combined Positive Score (CPS) by immunohistochemically measuring the PD-L1 protein expression level. Specifically, according to one aspect of the present invention, the PD-L1 protein expression level may be quantified by the CPS. More specifically, the CPS may be calculated by the following Formula 1:

[0027] [Equation 1]

[0028] .

[0029] According to one aspect of the present invention, CPS may have a range of 0 to 100, and specifically, in one aspect of the present invention, another composition may predict that the patient is likely to have improved survival with ICI treatment when the CPS is 5 or higher.

[0030] An evaluation of CPS according to one aspect of the present invention can be performed by a pathologist and quantified using an automated image analysis system.

[0031] A composition according to one aspect of the present invention may include a probe or antibody for detecting whether there is an infection with Epstein-Barr virus (EBV).

[0032] According to one aspect of the present invention, whether or not there is an EBV infection can be confirmed through an in situ hybridization (ISH) reaction that detects EBV-encoded RNA (EBER). Specifically, the presence or absence of EBV infection can be confirmed by an EBV-in situ hybridization (Epstein-Barr virus-encoded RNA in situ hybridization) reaction.

[0033] A probe according to one aspect of the present invention may be a nucleic acid probe that binds complementarily to an EBER1 or EBER2 sequence, and may be attached with a fluorescent label or an enzyme label.

[0034] According to one aspect of the present invention, "probe" means a polynucleotide having a sequence of bases capable of binding complementarily to a target site of a gene, a variant thereof, or a polynucleotide and a labeling substance bound thereto.

[0035] "Hybridization" according to one aspect of the present invention means that two single-stranded nucleic acids form a duplex structure through the pairing of complementary base sequences. Hybridization can occur not only when the complementarity between the single-stranded nucleic acid sequences is perfect match, but also when some mismatched bases are present.

[0036] In one aspect of the present invention, examples of tools or reagents include, but are not limited to, suitable carriers, solvents, cleaning agents, buffers, stabilizers, etc. If the labeling substance is an enzyme, it may include a substrate capable of measuring enzyme activity and a reaction stopping agent. The carrier may be a soluble carrier or an insoluble carrier, and an example of a soluble carrier is a physiologically acceptable buffer known in the art, e.g., PBS, and an example of an insoluble carrier may be polystyrene, polyethylene, polypropylene, polyester, polyacrylonitrile, fluoropolymer, cross-linked dextran, polysaccharide, other paper, glass, metal, agarose, and combinations thereof.

[0037] A composition according to one aspect of the present invention can be determined to be EBV positive when a specific signal is observed within the nucleus of a tumor cell in an EBV-in situ hybridization reaction. Specifically, the EBV positivity may be a case where tumor cells show a positive reaction in an EBV-in situ hybridization (ISH) test result in a gastric tissue specimen obtained through surgery or an endoscopic procedure such as endoscopic mucosal resection (ESD).

[0038] A composition according to one aspect of the present invention may be for predicting the responsiveness to immune checkpoint inhibitor (ICI) treatment in male gastric cancer patients.

[0039] A male gastric cancer patient according to one aspect of the present invention may be a male gastric cancer patient who is EBV positive and / or whose tumor location is in the gastric antrum.

[0040] An immune checkpoint inhibitor according to one aspect of the present invention may inhibit the signaling pathway of PD-1 or PD-L1. Specifically, the immune checkpoint inhibitor may be one or more selected from the group consisting of pembrolizumab, nivolumab, and tislelizumab. More specifically, the immune checkpoint inhibitor may be administered alone or in combination with one or more anticancer chemotherapy agents selected from the group consisting of pembrolizumab, nivolumab, and tislelizumab. The anticancer chemotherapy agents may include fluoropyrimidine-based drugs and platinum-based drugs.

[0041] A composition according to one aspect of the present invention may be a male gastric cancer patient who is EBV-positive and / or whose tumor location is in the gastric antrum, and who is predicted to have a high probability of survival improvement with ICI treatment if the CPS calculated with the anti-PD-L1 antibody is 5 or higher.

[0043] In another aspect, the present invention provides a composition for treating gastric cancer in a male gastric cancer patient, wherein the composition comprises an immune checkpoint inhibitor (ICI), said male gastric cancer patient is Epstein-Barr virus (EBV) positive and / or the tumor location is the gastric antrum, and said male gastric cancer patient has a Combined Positive Score (CPS) of PD-L1 protein expression calculated by Formula 1 below of 5 or higher:

[0044] [Equation 1]

[0045] .

[0046] The descriptions regarding the above male gastric cancer patient, ICI, EBV positivity, PD-L1 protein expression, CPS, etc., can be applied to the above composition for predicting ICI treatment responsiveness.

[0047] An immune checkpoint inhibitor according to one aspect of the present invention may inhibit the PD-1 or PD-L1 signaling pathway. Specifically, the immune checkpoint inhibitor may be a PD-L1 inhibitor, and more specifically, the PD-1 inhibitor may be one or more selected from the group consisting of pembrolizumab, nivolumab, and tislelizumab.

[0048] A composition for treating gastric cancer according to one aspect of the present invention may be administered with the immune checkpoint inhibitor alone or in combination with an anticancer chemotherapy agent. The anticancer chemotherapy agent may include a fluoropyrimidine-based drug and a platinum-based drug.

[0050] The structure and effects of the present invention will be explained in more detail below through examples and experimental examples. However, the following examples and experimental examples are provided for illustrative purposes only to aid in understanding the present invention, and the scope and range of the present invention are not limited by them.

[0052] Materials and Methods

[0054] [Example 1] Research group

[0055] Of the 15,021 gastric cancer patients who participated in a prospective cohort study conducted at Seoul National University Bundang Hospital (SNUBH) from May 2003 to January 2024, 468 patients who underwent PD-L testing were included in this study (Fig. 1). Information on age, sex, tumor location, histological type (WHO and Lauren classification criteria), molecular characteristics (MSI and EBV positivity), stage (American Joint Committee on Cancer [AJCC] 8th edition criteria), treatment method, survival information, and cause of death was collected and analyzed through the medical records of the surgery and medical cohorts established in 2003.

[0057] [Example 2] PD-L1 Status Analysis

[0058] Formalin-fixed, paraffin-embedded (FFPE) tissue sections (4 μm thickness) were deparaffinized and rehydrated, followed by heat-induced antigen recovery treatment using EnVision FLEX Target Retrieval Solution (high pH) at 97°C for 20 minutes. Endogenous peroxidase activity was inhibited by treatment with EnVision FLEX Peroxidase-Blocking Reagent for 5 minutes, followed by incubation with the primary monoclonal anti-PD-L1 antibody (clone 22C3) at room temperature for 20 minutes. PD-L1 expression was visualized by staining with EnVision FLEX DAB+ Chromogen for 10 minutes, and a brown precipitate was formed in PD-L1-positive cells. The sections were counterstained with hematoxylin for 5 minutes, rinsed, dehydrated, and then mounted using a permanent mounting medium. Tumor PD-L1 expression was evaluated using the Combined Positive Score (CPS). The CPS was calculated by dividing the number of PD-L1-positive tumor cells, lymphocytes, and macrophages by the total number of viable tumor cells and multiplying by 100. A CPS of 5 or higher was considered PD-L1 positive.

[0060] [Example 3] EBV-in situ hybridization

[0061] In the diagnosis of EBV-associated gastric cancer (EBVaGC), the term "EBV positive" was used to refer to cases where tumor cells showed a positive reaction in EBV-in situ hybridization (ISH) tests in gastric tissue specimens obtained through surgery or endoscopic procedures such as endoscopic mucosal resection (ESD). EBV-ISH was performed using INFORM EBER (Epstein-Barr Virus Early RNA) probe (Product No.: 05278660001) manufactured by Roche, VENTANA ISH iVIEWBlue detection kit (Product No.: 05278511001), ISH protease 2 (Product No.: 05273323001), and red stain II (Product No.: 05272017001). FFPE tissues were sectioned to a thickness of 3 μm for EBV-ISH testing. Tissue sections were deparaffinized at 75°C and pretreated with ISH Protease 2 (Product No.: 05273323001) for 8 minutes. Hybridization and visualization were performed according to a pre-fixation protocol using a probe and detection kit, followed by counterstaining with Red Dye II for 4 minutes. For each case, a representative whole-section slide containing the deepest invasion site of the gastric cancer was selected to perform in situ EBER hybridization.

[0063] [Example 4] Data Variables

[0064] Tumor location was classified into superior, middle, and inferior regions based on surgical pathology results. All tumors classified according to the WHO classification system were reclassified into differentiated (well and moderate), undifferentiated (poor and signet ring cell carcinoma), mixed, and other types. Additionally, they were classified into intestinal, diffuse, mixed, and indeterminate types according to the Lauren classification system, with the classification confirmed after surgery or endoscopic procedures such as ESD. MSI status was assessed using two single nucleotide markers and three double nucleotide markers. MSI-H indicates instability in two or more of the five markers. The presence of EBV within cancer cells was evaluated using EBV-ISH. Early gastric cancer (EGC) was defined as cancer invasion confined to the submucosal layer, while advanced gastric cancer (AGC) was defined as invasion extending beyond the submucosal layer. Cancer staging was performed using the TNM staging system according to the 8th edition of the AJCC. The dates and causes of death of the study participants were reviewed against electronic medical records (EMR) and national statistics data for verification.

[0066] [Example 5] Clinical Results

[0067] The primary outcome of this study was to analyze clinicopathological factors influencing the PD-L1 positivity rate in relation to patient gender. The secondary outcome was to analyze overall survival and gastric cancer-specific survival rates based on PD-L1 positivity and treatment with immune checkpoint inhibitors (ICIs) in relation to patient gender. Overall survival (OS) was defined as the time from the date of gastric cancer diagnosis to death from all causes. The dates and causes of death for enrolled patients were verified using data from the National Statistical Office. Randomization information guaranteeing patient anonymity was collected through a third party and submitted to the National Statistical Office, which provided data regarding patient deaths. The follow-up period was up to five years; for patients whose follow-up was discontinued within five years, the period from diagnosis to the last follow-up was estimated as the survival period. For patients who received hospice care, the survival period was calculated based on the actual date of death.

[0069] [Example 6] Statistical Analysis

[0070] Statistical analysis was performed using IBM SPSS software (version 27.0; IBM Corp., Armonk, NY, USA) and R software (version 4.5.1; R Foundation for Statistical Computing, Vienna, Austria). Reference characteristics and variables were analyzed using chi-square tests and t-tests. In particular, Firth's penalized maximum likelihood estimation was applied to the multivariate logistic regression model for PD-L1 positivity to correct for small sample bias in the subgroup analysis by gender. Subgroup analysis by gender was also performed. Variables with a p-value < 0.2 in the univariate analysis were used as covariates in the multivariate analysis. Overall survival (OS) was estimated using the Kaplan-Meier method, and differences between curves were evaluated using the log-rank test. Additionally, Cox proportional hazards regression analyses were performed to evaluate the association between clinicopathological factors and overall survival. In multivariate analysis, a p-value of less than 0.05 was considered statistically significant.

[0072] [Example 7] Ethical Statement

[0073] This study was reviewed and approved by the Institutional Review Board (B-2006-618-004) of Seoul National University Bundang Hospital, and informed consent was waived. The initial cohort of this study was registered on ClinicalTrials.gov (NCT04973631).

[0075] Experimental results

[0077] [Experimental Example 1] Clinicopathological characteristics according to PD-L1 expression status

[0078] A total of 468 patients participated in the study, of whom 280 (59.8%) were PD-L1 positive and 188 (40.2%) were PD-L1 negative (Figure 1). The baseline characteristics of the patients are summarized in Table 1. The average age of the PD-L1 positive group was 61.3 years, and that of the negative group was 58.9 years, with no significant difference. Cases where the GC was located in the lower 1 / 3 occurred significantly more frequently in the group with high PD-L1 expression (43 cases, 23.0% vs. 104 cases, 37.1%) (p = 0.005), and there were also significant differences in intestinal histological findings according to Lauren classification (47 cases, 25.7% vs. 86 cases, 33.0%), and it occurred significantly more frequently in EBV-positive patients than in EBV-negative patients (4 cases, 2.4% vs. 19 cases, 7.4%) (p = 0.043).

[0079] [Table 1]

[0080] Basis characteristics and multivariate analysis of factors affecting PD-L1 expression

[0081]

[0082] Values ​​are expressed as mean ± standard deviation or persons (%). AGC, advanced gastric cancer; CI, confidence interval; EBV, Epstein-Barr virus (EBV); EGC, early gastric cancer; GC, gastric cancer; MSI-H, high microsatellite instability; MSI-L, low microsatellite instability; MSS, microsatellite stable; OR, odds ratio; PD-L1, programmed cell death ligand-1; SD, standard deviation.

[0084] [Experimental Example 2] Multivariate analysis of PD-L1 positivity related to clinicopathological factors

[0085] To evaluate the factors influencing PD-L1 expression in gastric cancer, a multivariate analysis using Firth's penalty logistic regression was performed. The results and statistical significance are summarized in Table 1. The multivariate analysis revealed that the lower 1 / 3 location of gastric cancer (odds ratio [OR] = 1.84, 95% CI: 1.07-3.21, p = 0.027), MSI-H (OR = 5.04, 95% CI: 1.17-47.10, p = 0.027), and EBV positivity (OR = 7.46, 95% CI: 2.26-28.22, p < 0.001) were significantly associated with PD-L1 positivity. Diffuse histological findings showed a negative correlation (OR = 0.22, 95% CI: 0.04-0.94, p = 0.041).

[0087] [Experimental Example 3] Survival Analysis Based on PD-L1 Positivity

[0088] Of the 468 patients initially enrolled, 443 were included in the survival analysis after excluding patients who received budigalimab (n=4, 3 men, 1 woman), avelumab (n=5, 5 men), and combination therapy of durvalumab and tremelimumab (n=11, 10 men, 1 woman) due to small sample sizes and potential bias. Only patients treated with pembrolizumab, nivolumab, or tislelizumab were included in the analysis. The distribution of ICI and ICI combination chemotherapy is presented in Table 2. During the follow-up period, 230 patients (51.9%) died from gastric cancer. Five patients (2 men and 3 women) were excluded from follow-up due to transfer to other cancer centers (2 women), return to their home country (1 woman), or insufficient outcome data (2 men). The relationship between PD-L1 expression and overall survival (OS) was analyzed in the entire cohort (Fig. 2A), men (Fig. 2B), and women (Fig. 2C) through survival analysis using the Kaplan-Meier method. The PD-L1 positive group had a lower survival rate at the 5-year mark, but there was no statistically significant difference (p = 0.481). Additionally, overall survival (OS) according to ICI treatment was analyzed in the entire cohort (Fig. 3A) and gender subgroups (men: Fig. 3B, women: Fig. 3C). Of the total patients, 177 received ICI treatment, while 266 did not. The median OS of the ICI treatment group (1,218 days) was longer than that of the non-treatment group (902 days), but there was no statistically significant difference (p = 0.136).

[0089] [Table 2]

[0090] Distribution of combination chemotherapy regimens by type of immune checkpoint inhibitor

[0091]

[0092] CapOX, capecitabine plus oxaliplatin; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; ICI, immune checkpoint inhibitor; SOX, S-1 plus oxaliplatin; SP, S-1 plus cisplatin; XELOX, capecitabine plus oxaliplatin; XP, capecitabine plus cisplatin.

[0094] [Experimental Example 4] Analysis of subgroups according to gender

[0095] To investigate differences in PD-L1 expression, the study groups were divided by gender. The baseline characteristics associated with PD-L1 positivity in men and women are summarized in Tables 3 and 4. In the male group, PD-L1 positive tumors tended to be located primarily in the lower one-third of the stomach (n=80, 42.3%), showing a statistically significant difference compared to PD-L1 negative gastric cancer (n=31, 27.7%) (p=0.039). In contrast, this tendency was less common in the female group, and no statistically significant difference was observed (n=24, 26.4% vs. n=12, 16%; p=0.254). According to the Lauren classification, intestinal-type tumors were observed more frequently in PD-L1 positive patients than in negative patients in both sexes, but a statistically significant difference was observed only in the female group (n=19, 21.1% vs. n=6, 8.0%; p=0.038). This trend was not statistically significant in the male group (n=67, 39.2% vs. n=41, 38.0%; p=0.101). MSI-H was more common in PD-L1-positive patients in both sexes, but there was no statistically significant difference. The EBV positivity rate was significantly higher in the PD-L1-positive group than in the negative group among male patients (n=18, 10.4% vs. n=1, 1.0%, p=0.007), but no such difference was observed in the female group. Other variables, such as age and TNM stage, did not show statistically significant differences from PD-L1 expression in either the male or female groups.

[0096] [Table 3]

[0097] Basic characteristics and multivariate analysis of factors affecting PD-L1 expression according to male gender

[0098]

[0099] AGC, advanced gastric cancer; CI, confidence interval; EBV, Epstein-Barr virus (EBV); EGC, early gastric cancer; GC, gastric cancer; MSI-H, high microsatellite instability; MSI-L, low microsatellite instability; MSS, microsatellite stable; OR, odds ratio; PD-L1, programmed cell death ligand-1.

[0101] [Table 4]

[0102] Basic characteristics and multivariate analysis of factors affecting PD-L1 expression according to female gender

[0103]

[0104] AGC, advanced gastric cancer; CI, confidence interval; EBV, Epstein-Barr virus (EBV); EGC, early gastric cancer; GC, gastric cancer; MSI-H, high microsatellite instability; MSI-L, low microsatellite instability; MSS, microsatellite stable; OR, odds ratio; PD-L1, programmed cell death ligand-1.

[0106] In addition, to investigate the factors influencing PD-L1 expression according to sex differences, multivariate analysis using Firth's penalty logistic regression was performed for each subgroup, and the results are summarized in Tables 2 and 3. The results of the multivariate analysis showed that in the male group, low tumor location (OR = 2.38, p = 0.011) and EBV positivity (OR = 36.27, p < 0.001) exhibited statistically significant differences. In the female group, MSI-H was the only factor significantly associated with PD-L1 positivity (OR = 11.63, p = 0.030). Although the Lauren classification did not show overall statistical significance, the mixed histological type of the Lauren classification showed borderline significant results (OR = 0.11, p = 0.072).

[0107] Survival analysis using the Kaplan-Meier method revealed differences in cumulative survival curves between the male and female groups. In the male group, the median survival of PD-L1-positive patients tended to be longer than that of PD-L1-negative patients (1,156 days vs. 961 days). However, this difference was not statistically significant (p = 0.941) (Fig. 2A). In the female group, the median survival of PD-L1-positive patients tended to be shorter than that of PD-L1-negative patients (825 days vs. 1,020 days). The p-value for the log-rank test was 0.193 (Fig. 2C).

[0108] In a survival analysis comparing patients treated with chemotherapy including ICI and those not treated, men showed a significantly longer survival time in the ICI treatment group (1,314 days vs. 950 days) compared to the non-ICI treatment group, a statistically significant difference (p = 0.005) (Fig. 3B). On the other hand, for women, there was no significant difference in survival time between the ICI treatment group and the non-ICI treatment group (897 days vs. 890 days, p = 0.415) (Fig. 3C). Further analysis was performed on ICI-treated patients regarding ICI treatment (Fig. 4), treatment stage (Fig. 5), and monotherapy versus combination therapy (Fig. 6). In particular, a significant difference in survival rates was observed in male patients depending on the ICI treatment (p = 0.038) (Fig. 4B).

[0109] To further evaluate the differential effects of ICI treatment according to gender, a multivariate Cox proportional hazards regression analysis including an interaction term between ICI and gender was performed (Table 5). The overall effect of ICI treatment showed a trend toward improved survival rates, but it was not statistically significant (hazard ratio [HR] = 0.70, p = 0.080). The interaction term (ICI × gender) suggested the possibility of gender-specific differences in ICI efficacy. Compared to male patients who did not receive ICI treatment (control group), the mortality hazard ratio for female patients treated with ICI was 1.77 (p = 0.073), suggesting the possibility that the survival benefit was somewhat reduced in women compared to men. Other significant covariates included in the model were female (HR = 0.63, p = 0.025) and EBV positivity (HR = 0.39, p = 0.036), both of which were associated with a reduced risk of death. On the other hand, histological subtypes other than the intestinal type were associated with increased mortality (HR = 2.28, p < 0.001). These factors were associated independently with overall survival. Although TNM staging was included in the multivariate model, no stage showed a statistically significant association with overall survival.

[0110] [Table 5]

[0111] Multivariate Cox proportional hazards regression analysis of factors related to overall survival

[0112]

[0113] ICIX HR for gender indicates how much the effect of ICI differs by gender. AGC, advanced gastric cancer; CI, confidence interval; EBV, Epstein-Barr virus (EBV); EGC, early gastric cancer; GC, gastric cancer; HR, hazard ratio; ICI, immune checkpoint inhibitor; MSI-H, high microsatellite instability; MSI-L, low microsatellite instability; MSS, microsatellite stable; PD-L1, programmed cell death ligand-1.

[0115] Discussion

[0117] The results of this study on gender differences in PD-L1 expression and prognosis after ICI treatment showed that EBV-ISH positivity and tumor location in the gastric antrum had a significant impact on PD-L1 expression in men, but not in women. These gender differences in PD-L1 expression are similar to the results of EBV-associated gastric cancer (EBVaGC) reported in previous studies. EBVaGC is classified as a highly immunoreactive tumor and is known to have a favorable prognosis, particularly in men

[16] . It accounts for 9.9% of all gastric cancers and occurs about four times more frequently in men (13.2%) than in women (3.34%)

[17] . Additionally, EBVaGC occurs more than twice as often in the body or cardia than in the gastric antrum

[17] . Furthermore, EBVaGC is characterized by high expression of PD-L1 / 2 and exhibits a unique histological subtype known as gastric cancer with lymphoid stroma (GC with lymphoid stroma)

[18] .

[0118] In this study, the increased PD-L1 expression in the gastric antrum of male patients can be explained by several factors. Under inflammatory conditions, PD-L1 expression is upregulated to prevent tissue damage

[19] . The gastric antrum is more frequently infected with Helicobacter pylori to avoid direct exposure to gastric acid secreted by parietal cells in the body of the stomach; therefore, chronic inflammation in this region can stimulate PD-L1 expression through a feedback mechanism. Intestinal gastric cancer, which has a higher PD-L1 positivity rate than diffuse gastric cancer, is also associated with MSI and occurs more frequently in the gastric antrum

[20] . These results suggest that PD-L1 expression is more widespread in intestinal gastric cancer in the gastric antrum where chronic inflammation persists. Thus, tumor location in the gastric antrum can independently influence PD-L1 expression, particularly in men. Histological subtypes such as intestinal gastric cancer may also contribute to this association, but they did not show a statistically significant independent effect in the cohort of this study.

[0119] MSI-H status was the only factor independently associated with PD-L1 expression in women. MSI-H tumors are characterized by a high mutational burden, which generates abundant neoantigens and triggers a potent immune response. This leads to increased infiltration of activated T cells and the upregulation of immune checkpoint molecules, including PD-L1, as an adaptive resistance mechanism. Sex-specific immune responses can amplify this association. The potent immune response induced by estrogen can further accentuate PD-L1 expression as a feedback mechanism to prevent excessive immune activation, particularly in response to the immunogenicity induced by MSI. Conversely, the anti-inflammatory effects of estrogen may reduce the influence of tumor location on PD-L1 expression

[11] . In contrast, testosterone, the major sex hormone in men, exerts immunosuppressive effects, suppressing immune activation and intensifying inflammatory responses. Due to the dominance of immunosuppressive mechanisms, overexpression of PD-L1 for immune evasion may be lower in men. Consequently, PD-L1 upregulation in men may occur as a response to chronic inflammation of the stomach rather than being influenced by histological type.

[0120] Overall, the high proportion of male patients in this study cohort is influenced by general trends

[21] . Therefore, factors that strongly affect PD-L1 expression in women, or factors that do not have a statistically significant effect in women, may have been masked due to the larger number of male patients.

[0121] The relationship between PD-L1 expression and prognosis remains controversial, with some reports presenting conflicting views on prognosis—some suggesting a good prognosis, others a poor one—or even claiming there is no significant correlation

[14] . In this study, PD-L1 expression did not show a significant association with prognosis in either the overall cohort or the gender subgroup analysis. Therefore, PD-L1 positivity, particularly CPS ≥ 5, was found not to act as an independent prognostic factor for gastric cancer patients on its own. However, multivariate Cox proportional hazards regression analysis revealed that several other clinicopathological factors were associated with a significant association with overall survival. Specifically, female gender, EBV positivity, and intestinal-type gastric cancer were associated with reduced risk. Although PD-L1 expression alone did not show a significant association with prognosis in this study cohort, it may possess prognostic value when interpreted in conjunction with other immunological or molecular characteristics. Previous studies have reported that PD-L1 expression alone cannot predict prognosis, but that it can provide prognostic value when considered alongside other immunological or molecular factors. Furthermore, in gastric cancer, the tumor microenvironment and immune cell composition vary depending on the molecular subtype, and in particular, EBV-positive and MSI-H tumors are characterized by increased lymphocyte infiltration and increased immune checkpoint molecule expression

[22] . For example, Choi et al. reported that the combination of PD-L1 expression and CD8 tumor-infiltrating lymphocytes (TILs) acts as an independent prognostic factor in gastric cancer

[23] . Tumors showing PD-L1-positive cells, PD-L1-negative immune cells, and low CD8 TIL infiltration are associated with a poor prognosis

[23] . Additionally, Noh et al. reported that tumors that are PD-L1-positive with high CD8 TIL density showed the best prognosis, while tumors that are PD-L1-positive with low CD8 TIL density showed the worst prognosis

[24] .These findings suggest that high TIL density is associated with a favorable prognosis, and that PD-L1 expression shows a strong correlation with TIL density. The favorable prognosis of EBV-associated gastric cancer (EBVaGC) and high PD-L1 expression may also be associated with high TIL density [5, 17]. Previous studies have used various approaches to investigate whether the prognosis of gastric cancer differs depending on PD-L1 status. However, since PD-L1 is an immune-related ligand and the immune system exhibits distinct differences based on sex, an analysis focusing on sex differences is essential. Most previous gastric cancer studies have not included such sex-related analyses. In contrast, this study comprehensively evaluated sex differences, including EBV and MSI status, at a single institution through a relatively large-scale cohort study. Therefore, the key strength of this study is that it conducted an in-depth analysis of sex differences in gastric cancer, including survival outcomes related to ICI treatment.

[0122] Gender differences in the efficacy of ICI-based chemotherapy have been reported in melanoma and non-small cell lung cancer (NSCLC)

[24] . In this study, male patients who received ICI-containing chemotherapy showed significantly improved survival rates, but this effect was not observed in female patients. In particular, although the interaction term between gender and ICI treatment in the Cox model did not reach a statistically significant level, it showed a meaningful trend indicating gender-specific differences in treatment efficacy. This trend is consistent with the results of the Kaplan-Meier survival analysis, in which a significant survival benefit of ICI treatment was observed in male patients, whereas this benefit was not observed in female patients. This difference may be attributed to differences in immune mechanisms between genders. In male patients, immunosuppression caused by the influence of androgens can lead to reduced T-cell surveillance function and the sustained proliferation of highly immunogenic cancer cells. Therefore, exhausted T cells with high PD-L1 or CTLA-4 expression are relatively more abundant in men, inducing an immune response that leads to a more favorable response to ICI [8, 25, 26]. In contrast, female patients exhibit more activated innate and adaptive immune responses, including increased CD8+ T cell differentiation and activation

[27] . Estrogen and selective estrogen receptor modulators (SERMs) can affect immune checkpoints such as PD-L1, CTLA-4, and PD-1, suggesting that they function similarly to ICIs

[28] . Castro et al. suggested that in this environment, early cancers in female patients may be eliminated more aggressively than in male patients. Consequently, tumors in female patients undergo potent immune editing to avoid this elimination, resulting in the accumulation of mutations that are not properly expressed by MHC-I and MHC-II molecules. Consequently, tumors with low immunogenicity may survive and progress [29, 30].Therefore, compared to GC in male patients, advanced GC in female patients may have lower immunogenicity and stronger immune evasion mechanisms, which may be the reason why female GC patients show resistance to ICI

[30] .

[0123] Furthermore, because there are more immune cells in the tumor microenvironment of women, the expression of other immune checkpoint molecules such as TIM3, TIGIT, and VISTA, as well as immunosuppressive cells such as cancer-associated fibroblasts (CAF) and myeloid-derived suppressor cells (MDSC), is higher, which contributes to resistance to ICI therapy

[24] . Despite more efficient initial immune recognition and response to early-stage cancer, female patients exhibit more complex and redundant resistance mechanisms, such as those resulting from a greater depletion of CD4+ and CD8+ T cells within the tumor.

[0124] In conclusion, this study highlights gender differences in PD-L1 expression and immunotherapy outcomes in gastric cancer patients. EBV infection and tumor location were significant factors influencing PD-L1 positivity in men, but not in women. MSI was a significant factor influencing PD-L1 positivity only in women. Unlike women, men demonstrated a significant survival benefit from ICI-containing chemotherapy. These results highlight the need for gender-specific strategies in gastric cancer treatment and raise the need for further research to validate and expand upon these findings.

[0126] Through this, it was confirmed that PD-L1 expression and the therapeutic effect of ICI differ according to gender in gastric cancer. In men, EBV infection and gastric tumor location were independent risk factors, while in women, MSI-H status was a significant risk factor. These results emphasize the importance of gender-specific immunobiological characteristics in customizing gastric cancer treatment strategies. Thus, it was found that the composition for predicting ICI treatment responsiveness according to one aspect of the present invention can easily predict ICI treatment responsiveness in male gastric cancer patients by immunohistochemically measuring the PD-L1 (Programmed death-ligand 1) protein expression level in male patients who are EBV-positive and / or whose tumor is located in the gastric antrum to calculate a Combined Positive Score (CPS) and detecting the presence of Epstein-Barr virus (EBV) infection.

[0128] In this specification, [N] (where N is an integer) represents the number of the reference below.

[0129]

Claims

Claim 1 A composition for predicting the responsiveness to an immune checkpoint inhibitor (ICI) treatment in male gastric cancer patients, wherein the composition comprises an anti-PD-L1 antibody for calculating a Combined Positive Score (CPS) by immunohistochemically measuring the PD-L1 (Programmed death-ligand 1) protein expression level; and a probe or antibody for detecting Epstein-Barr virus (EBV) infection, wherein the male gastric cancer patient is EBV positive and / or the tumor location is the gastric antrum. Claim 2 In claim 1, the composition for predicting ICI therapeutic responsiveness, wherein the CPS is calculated by the following Formula 1: [Formula 1] . Claim 3 In claim 1, the composition is a composition for predicting ICI treatment responsiveness, which predicts a patient with a high probability of survival improvement to ICI treatment when the CPS is 5 or higher. Claim 4 A composition for predicting ICI therapeutic responsiveness according to claim 1, wherein the presence of EBV infection is confirmed by an EBV-in situ hybridization (Epstein-Barr virus-encoded RNA in situ hybridization) reaction. Claim 5 A composition for predicting ICI treatment responsiveness, wherein the immune checkpoint inhibitor is one or more selected from the group consisting of pembrolizumab, nivolumab, and tislelizumab. Claim 6 A composition for treating gastric cancer in male gastric cancer patients, wherein the composition comprises an immune checkpoint inhibitor (ICI), the male gastric cancer patient is Epstein-Barr virus (EBV) positive and / or the tumor location is the gastric antrum, and the male gastric cancer patient has a Combined Positive Score (CPS) of PD-L1 protein expression calculated by Formula 1 below of 5 or higher: [Formula 1] . Claim 7 In claim 6, the above immune checkpoint inhibitor is a PD-L1 inhibitor, a composition for treating gastric cancer. Claim 8 A composition for treating gastric cancer according to claim 7, wherein the PD-1 inhibitor is one or more selected from the group consisting of pembrolizumab, nivolumab, and tislelizumab. Claim 9 In claim 6, the above composition is a composition for treating gastric cancer administered in combination with an anticancer chemotherapy agent.

Images