A study on the expression of EZH2, Bcl-2 and Ber-EP in BCC

317 A study on the expression of EZH2, Bcl-2 and Ber-EP in BCC Boyana Anatolieva1, Ivan Ivanov1, Dimitar Gospodinov1 1 Medical University of Pleven, Pleven, Bulgaria Corresponding author: Boyana Anatolieva (b.todor[email protected]) Copyright: © Boyana Anatolieva et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International – CC BY 4.0). Research Article Summary Basal cell carcinoma is the most common malignant tumour in humans. In cases with indistinct morphology on H&E-stained slides, immunohistochemistry may help distinguish basal cell carcinoma from other similar-appearing lesions. Our study aimed to investigate the expression of a marker panel comprising EZH2, Bcl-2, and Ber-EP4 in morphologically diagnosed, CK20-verified cutaneous basal cell carcinomas. Materials and methods: A cross-sectional study of 50 histologically confirmed cases of basal cell carcinoma was conducted. Immunohistochemical staining was performed using the following markers: EZH2, Bcl-2, Ber-EP4, and CK20. Due to the lack of a standardised method for evaluating markers, we adopted and modified the staining index (SI), which semi-quantitatively combines staining intensity and the percentage of positive cells. The results were systematised and interpreted using IBM SPSS. Results: All 50 examined tumours tested negative for CK20 (100%), thereby excluding mimics. All 50 tumours stained positive for EZH2 and Bcl-2 (100%), and only one stained negative for Ber-EP4 (98% positive). We found no association between histological type and EZH2 (p = 0.376), Bcl-2 (p = 0.376), and Ber-EP4 (p = 0.318), respectively, or their co-expression (p = 0.258). High co-expression of two of the three markers was observed in 33 of the 50 examined cases (66%), and a low co-expression in 4 cases (8%). Conclusion: The marker panel demonstrates co-expression of the three markers in the context of negative CK20 in over 90% of the cases. In challenging cases, it is important to consider clinical, morphological, and immunohistochemical features together. Key words: Basal cell carcinoma, Bcl-2, Ber-EP4, CK20, co-expression patterns , EZH2 Introduction Basal cell carcinoma (BCC) is the most common malignant tumour in humans worldwide (Chinem and Miot 2011). It is one of the so-called keratinocytic tumours (together with squamous cell carcinoma), whose incidence surpasses that of all other neoplasms (Albert and Weinstock 2003). The rising frequency and decreasing age of onset in recent years have made it a socially significant disease, associated with high morbidity and costs (Hu et al. 2022; Sendín-Martín et al. 2025). Clinically and histologically, basal cell carcinoma is a heterogeneous tumour that presents in different variants - nodular, cystic, Academic editor: Ivelina Yordanova Received: 14 October 2025 Accepted: 24 November 2025 Published: 18 December 2025 Citation: Anatolieva B, Ivanov I, Gospodinov D (2025) A study on the expression of EZH2, Bcl-2 and BerEP in BCC. Journal of Biomedical and Clinical Research 18: 317–330. https://doi.org/10.3897/jbcr.e174917 Journal of Biomedical and Clinical Research 18: 317–330 (2025) DOI: 10.3897/jbcr.e174917 Journal of Medical University of Pleven 318 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC morpheaform, infiltrative, micronodular, superficial, pigmented, and others. These different subtypes of BCC exhibit diverse biological behaviour, clinical and histological profiles, and prognoses (Raasch et al. 2006; Cameron et al. 2021). The basosquamous, infiltrative, morpheaform, and micronodular types are among the most aggressive (Peris et al. 2023). Currently, the gold standard in the histological diagnosis of BCC is hematoxylin and eosin (H&E) staining. However, this examination does not always reliably distinguish some types of BCC from other carcinomas, such as basosquamous carcinoma, whose treatment differs substantially from that of BCC due to its higher risk of metastasis (Karahan et al. 2006; Sunjaya et al. 2017). In practice, certain immunohistochemical (IHC) markers are used to support morphological diagnosis. The most commonly applied among them are Cytokeratin 20 (CK20), which should be negative in BCC, and positive markers, including Bcl-2, Ber-EP4, EZH2, among others (Crowson et al. 1996; Ramdial et al. 2000; Ramezani et al. 2016; Rao et al. 2016; Ozkanli 2023). Numerous studies have aimed to identify optimal diagnostic IHC panels for distinguishing BCC from other morphologically similar lesions. For example, (Ramezani et al. 2016) showed that positivity for Bcl-2 and CD10 differentiates BCC (from squamous cell carcinoma) with an accuracy of 88% and specificity of 100%. It is worth noting that the authors reported Bcl-2 positivity in 3.5% of squamous cell carcinomas. In their study, Bcl-2 positivity was observed in all BCCs examined. Other studies do not support the claim that all BCCs are positive for Bcl-2. Moreover, Crowson AN et al. have noted that Bcl-2 is expressed at varying levels, and this variability may be linked to tumour biological features or tumour progression. The pattern of positivity, however, is often weak and diffuse (Bcl-2), or incomplete and patchy with variable intensity -Ber-EP4 or EZH2 (Crowson et al. 1996; Ramdial et al. 2000; Rao et al. 2016; Ozkanli 2023). According to Ozkanli (2023), Ber-EP4 expression can differ between morphological variants and within different tumour areas of BCCs. There is relatively limited experience with using a panel of markers, including Bcl-2, Ber-EP4, EZH2, and CK20, for BCC, as well as analysing their expression and intensity both among BCCs in general and within their histological variants. Objective To investigate the expression of a marker panel including EZH2, Bcl-2, and Ber-EP4 in morphologically diagnosed and CK20-verified cutaneous basal cell carcinomas. Materials and methods Sample selection A cross-sectional study was conducted at the Dr Georgi Stranski University Hospital and Medical University - Pleven. Fifty histologically confirmed cases of basal cell carcinoma were randomly selected during 2023–2024 from the GAMMA Codemaster system of the University Hospital. The tumours were grouped by histological subtype. Clinical data, including sex and age at diagnosis, were collected. 319 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Histologic examination Corresponding H&E-stained slides were reviewed by a pathologist at the Department of Pathology. Histological types were confirmed. Immunohistochemistry At the Competence Centre of Medical University - Pleven, standard histological sections were prepared from paraffin blocks on adhesive slides. Immunohistochemical staining was performed using the studied markers: EZH2, Bcl-2, Ber-EP4, and CK20. Sections from formalin-fixed, paraffin-embedded tissue samples were cut at 3 μm thickness. After deparaffinisation and rehydration, staining was performed. Reagents and protocols from Biocare Medical (USA) were mainly used; the EZH2 antibody was from GenomeMe (Canada). Staining was carried out with antibodies CK20 (clone Ks20.8), Ber-EP4 (clone Ber-EP4), Bcl-2 (clone 100/5D), and EZH2 (clone IHC 770). Visualisation was done with the MACH 1 Universal HRP-polymer detector (Biocare Medical, USA). Controls and interpretation followed the manufacturer’s documentation. Expression grading The evaluation was performed by a histopathologist using a light microscope equipped with standard magnifications and a digital camera. Normal epidermis in each section served as an internal negative control. Only histological areas including the tumour were assessed. Due to the lack of a standardised method for evaluating the markers EZH2, Bcl-2, and Ber-EP4, we applied a modified method of our own. The H-score (0–300), used by some authors, would yield highly heterogeneous results with the small sample size of 50 cases. Therefore, we did not consider it suitable for this study (Petronilho et al. 2021). Instead, we adopted another method reported in the literature: the staining index (SI) – semi-quantitatively summing the staining intensity (0 – negative, 1 – weak, 2 – moderate, 3 – strong) and the percentage of positive cells (1 – <33%, 2 – 34–66%, 3 – >67%). We assessed the tumour areas with the highest percentage of positivity and recorded the staining intensity in these areas. A total score >4 was considered high expression of the marker, while a total score ≤4 was considered low expression (Van Rijn 1994; Bachmann et al. 2006; Puizina-Ivić et al. 2008; Rao et al. 2016; Mendez-Flores et al. 2022). The positivity pattern was recorded as predominantly peripheral or predominantly diffuse (modified from Ozkanli 2023). CK20 was recorded as either positive or negative. The obtained results were summarised and analysed. Statistical analysis The results were systematised and interpreted using IBM SPSS® Statistics version 26. To assess the presence of an association between categorical data, the chi-square test was used. Values of p < 0.05 were considered statistically significant. 320 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Ethical approval This study was approved by the Ethics Committee of Scientific Research (ECSR) at MU Pleven (Ref. No. 782/ECSR/14.06.24, Protocol No. 79) and conducted in accordance with the Declaration of Helsinki. Results СК20 All 50 examined tumours tested negative for CK20, thereby confirming that they are basal cell carcinomas, with no trichoblastomas or Merkel cell carcinomas (MCC) present (Fig. 1). Individual expression of EZH2, Bcl-2, and Ber-EP4 All 50 tumours were positive for EZH2. Low expression (≤4) was observed in 15 cases, while high expression (>4) was found in 35 cases. No negative samples were recorded (Fig. 2). In larger tumours, a less intensely stained centre of the tumour nests was observed compared to the more intensely stained periphery. All 50 tumours were positive for Bcl-2, with no negative samples. Tumour expression of Bcl-2 was low (≤4) in 14 cases, and high (>4) in 36 cases (Fig. 3). Again, in larger tumours, a less intensely stained centre of the tumour nests was seen in contrast to the more intensely stained periphery. Figure 1. Cytokeratin 20 (CK20) expression in BCC – Superficial BCC (*) with no positive cells for CK20 staining (original magnification ×100). Normal epidermis is also CK20-negative (internal control). 321 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Figure 2. EZH2 expression in BCC – The same BCC (*), with apparent positive staining with strong intensity (6/6) for EZH2 in the majority of the tumour cells (original magnification ×100). Figure 3. Bcl-2 expression in BCC – The same BCC (*) with intense homogenous staining for Bcl-2 (6/6) in the majority of the tumour cells (original magnification ×100). 322 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Of all samples (n = 50), 49 were positive for Ber-EP4, and one was negative. Low expression (≤4) was found in 14 carcinomas. In contrast, high expression (>4) was observed in 35 cases (Fig. 4). Here, too, in larger tumours, a less intensely stained centre of the tumour nests was seen compared to the more intensely stained periphery. In smaller tumours, the staining was more homogeneous, without the aforementioned variation in staining for the investigated markers. This trend, however, was not absolute, and it is difficult to precisely evaluate in a small series of cases. Coexpression of EZH2, Bcl-2 and Ber-EP4 High co-expression of two of the three studied markers was observed in 33 of the 50 examined cases. At the same time, low co-expression was observed in only 4 cases. Details are presented in Table 1. The majority of tumours showed high expression of all three or at least two of the markers with high intensity in a large percentage of tumour cells (score >4). Table 1. Co-expression of EZH2, Bcl-2, and Ber-EP4. Combined expression of the three markers Number of tumours EZH2>4, BCL2>4, BEReP4>4 18 EZH2<4, BCL2>4, BEReP4>4 2 EZH2<4, BCL2<4, BEReP4>4 2 EZH2<4, BCL2>4, BEReP4<4 7 EZH2>4, BCL2>4, BEReP4<4 9 EZH2<4, BCL2<4, BEReP4<4 4 EZH2>4, BCL2<4, BEReP4>4 4 EZH2>4, BCL2<4, BEReP4<4 4 Figure 4. Ber-EP4 expression in BCC – Superficial multifocal BCC (*) with positive (6/6) staining for Ber-EP4 with strong intensity diffusely in the whole tumour (original magnification ×100). Normal epidermis is Ber-EP4 negative (internal control). 323 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Individual expression of EZH2, Bcl-2 and Ber-EP4 in the different morphological variants of BCC After comparing EZH2 expression across the different morphological variants of BCC, we found no association between histotype and EZH2 (ChiSquare = 7.529, Df = 7; p = 0.376) (see Table 2). Here, too, the distribution was not statistically reliable enough. Table 2. EZH2 staining in the different histologic subtypes of BCC. Histotype EZH2_score </=4 – weak expression >4 – strong expression superficial 3 5 nodular/solid (= pigmented, keratotic) 7 19 micronodular 1 1 infiltrative (sclerosing, morpheaform) 2 5 basosquamous 0 1 cystic 0 1 mixed 2 0 NOS (not otherwise specified) 0 3 Table 3. BCL-2 staining in the different histologic subtypes of BCC. Histotype Bcl-2_score </=4 – weak expression >4 – strong expression superficial 2 6 nodular / solid (= pigmented, keratotic) 8 18 micronodular 0 2 infiltrative (sclerosing, morpheaform) 1 6 basosquamous 0 1 cystic 1 1 mixed 0 2 NOS (not otherwise specified) 2 1 70% of the examined tumours (35 out of 50) showed strong expression (>4) of EZH2 across the different morphological subtypes. When comparing Bcl-2 expression across the different histological types of BCC, no association was found between histotype and Bcl-2 (Chi-Square = 7.528, Df = 7; p = 0.376) (see Table 3). Again, the statistical result is not reliable enough due to the small number of cases. Of the 50 tumours, 36 (72%) showed intense Bcl-2 staining, and this was observed across the majority of histotypes. After comparing Ber-EP4 expression across the different morphological variants of BCC, no association was found between histotype and Ber-EP4 (Chi-Square = 8.168, Df = 7; p = 0.318) (see Table 4). The small sample size (n = 50) and its distribution render the statistical results insufficiently reliable. Overall, the tumour type does not determine the intensity of Ber-EP4 expression. 26 of 50 tumours showed strong expression of Ber-EP4 (52%), and only one stained negative. 324 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC Co-expression of the markers in the different morphological variants of BCC The combination of EZH2, Bcl-2, and Ber-EP4, along with their levels of expression, did not show a statistically significant association with histological distribution (Pearson Chi-Square = 54.994; Df = 49; p = 0.258). It is notable that, most often, all three markers, or at least two of them, showed intense staining (see Table 5). There was no clear correlation between the expression of the three markers and histologic subtype. It is noteworthy, however, that, in most cases, either all three markers or at least two of them showed intense staining. Discussion Morphologically, basal cell carcinomas are distinct tumours composed of atypical basaloid cells, with some variants exhibiting a clearly palisaded arrangement of tumour cells along the periphery of the tumour nests. Often, the tumours show a fibro-myxoid stroma, and occasionally contain melanin pigment. The differential diagnosis for these tumours can be quite broad, including Merkel cell carcinoma, sebaceous carcinoma, basaloid follicular neoplasms, and basaloid squamous cell carcinomas. Frequently, markers such as Bcl-2, BerEP-4, CK20, CD10, p40, CK5/6, p63, AR, EMA, CEA, and others are used depending on the case (Liu et al. 2024; Ramezani et al. 2016; WHO Classification of Skin Tumours 2023). According to Sari Aslani et al. (2013), CD10 expression by tumour cells in basal cell carcinoma was observed in 42 of 55 cases; meanwhile, in trichoepithelioma, Table 4. Ber-EP4 staining in the different histologic subtypes of BCC. Histotype Ber-EP4_score </=4 – weak expression >4 – strong expression superficial 4 4 nodular / solid (= pigmented, keratotic) 11 15 micronodular 2 0 infiltrative (sclerosing, morpheaform) 2 5 basosquamous 1 0 cystic 0 1 mixed 2 0 NOS (not otherwise specified) 2 1 Table 5. Co-expression of EZH2, Bcl-2, and Ber-EP4 and histologic subtype of BCC. BCC histologic subtype EZH2>4; Bcl2>4; BerEP4>4 EZH2<4; Bcl2>4; BerEP4>4 EZH2<4; Bcl2<4; BerEP4>4 EZH2<4; Bcl2>4; BerEP4<4 EZH2>4; Bcl2>4; BerEP4<4 EZH2<4; Bcl2<4; BerEP4<4 EZH2>4; Bcl2<4; BerEP4>4 EZH2>4; Bcl2<4; BerEP4<4 Superficial 3 1 2 2 Nodular 10 1 2 3 4 1 2 3 Micronodular 1 1 Infiltrating 5 1 1 Basosquamous 1 Nodulocystic 1 Mixed 2 NOS 1 1 1 325 JBCR 18: 317–330 (2025), DOI: 10.3897/jbcr.e174917 Boyana Anatolieva et al.: A study on the expression of EZH2, Bcl-2, and Ber-EP in BCC no expression was observed in the neoplastic epithelial cell population. The mentioned fact suggests that positivity for CD10 may have real clinical value. The use of negative markers, such as EMA, for identifying squamous carcinomas among basal cellasaloid epidermal tumours of the skin is generally useful. However, areas of squamous differentiation are positive and may be confusing to the inexperienced in dermatopathology (Hussein et al. 2022). Furthermore, not all squamous cell carcinomas are EMA-positive, which may further complicate interpretation (Beer et al. 2000). The androgen receptor (AR) can be added to the diagnostic panel of markers to distinguish basal cell carcinoma, as it is generally positive. In contrast, trichoblastoma is usually negative (Izikson et al. 2005). When used in a panel in the proper context, EMA, AR, and CD10 may provide diagnostic benefit. The negativity or positivity of tumours for Cytokeratin 20 (CK20) is an important aspect in the immunohistochemical diagnostic clarification of basaloid skin tumours. CK20 is part of the cytoskeleton of epithelial cells. It is used to distinguish trichoblastomas from basal cell carcinomas in difficult-to-assess cases. Basal cell carcinoma expresses a cytokeratin profile similar to that of follicular germinative cells, characterised by CK5/6 and CK14 expression and the absence of CK20. In some trichoblastomas, a small number of Merkel cells are positive for CK20, scattered and few in number, unlike in MCC, where diffuse perinuclear positivity is observed. Most MCCs are CK20-positive, while BCCs are CK20-negative (Yang et al. 2004). EZH2 and Ki67 have been identified as biomarkers associated with aggressive BCC types, suggesting that EZH2 may represent a potential therapeutic target. Further research is needed to determine its predictive and prognostic value regarding treatment response. The results of the present study highlighted several important aspects of the immunohistochemical expression of EZH2, Bcl-2, and Ber-EP4 in BCC. First, no tumour was found to be simultaneously negative for all three markers, with only one case being entirely negative for Ber-EP4. The pattern of positivity and the staining intensity (although presented here in a more summarised form) support the data reported in the literature (Ozkanli, 2023). Nevertheless, the variability in expression among individual tumours limits their use as standalone markers, thus underscoring the need for a standardised diagnostic panel for BCC. EZH2 (Enhancer of Zeste Homolog 2) is a key enzyme that regulates gene expression by repressing tumour suppressor genes, thereby facilitating tumour growth. The expression of EZH2 in basal cell carcinomas has been described by other authors and is considered a marker of tumour aggressiveness (Rao et al. 2016, 2018). EZH2 is expressed in a variety of skin tumours, and its positivity as a standalone IHC marker is not sufficient to determine histogenesis (Xie et al. 2014). Bcl-2 (B-cell lymphoma 2) is a mitochondrial proto-oncogene that blocks apoptosis in the pro-B lymphocyte cell line (Hockenbery et al. 1990). Like CK20, Bcl-2 distinguishes basal cell carcinomas (in which it is positive) from trichoblastomas, in which Bcl-2 is usually negative (Rijn 1994). Several other skin neoplasms can also express Bcl-2. Notably, the pattern of Bcl-2 expression varies among the different variants of basal cell carcinoma (Puizina-Ivić et al. 2008). Bcl-2 is expressed in basal cell carcinomas predominantly with moderate and weak intensity over a large proportion of the tumor volume (diffusely).