1 Background

Prostate cancer is the most common cancer among men in Western societies and the 2nd most diagnosed cancer in men worldwide [1]. There are an estimated 288,300 new prostate cancer cases and 34,700 deaths expected for 2023 in the United States [2].

In Egypt, according to GLOBOCAN 2020, prostate cancer is the 7th most prevalent cancer, with 4767 new cases and 2227 deaths. From the age of 50 onward, there is an estimated 22% frequency of prostate cancer among Egyptian men [3].

Age, family history, and a few genetic abnormalities (such as BRCA1 and BRCA2) are the only factors that have been proven to increase the risk of developing prostate cancer. Prostate cancer is significantly more likely in people over 50. Sixty to 70 years old is when the incidence peaks. Additional risk factors for advanced prostate cancer include smoking, being overweight, and certain dietary components [4].

Depending on the clinical context, multiple therapeutic approaches are needed for prostate cancer because it is a clinically and molecularly heterogeneous disease [5]. An important lineage-specific, carcinogenic transcriptional pathway in prostate cancer is triggered when the androgen steroid hormone interacts with the androgen receptor (AR). This fact has been used therapeutically for many years to treat de novo or recurrent metastatic disease following initial surgery or radiotherapy. Castration-resistant prostate cancer (CRPC) eventually develops in most patients who first respond to androgen deprivation therapies, even though these treatments stopped tumor growth initially [6].

Recently, new techniques were adopted for better understanding different pathways that promote tumor growth and progression, with the subsequent development of promising therapeutic strategies for the treatment of prostate cancer [5].

The Claudin family comprises at least 27 transmembrane proteins, which are important parts of tight junctions [7]. According to their sequence, they are classified into classic and non-classic types [8]. CLDNs 1–10 are examples of classic types, while CLDN16 and CLDN18 are examples of non-classic types [9]. CLDNs are expressed differentially in a variety of tissues, such as gastric, pancreatic, and lung tissues, and the formation of cancers in these tissues depends on their altered tissue function [10].

Claudin-18.1 is one of the two splicing variants of claudin-18, which is expressed in the lung, whereas claudin-18.2, which is only expressed in differentiated gastric mucosal cells, has very little expression in other healthy normal tissues [11]. Several gastrointestinal, ovarian, and non-small cell lung carcinomas expressed claudin-18.2 [12,13,14]. Recently created monoclonal antibodies (mAb) and chimeric antigen receptor modified T-cells (CAR-T) specific for claudin-18.2 have been used with encouraging outcomes for advanced pancreato-biliary tract and gastroesophageal tract malignancy management [15, 16].

The role of claudin-18 expression in prostatic carcinoma has not been evaluated.

The purpose of this study was to examine the expression of claudin-18 and the androgen receptor in prostatic carcinoma and to link these findings with other clinicopathological characteristics.

2 Methods

This retrospective observational study was carried out on 120 cases of prostatic lesions, including 100 cases of prostatic carcinoma and 20 cases of benign prostatic hyperplasia. The studied cases included archival formalin-fixed, paraffin-embedded tissue blocks processed during the years 2015–2023 by the Pathology Department of the Faculty of Medicine. All the specimens were obtained by prostatectomy. The research ethics committee of the faculty of medicine approved the study (NO:RC1-8-2023).

2.1 Histopathological analysis

Four-micron-thick sections of each tissue block were stained with hematoxylin and eosin (H&E) stain and revised by two independent pathologists to confirm the diagnosis. For grading, we applied the ISUP 2019 Gleason grading system [17] and staged using the American Joint Committee on Cancer (AJCC) 2017 TNM staging system [18].

2.2 Immunohistochemical analysis

Anti-claudin18 and anti-AR immunostaining were done for each case using the Avidin–Biotin complex technique according to the manufacturer's instructions. Antigen retrieval was performed using 10 mmol/1 citrate buffer (ph. 6.0). Primary antibodies, anti-claudin18 antibody (Invitrogen, Carlsbad, CA; polyclonal, 1:100), and Anti-AR antibody (Proteintech, Wuhan, China, monoclonal,1:100), were used. Primary antibodies were then added and incubated at room temperature overnight in a humidity chamber. Diaminobenzene (DAB) was used as a chromogen.

2.3 Positive control

Apparently normal gastric tissue was used as a positive control for CLDN18 [19], and for AR, apparently normal prostatic tissue was used [20]. Negative controls were prepared by omitting the primary antibody during staining and replacing it with saline or phosphate buffer.

2.4 Immunohistochemical assessment

2.4.1 Assessment of CLDN18 immunostaining

CLDN18 immunostaining was graded as negative, weakly positive, moderately positive, or strongly positive, according to Tanaka et al. [21].

2.4.2 Assessment of AR immunostaining

The immunostaining of AR was evaluated based on the percentage of nuclear positive staining cells; it is considered positive if (≥ 10%) of cells showed nuclear staining or negative (< 10%) [22].

2.5 Statistical analysis

The Statistical Package for the Social Sciences (SPSS) program, version 22 (SPSS Inc., Chicago, Illinois, USA), was used to analyze the results. Categorical data were presented as numbers and percentages using the Chi square test (χ2 test) or Fisher Exact test for their analysis. A P value < 0.05 was judged statistically significant, and a P value < 0.01 was considered statistically highly significant.

3 Results

3.1 Histopathological results

The current work included 120 cases of prostatic lesions, including 100 cases of prostatic carcinoma and 20 cases of benign prostatic hyperplasia. The mean age for prostatic carcinoma cases was 65 years. The clinicopathological data are detailed in Table 1.

Table 1 Distribution of different clinicopathological data of the studied prostatic carcinoma cases (N = 100)
Full size table

3.2 Immunohistochemical Results

3.2.1 Immunohistochemical assessment results of claudin-18 expression

Claudin-18 expression was detected in the membrane of 30 (30%) of studied carcinoma cases. Benign prostatic hyperplasia cases showed complete absence of staining (Fig. 1). There was a statistically significant correlation between claudin-18 expression & the studied groups (p = 0.046) (Table 2).

Fig. 1
figure 1

Representative samples of claudin-18 expression: A, B Claudin-18 expression in prostatic carcinoma with a low Gleason score (A, ABC X200; B, ABC X200), C Claudin-18 expression in prostatic carcinoma with a high Gleason score (ABC X400), and D Benign prostatic hyerplasia showing negative expression for claudin-18 (ABC X200)

Full size image
Table 2 Differences in Claudin-18 marker level between the studied groups
Full size table

3.2.2 Comparison of Claudin-18 expression with the clinic-pathological data of the studied prostatic carcinoma cases

A statistically significant correlation was found between claudin-18 expression and nodal metastasis (p = 0.000), tumor stage (p = 0.03), and Gleason grade group (p = 0.000), while no correlation was found with other variables (Table 3).

Table 3 Differences in claudin-18 marker level between the studied prostatic carcinoma cases regarding clinic-pathological data
Full size table

3.2.3 Immunohistochemical assessment results of androgen receptor

The expression of Androgen receptor was detected in the nucleus of 96 (96%) of studied carcinoma cases and in 18 (90%) cases of benign prostatic hyperplasia (Fig. 2). No statistically significant correlation was found between the studied groups and AR expression (p = 0.427) (Table 4).

Fig. 2
figure 2

Representative samples of androgen receptor (AR) expression: (A), AR expression in prostatic carcinoma with a low Gleason score (A, ABC X200), B, C AR expression in prostatic carcinoma with a high Gleason score ( B, ABC X400; C,ABC X200), and D Benign prostatic hyerplasia showing positive expression for AR (ABC X200)

Full size image
Table 4 Differences in AR marker level between the studied groups
Full size table

3.2.4 Comparison of androgen receptor expression with the clinic-pathological data of the studied prostatic carcinoma cases

A statistically significant correlation was found between AR expression and stage (p = 0.01), and Gleason grade group (p = 0.03), while no correlation was found with other variables (Table 5).

Table 5 Differences in AR marker level between the studied prostatic carcinoma patients regarding clinic-pathological data
Full size table

3.2.5 correlation between androgen recetor expression and claudin18 expression

A statistically significant correlation was found between androgen  receptor expression and claudin18 expression (p = 0.002) (Fig. 3).

Fig. 3
figure 3

Correlation between Caudin18 and androgen receptor markers expression

Full size image

4 Discussion

Claudin-18 is one of the claudin family that participates in tight junction strands in epithelial cells [11]. It was recognized in a variety of gastrointestinal, ovarian, and non-small cell lung carcinomas [13, 14, 23, 24]. The role of claudin-18 in prostatic carcinoma has not been clarified.

In the current study, benign prostatic tissue failed to stain with CLDN18, while in prostatic carcinoma, lower rates of CLDN18 positivity were recorded (30%), and this difference was of statistical significance (p = 0.046). The pattern of expression for CLDN18 from benign to malignant may suggest that CLDN18 could have a role in the genesis and development of prostatic carcinoma.

The claudin protein family has previously been examined, demonstrating a putative function for claudins in the development of prostatic cancer, which is compatible with our findings. Sheehan et al. [25] discovered that 41% of carcinomas had a higher level of claudin-1 than normal prostatic glands in 41% of tumors. Väre et al. [26] found claudin-1 positivity in 97% of cancers, with 43% showing strong immunostaining. According to Kind et al. [1], claudin-1 was overexpressed in a subset of prostate cancers. Landers et al. [27] observed that claudin-4 levels were increased in primary and metastatic prostate cancer. Claudin-8 was expressed in malignant tissues relative to normal ones, with a significant difference, according to Ashikari et al. [28]. Claudin-3 loss of expression was a prognostic marker in castration-resistant prostate cancer according to Orea et al. [29].

Analysis of CLDN18 expression was linked to lower Gleason grade score (p = 0.000). To what we know, this is the first study to evaluate this relationship, so we looked into the relationship of CLDN18 to tumor grades in additional organs. Pellino et al. [30] found similar associations in studied cases of advanced gastric and gastroesophageal junction adenocarcinomas. Analysis of CLDN18 by Kayikcioglu et al. [31] was high in lower grades of pancreatic ductal adenocarcinoma but did not reach a significant level.

Regarding the relationship of various claudins in prostatic carcinoma to the Gleason score, Landers et al. [27] reported that claudin-4 was expressed more in primary tumors with a Gleason score of 6 than higher Gleason scores. The low expression of claudin-1 was detected in Gleason scores of 7 or higher by Seo et al. [32]. Claudin-3 expression was found to be considerably lower in tissues of individuals with a Gleason score (≥ 8) by Orea et al. [29]. This runs in parallel with the hypothesis that tight junctions are destructed during tumorigenesis with disruption of cell adhesion molecules, contributing to cell invasiveness and metastases [33].

Claudin-18 expression was shown to be higher in advanced-stage tumors (p = 0.03). CLDN18 expression may play a role in tumor formation and progression. CLDN18 expression during tumor growth causes tight junctions to loosen, which may promote tumor cell motility and invasiveness [34].

To our awareness, no studies in the literature investigated this relation in prostatic carcinoma; however, studies in gastric and gastroesophageal junction adenocarcinomas by Pellino et al. [30] which matched our results, and in pancreatic ductal adenocarcinoma by Kayikcioglu et al. [31], where high CLDN18 expression was detected in advanced stages but did not reach a significant level.

The relation between claudin-4 expression and advanced stages of prostatic carcinoma was evaluated by Sheehan et al. [25] and reported a significant correlation.

Regarding the association between CLDN18 and lymph node metastasis, there was a statistically highly significant association (p = 0.000). This was compatible with Phattarataratip and Sappayatosok [35], who discovered that claudin-7 had a significant effect on oral squamous cell cancer. Claudin-7 alterations were linked to pathological grade, tumor size, and advanced TNM stage. Moreover, Wöll et al. [36] in pancreatic carcinoma detected higher CLDN18 expression in the lymph node metastasis-positive group. This observation could be beneficial in the development of novel targeted therapies for patients with lymph node metastases.

Androgen receptor plays pivotal roles in prostate cancer. AR has already been identified as the primary driver in the genesis and progression of prostate cancer [37].

In the present study, AR expression was detected in the nucleus of 96 (96%) of the studied carcinoma cases and in 18 (90%) cases of benign prostatic hyperplasia. No statistically significant difference was detected between AR expressions in both groups (p = 0.427). Similar results by Lai et al. [38] and Navaei et al. [39] were obtained. Lekshmy and Prema [40] reported AR expression in almost all prostate cancer cases as well.

Androgens greatly affect prostate cancer growth rates and development from preclinical to clinically relevant forms, which may be due to altered androgen metabolism.

We demonstrated that AR was correlated with the Gleason grade group (p = 0.03). In a study by Kwang et al. [41], the frequency of AR expression was higher in the group with the highest Gleason scores; similar results were obtained by Lai et al. [38]. Hashmi et al. [42] also found that low-grade tumors did not show strong AR expression, while patients in high-grade group showed strong AR expression. Hermien et al. [43] found high scores of AR expression, especially in the WHO grade groups III–V. Increased levels of AR may be especially important in driving tumor cell proliferation [44].

An association between AR and high stages of prostatic carcinoma (p = 0.01) was illustrated in our study. Similarly, Li et al. [45] documented that a high AR expression level was correlated with clinical stage. Heinlein and Chang [37] also found that high AR expression correlates with disease progression. In prostate cancer, the rate of proliferation exceeds that of cell death, where AR regulates the proliferation-death ratio of these cells. Therefore, the increased expression of AR is associated with the aggressiveness of prostate adenocarcinoma, including high grades and advanced stages.

The relationship between AR expression and LN metastasis was not found to be significant (p = 0.090). In a study conducted by Kwang et al. [41], the relationship was not statistically significant. On the contrary, Li et al. [45] documented that high AR expression levels were associated with the presence of LN metastasis.

We found a significant relationship between CLDN18 and AR expression (p = 0.002), indicating that CLDN18 may be regulated by AR. Meng et al. [46] showed that testosterone regulates the expression of CLDN8 in the prostate of mice, stating that castration resulted in lower levels of CLDN8 and a loss of the tight junction barrier, resulting in a loss of immunological privilege, inflammation, and an autoimmune reaction. It is generally understood that inflammatory responses and immunological processes play an important role in cancer genesis and progression. CLDN8-mediated androgen-dependent tight junction system barrier may preserve cellular homeostasis and cytoskeleton structure [46].

In addition, Ashikari et al. [28] showed that increased CLDN8 expression boosted prostate cancer cell growth and invasion and that AR regulates CLDN8. CLDN8 has been identified as an AR target gene in breast cancer cells, with AR being up-regulated following DHT therapy [47]. Zhang et al. [22] discovered a similar parallel relationship between CLDN8 and AR in breast cancer.

5 Conclusion

We demonstrated positive CLDN18 expression in 30% of the studied prostatic carcinoma cases, in contrast to benign prostatic tissue. This pattern of expression may suggest that CLDN18 could have a role in the development of prostatic carcinoma. The significant relationship between CLDN18 expression and Gleason grade group, tumor stage, and nodal metastasis suggests a possible role of CLDN18 in tumor aggressiveness and adverse patient outcomes. The significant correlation between CLDN18 and AR expression indicates that CLDN18 may be regulated by AR to enhance prostate cancer progression. CLDN18 could be a candidate therapeutic target for the treatment of prostatic carcinoma.