Immunoexpression of Claudin-4 In Non-Small Cell Lung Carcinomas

Abstract

Non-small cell lung carcinomas (NSCLC) are the most common histological forms of lung cancer, lesions that by their incidence and associated mortality constitute a health problem worldwide. Alteration of the expression of claudins with a role in the stability of cell adhesion constitutes one of the complex biomolecular mechanisms involved in tumor initiation and progression. In this study, Claudin-4 immunoexpression was investigated in 52 cases of NSCLC in relation to epidemiological and histopathological prognostic parameters of the lesions. Although the staining scores were higher in patients in younger age groups, in women, in smokers and in squamous cell carcinomas (SCC) and large cell carcinoma (LLC) compared to adenocarcinomas (ADK), the aspects were not statistically significant. We found significantly higher differences in micropapillary, cribriform and solid ADK compared to other histological subtypes, in well and poorly differentiated lesions (G1/G3), with vascular invasion and in advanced tumor stages. Reaction analysis indicated translocation of Claudin-4 signals from the membrane level to the cytoplasm and nucleus in aggressive histological subtypes. The study indicated differences in Claudin-4 expression, which can be used to identify aggressive NSCLC and to stratify patients for specific therapy.

Introduction

Lung cancer is one of the most common and lethal types of cancer worldwide, ranking second in terms of global incidence of malignant neoplasms and first as the leading cause of cancer death [1].

Lung cancer is a heterogeneous disease, which refers to several histopathological types, the most common being non-small cell lung carcinomas (NSCLC), which include squamous cell carcinoma (SCC), adenocarcinoma (ADK) and large cell carcinomas (LCC), each of these types having distinct characteristics and requiring personalized therapeutic approaches [2, 3].

Lung cancer classification is based on several factors, such as cellular morphology, tumor location, and molecular characteristics [4].

Among these, the expression of molecules belonging to cell adhesion systems and ensuring epithelial stability have been analyzed in the hope of identifying effective therapeutic targets and criteria for identifying aggressive lesions.

In this context, claudin expression seems to provide essential information about tumor behavior and metastatic potential, factors that influence both prognosis and therapeutic options [1, 5].

However, there are relatively few studies that have analyzed the expression and localization of claudins in lung tumors.

Claudin-4 is a transmembrane protein belonging to the claudin family, essential components of tight junctions in epithelial cells, which play a crucial role in maintaining the integrity of the cell barrier, ensuring selective permeability between tissue compartments and maintaining cell polarity [6, 7].

Current evidence indicates an association between levels of Claudin-4 overexpression and clinicopathological features of NSCLC, such as aggressive tumor behavior, metastatic potential and sensitivity to various oncological therapies [6, 8].

Despite advances in the diagnosis and treatment of NSCLC, the long-term survival rate remains relatively low, which highlights the need for new diagnostic and therapeutic strategies [1, 9].

Detailed semiquantitative and localization characterization of Claudin-4 expression in NSCLC may contribute to improving patient stratification criteria for therapy and identifying lesions with poor prognosis.

In this context, we analyzed Claudin-4 immunoexpression in relation to epidemiological and histopathological prognostic parameters of NSCLC.

Materials and Methods

In this study, 52 cases of non-small cell lung carcinomas (NSCLC), which were diagnosed in the Craiova County Emergency Clinical Hospital, were analyzed.

The biological material was collected over eight years (2015-2022) and was represented by partial or total lung resection specimens, which were fixed in 10% buffered formaldehyde and classically processed within the Pathology Department of the hospital.

After standard staining with Hematoxylin Eosin, the tissue fragments were investigated for diagnosis and reanalyzed according to the current World Health Organization (WHO) criteria for lung cancers [4].

In addition to establishing histological prognostic parameters represented by tumor type and subtype, tumor grade and stage, and vascular invasion, age groups, gender, and smoking status were also analyzed, all in relation to the immunoexpression of Claudin-4, a marker involved in the intercellular adhesion system [6].

Immunohistochemical (IHC) analysis was performed on serial sections from paraffin blocks, which were deparaffinized in xylene, hydrated with alcohol and washed in distilled water, subsequently subjected to boiling in citrate buffer solution pH 6 for 20 minutes for antigen retrieval.

In this study, the rabbit antihuman Claudin-4 polyclonal antibody (Invitrogen, code PA1-37471) was used, in a dilution of 1:150, which was positive on the external control section represented by breast ductal carcinoma.

After application of the secondary antibody from the EnVision™ FLEX+ System, the reactions were visualized with the chromogen 3,3’-Diaminobenzidine (DAB).

Washing of the sections after each working step and stopping the reactions was performed in PBS (phosphate-buffered saline) solution, which was followed by dehydration and clarification in alcohol and xylene and finally permanent mounting.

IHC reactions were quantified based on scores, taking into account the number of labeled cells reported as a percentage average on 10 MFs (microscopic fields) of 40x, representative for each case (1-25%- score 1, 26-50%- score 2, >50%- score 3) and the intensity of the reactions (weak-score 1, moderate-score 2, high-score 3).

The two types of scores were multiplied and final staining scores (FSS) were thus obtained, which had values ranging from 1-9.

In this study, all membranous, cytoplasmic and nuclear IHC signals of Claudin-4 were quantified.

In addition to quantification, a descriptive analysis of the localization of IHC staining in the tumor areas was also made.

For statistical analysis, low (FSS 1-4) and high (FSS 6-9) scores were investigated in relation to the epidiemiological and histopathological parameters using the statistical package for the social sciences (SPSS) 12 comparison tests (chi square-χ2 and Fisher).

Values were considered significant if p<0.05.

In this study, approved by the Local Ethics Committee, the ethical norms in research were respected (No.194/14.10.2022).

Results

Patients diagnosed with non-small cell lung carcinomas (NSCLC) belonged to a wide age range between the 5th and 8th decades of life, and with a mean age of 65.8 (±8.8) years, with most cases being in the 70-79 years range (23 cases).

The 52 cases were mostly males (40 cases) and belonged mostly to the smoking category (47 cases).

The majority of cases were represented by adenocarcinomas (ADK) and non-keratinized squamous carcinomas (NKSCC) (27 and 22 cases), with a relatively small number of large cell carcinomas (LCC) (3 cases) being identified.

There were numerous tumor subtypes that were dominated by the acinar (tubular) type (8 cases) in the case of ADK and NKSCC (17 cases) in the case of SCC.

Additionally, papillary, cribriform and solid ADK aspects were present (in 4 cases each), but also micropapillary (3 cases), lepidic and mucinous (in 2 cases each), as well as basaloid SCC (BSCC) (5 cases).

The majority of lesions were high grade, moderate (G2, 22 cases) and poorly differentiated (G3, 28 cases), and were in stage III/IV (16/14 cases), while vascular invasion was observed in just under half of the cases (20 cases).

Claudin-4 immunoexpression was identified in all analyzed cases, at the tumor level with membrane, cytoplasmic and/or nuclear localization.

In the case of normal structures (bronchioles, alveolar epithelium), the staining was complete membrane and apical cytoplasmic with a uniform appearance.

In addition, the reactions were also present at the level of some stromal elements represented by plasma cells, lymphocytes, fibroblasts, macrophages.

The analysis of Claudin 4 expression in relation to the analyzed parameters showed some differences.

Thus, regarding the age groups, for cases <50 years the number of labeled cells was 67.5%, with moderate/high intensity and FSS of 7.5, while for the intervals of 50-59 and 60-69 years, 51.6±19.5 and 51.1±20.1 cells were positive, with moderate/high intensity reaction and FSS of 5.8 and 6.0, respectively.

In comparison, for the group 70-79 years 46.5±22 cells were stained, the reactions having variable intensity and a final score of 4.7 (Table 1).

Table 1.

Claudin-4 immunoexpression depending on investigated parameters

Parameter/ p value		Claudin-4 (FSS)
Age (years)	<50	7.5
	50-59	5.8
	60-69	6.0
	70-79	4.7
	p value (χ2 test)	0.509
Gender	M	5.4
	F	5.7
	p value (Fisher’s test)	0.741
Smoking status	Smoker	5.6
	Non-smoker	4.0
	p value (Fisher’s test)	0.383
Tumor type	Squamous cell carcinoma	6.1
	Adenocarcinoma	4.9
	Large cell carcinoma	5.6
	p value (χ2 test)	0.211
Tumor subtype	Acinar	2.2
	Papillary	2.2
	Lepidic	6.0
	Mucinous	6.5
	Cribriform	7.5
	Micropapillary	7.0
	Solid	7.7
	Basaloid squamous cell carcinoma	6.4
	Non-keratinizing squamous cell carcinoma	6.1
	Large cell carcinoma	5.6
	p value (χ2 test)	0.002
Tumor grade	G1 (well differentiated)	6.0
	G2 (moderate differentiated)	4.4
	G3 (poorly differentiated)	6.3
	p value (χ2 test)	0.011
Vascular invasion	Absent	4.5
	Present	7.0
	p value (Fisher’s test)	<0.001
Tumor stage	I	3.6
	II	3.9
	III	6.6
	IV	6.8
	p value (χ2 test)	0.001

Although FSS seemed to decrease with increasing age, the aspects described were not statistically significant (Table 1).

In relation to gender, reactions for males compared to females were present in 49.8±21.5 vs. 49.6±18.7 tumor cells, with variable intensity and with FSS of 5.4 and 5.7, without statistical differences (Table 1).

Also, smoking status was associated with higher Claudin-4 reactions compared to non-smokers, the mean number of labeled cells being 50.6±21.1 vs. 42±16, with variable intensity and mean scores of 5.6 and 4, but without statistical differences (Table 1).

Regarding histopathological parameters, for the tumor type, the highest reactions were identified in the case of SCC, which presented 55.6±16.4 labeled cells, with variable intensity but predominantly moderate/high and mean FSS of 6.1.

In the case of ADK and LCC, the number of labeled cells was 45±23.2 and 50±20, also with moderate/high intensity and mean FSS of 4.9 and 5.6.

In relation to the tumor subtype, we found variations in both reaction scores and signal localization.

The weakest reactions were present in the case of acinar and papillary ADK, which presented 23.7±6.9 and 18.7±10.3 labeled cells, with variable intensity, respectively exclusively weak/moderate and FSS of 2.2 in both cases.

In these cases, the reactions were predominantly incomplete membranous, in addition in the case of tubular ADK focal cytoplasmic staining being also present (Table 1, Figure 1A-B).

Figure 1.

Claudin-4 immunoexpression, NSCLC. A. Acinar ADK, x400; B. Papillary ADK, x400; C. Lepidic ADK, x200; D. Mucinous ADK, x200; E. Cribriform ADK, x200; F. Micropapillary ADK, x200; G. Solid ADK, x200; H. BSCC, x200; I. NKSCC, x200.

In the case of lepidic and mucinous ADK the number of labeled cells was 60% and 50%, with moderate/high intensity and with average scores of 6.0 and 6.5, the reactions being membranous and cytoplasmic (Table 1, Figure 1C-D)

The highest Claudin-4 reactions were identified in cribriform, micropapillary and solid ADK, in which the number of labeled cells was 68.7±8.5, 71.6±12.5 and 60±10.8, with moderate/high intensity and mean FSS of 7.5, 7.0 and 7.7 (Table 1, Figure 1E-G).

In these cases, the reactions were predominantly cytoplasmic and focally incompletely membranous.

In SCC, the mean number of positive cells was 58±14.4 in BSCC with high intensity and 55±17.3 in NKSCC with variable intensity, predominantly moderate/high; in BSCC, the labeling was membranous and cytoplasmic, and in NKSCC, it was predominantly nuclear and cytoplasmic (Table 1, Figure 1H-I).

Overall, we found the translocation of Claudin-4 expression from a predominantly membrane level (tubular and papillary ADK) to a membrane/cytoplasmic level (lepidic and mucinous ADK, BSCC), to a predominantly cytoplasmic level (cribriform, micropapillary and solid ADK) and to a nuclear/cytoplasmic level (NKSCC).

Claudin-4 reactions were maximal in well and poorly differentiated NSCLC (G1/G3) in which 60% and 56±18.6 labeled cells were identified, with moderate/high intensity and FSS of 6.0 and 6.3, compared to moderately differentiated tumors (G2), in which the number of stained cells was 40.9±21, with predominantly moderate intensity and FSS of 4.4 (Table 1).

We believe that Claudin-4 expression was maximal in the case of membrane/cytoplasmic immunostaining that was comparable to expression on normal structures in papillary, tubular, mucinous, lepidic NSCLC and in the case of high-grade tumors, in which expression had already translocated to the cytoplasm/nucleus in the case of micropapillary, cribriform, solid or NKSCC forms.

Regarding Claudin expression in relation to vascular invasion, the number of labeled cells was higher in the presence compared to the absence of the parameter, respectively 64.2±13.5 vs. 40.7±19.4, the intensity of the reactions being predominantly moderate/high and average FSS of 7.0 and 4.5 (Table 1).

Claudin-4 values were higher in stages III/IV, respectively 60.7±17.7/ 60.9±17.4 labeled cells, compared to stages I/II where we found 37±22.5/ 34.1±12.7 labeled cells.

In all cases the intensity of reactions was predominantly moderate/high and the FSS values were for stages III/IV 6.6/6.8, and for I/II 3.6/3.9 (Table 1).

Statistical analysis of Claudin-4 expression in relation to the analyzed histopathological parameters indicated non-significant differences in relation to the tumor type (χ2 test, p=0.211) and significantly higher values in micropapillary, cribriform and solid ADK, compared to the other forms of NSCLC (χ2 test, p=0.002) (Table 1, Figure 2A-B).

Figure 2.

A. Claudin-4 FSS according to tumor type; B. Claudin-4 FSS according to tumor subtype; C. Claudin-4 FSS according to tumor grade; D. Claudin-4 FSS according to tumor stage.

Also, Claudin-4 values were significantly higher in NSCLC G1/G3 (χ2 test, p=0.011), in lesions that presented vascular invasion (Fisher's test, p<0.001) and in tumor stages III/IV (χ2 test, p<0.001) (Table 1, Figure 2C-D).

Discussion

Lung cancer has a specific epidemiological profile, representing the main cause of cancer morbidity and mortality in men and ranking third in incidence in women, after breast and colorectal cancer, and second in mortality, after breast cancer [1, 2].

Also, the incidence of lung cancer varies depending on geographical factors, but also on the lifestyle of the population, smoking being the most important known risk factor [1, 9].

In our study, the average age of diagnosis of NSCLC was in the 7th decade of life, with most cases being associated with men (76.9%) and smokers (90.3%).

Also, most cases belonged to ADK, high-grade, many with vascular invasion and in advanced stages, which is consistent with the histopathological profile of NSCLC [4, 10].

In this study, we did not find differences in Claudin-4 expression in relation to epidemiological parameters.

Regarding Claudin-4 expression in relation to histological parameters, there are some few and controversial data.

Thus, some authors have indicated that Claudin-4 expression is associated with disruption of tight junction integrity in cancer cells, which may contribute to tumor progression and invasion and although it is not directly associated with survival, there is a differential expression in the types of NSCLC [6].

The study conducted by Naso JR et al. that analyzed the immunoexpression of Claudin-4 in lung tumors revealed the presence of the protein in 82% of cases of ADK, SCC and LCC, without identifying significant differences between the different histopathological types of NSCLC [11], results that are also supported by the study conducted by Paschoud S et al. [12].

In our study, Claudin-4 was expressed in all cases and although the reaction scores were higher in SCC and LCC compared to ADK, the results were without statistical significance.

In the study conducted by Hruaii V et al. it was demonstrated that the molecule is an immunohistochemical marker with high sensitivity and specificity for malignant tumor cells, being expressed in all cases of metastatic adenocarcinomas of primary pulmonary origin [13].

Research by Chae MC et al. analyzed the role of Claudin-4 as a prognostic factor in lung adenocarcinoma, finding that increased expression is a negative prognostic factor, associated with a higher risk of relapse, even in early stages of the disease, suggesting that patients with increased expression may benefit from adjuvant treatments, such as chemotherapy, to improve long-term survival [14].

However, in our study we found different patterns of Claudin-4 positivity in relation to ADK and SCC subtypes.

Thus, if in the case of tubular and papillary ADK the expression was predominantly membranous, similar to the expression in normal structures, in the case of lepidic and mucinous ADK and in the case of basaloid SCC (BSCC) the expression was membranous and cytoplasmic, compared to cribriform, micropapillary and solid ADK in which the expression was predominantly cytoplasmic and non-keratinized SCC (NKSCC) with nuclear/cytoplasmic expression.

As such, we consider that the localization of Claudin-4 expression is at least as important as the quantity of the staining, aspects that are very little addressed in the specialized literature.

In our study, Claudin-4 expression presented significant differences in relation to tumor subtypes.

The study conducted by Patel A et al. demonstrated the utility of Claudin-4 in diagnosing pulmonary ADK and could improve risk stratification among patients with NSCLC [15].

The authors observed that Claudin-4 expression was correlated with advanced disease stage and increased risk of metastasis, making this protein a valuable indicator in determining the prognosis of patients with lung adenocarcinoma [15].

Research on Claudin-4 expression revealed that immunoexpression was not associated with tumor stage or degree of differentiation [12].

In our study, there were significantly higher differences in Claudin-4 scores in G1/G3 NSCLC, with vascular invasion and in advanced stages.

In various types of cancer, such as ovarian, colorectal, pancreatic and gastric, Claudin-4 is often overexpressed, which may promote tumor progression and invasiveness [7].

In breast cancer, its expression is associated with hormone receptor-negative tumors and high mitotic index, and in prostate cancer, increased Claudin-4 immunoexpression is correlated with advanced tumor stage [8, 16].

Recent data from the literature highlight the importance of claudins in the progression and metastasis of NSCLC, highlighting their role in maintaining the epithelial barrier and influencing the tumor environment [5].

In particular, Claudin-1 has been associated with tumor aggressiveness and the presence of metastases, while Claudins 3, 4 and 7 have shown variable expression patterns depending on the stage and histological type of the tumor [5].

The multitude of staining patterns identified in this study in relation to NSCLC types and subtypes, the translocation of its expression from the membrane level to the nuclear level and the relationship with histological parameters of lesion aggressiveness, designate Claudin-4 as a marker with potential for use as a molecular target and for lesion classification, with future studies being necessary for confirmation.

Conclusion

There are differences in Claudin-4 localization according to tumor subtype, with decreased membrane expression and acquisition of cytoplasmic and nuclear staining in micropapillary, cribriform, solid, and nonkeratinized squamous tumors, which represented the majority of high-grade and advanced NSCLC.

The preservation of membrane expression of Claudin-4 appears to be a protective factor in NSCLC.

Determination of the localization of Claudin-4 expression, along with quantification of staining, can be used to improve the criteria for identifying aggressive NSCLC and may contribute to improving prognosis.

Conflict of interests

The authors have no conflict of interest to declare.