Keratin 24 Predicts Poor Prognosis in Breast Cancer Open Access

Breast cancer (BC) exhibits diverse histological, molecular, and clinical characteristics. Therefore, it is still challenging to comprehend BC biology and define an appropriate treatment approach. Exploring the biological process and molecular mechanisms underlying BC progression is crucial for developing better treatment strategies and eventually improving patient outcomes.

Eukaryotic cells contain three main kinds of cytoskeletal filaments, namely, intermediate filaments (IFs), microfilaments, and microtubules that form structural framework within the cellular cytoplasm providing physical stability and protecting cell surfaces during cell stress [1] in addition to other cellular functions [2]. KRTs, which are IF proteins [3], are the main cytoskeletal proteins of epithelial cells. KRT proteins can form IFs either inside or outside epithelial cells by producing polymeric filaments via the division of type I (K9–K28, K71–K86) and type II (K1–K8, K31–K40) KRT proteins [3, 4]. These proteins respond strongly to their cellular environment, affecting their expression when confronted with cell stress. Additionally, KRTs function in several biological processes, including cell migration, proliferation, adhesion, epithelial polarity, and inflammatory regulation of protein catabolism in multicellular compartments between the nucleus and extracellular matrix [5].

Keratin 24 (KRT24) is a member of the type I KRT family composed of 525 amino acids [3, 6], bears high similarity to the type I KRTs, and displays a unique expression profile [3]. KRT24 is strongly expressed in keratinocytes, colon, spleen, and placenta [6]. Previous research has suggested that KRT24 potentially acts as a oncogene in the early stages of colorectal cancer [6, 7]. However, the biological roles of KRT24 in BC are still unknown. Thus, this study aimed to evaluate the expression of KRT24 in BC and determine its significant association with clinicopathological variables and patient outcomes as a potential prognostic factor and therapeutic target in BC.

The Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) data set (n = 1,980) was used as a discovery cohort to investigate the prognostic value of mRNA expression of KRT24 at the genomic level [8]. Additionally, the Kaplan-Meier (KM) plotter (n = 3,951) online data set (https://kmplot.com/analysis/) [9] and Breast Cancer Gene-Expression Miner v5.1 (bc-GenExMiner v5.1) database (n = 6,240 – all oestrogen receptor [ER] and all progesterone receptor [PR]) [10] were used to validate the prognostic value of KRT24 for mRNA expression.

Differential Gene Expression and Pathway Analysis. Differential gene expression analysis was conducted on primary tumour data from The Cancer Genome Atlas (TCGA) data set (n = 854) using DESeq2 [11]. Samples were dichotomised by KRT24 median mRNA expression into high and low, and significantly differentially expressed genes (DEGs) were identified (fold change ±2 and FDR-corrected p value <0.05). The WEB-based GEne SeT AnaLysis Toolkit (WebGestalt) was used to investigate overrepresentation analysis and the enrichment of KEGG pathways with up- and downregulated DEGs [12], and the Nucleic Acid SeQuence Analysis Resource (NASQAR), a web-based platform for high-throughput sequencing data analysis and visualisation, was used to visualise the data [13].

A large well-characterised BC cohort with (n = 951 cases) was used with extensive clinical information comprising patients who presented at Nottingham City Hospital, Nottingham, UK [14]. However, only informative tumour microarray (TMA) cores that harbour >15% of cancer cells in the core were considered to assess KRT24 protein expression using immunohistochemistry (IHC). Therefore, 832 cases were identified to perform cytoplasmic and membranous expressions [15‒17]. For molecular classification of the cohort, the ER (≥1% positive), PR (≥1% positive), human epidermal growth factor receptor (HER) 2, and Ki67 were used as previously reported [18‒21]. The BC molecular subtypes based on this IHC profile were categorised as follows: luminal A-like BC: ER+, PR+, HER2−, with low proliferation (Ki67 <10%); luminal B-like BC: ER+, PR+, HER2−, with high proliferation (Ki67 ≥10%), or ER+, PR+, HER2+; HER2-positive BC: HER2+ and ER−, and triple-negative BC: ER−, PR−, and HER2− [22].

Both ER status and the Nottingham Prognostic Index (NPI) were applied to classify patients into clinically relevant groups for treatment decision-making. Patients with poor prognostic NPI scores (>3.4) and positive ER status received endocrine treatment (295/832; 35%), while those with negative ER status received chemotherapy (172/832; 20%) (classical cyclophosphamide, methotrexate and 5-fluorouracil). Patients with good prognostic NPI scores (≤3.4) received no adjuvant therapy. No anti-human epidermal growth factor receptor 2 (HER2) target therapy or neoadjuvant therapy was used in this study. The clinicopathological features of this study cohort series were summarised previously [14, 16].

To explore the correlation between KRT24 expression and other related biomarkers, previously collected data from the same cohort [21, 23, 24] were used. These include a tumour suppresser marker (P53), proliferation marker (Ki67), adhesion molecules (E-cadherin [CDH1] and N-cadherin [CDH2]), basal phenotype markers (CK5, CK14, CK17 positive; ER negative), and epidermal growth factor receptor (EGFR).

Prior to IHC staining, the primary anti-mouse KRT24 antibody (1:1,500; OTI4E8, Thermo Fisher, UK) was validated and assessed using an immunoblotting technique. To validate the specificity of KRT24, both luminal A type (MCF7) and HER2-positive ER− (SKBR3) human BC cell lines (obtained from the American Type Culture Collection, Rockville, MD, USA) were used, as recommended by the manufacturer. The rabbit anti-β-actin antibody (A5441, clone AC-15, Sigma-Aldrich, Gillingham, UK) at 1:5,000 was used as a housekeeping protein and showed a band at approximately 42 KDa. A single specific band for KRT24 protein was observed at the expected molecular weight of ∼55 KDa after incubation overnight (Fig. 1a).

To assess and evaluate the distribution of KRT24 expression, full-face section slides of BC cases (n = 14), representing several tumour grades and different molecular subtypes, including normal breast tissue with terminal ductal lobular units, ductal carcinoma in situ, and invasive BC tissue, were used (Fig. 1a–f). Patients’ samples were arranged in TMAs, as previously described [25]. To assess the immunoreactivity of KRT24, a Novolink Max Polymer Detection kit (Leica, Newcastle, UK) was utilised [14]. Prior to incubating the samples overnight at 4°C with KRT24 antibody diluted 1:1,500, the samples were exposed to citrate antigen retrieval (pH 6.0). A high-resolution digital imager (NanoZoomer; Hamamatsu Photonics, Welwyn Garden City, UK) was used to scan the KRT24 slides at ×20 magnification and visualised using viewing software (Xplore; Philips, UK), to determine the protein expression level. The histochemical score (H-score) [26] was detected semi-quantitatively using the equation for the combination with staining intensity (0–3), multiplied by the proportion of tumour cells (0–100). Staining intensity was categorised into four groups: 3 (strong staining); 2 (moderate staining); 1 (weak staining); and 0 (no staining) (Fig. 1g–k). The final H-score was determined in a range of 0–300. No assessment was made for cores with folded tissue and/or with less than 15% of tumour areas. Inter-observer concordance was checked by conducting blind double-scoring with two researchers (Y.K. and M.A.). To illustrate, one observer (Y.K.), who was blind to clinicopathological and patient outcome data, scored the data. Furthermore, the scoring’s repeatability among observers was assessed. Another observer (MA) picked a subset of cores (30%) at random and double-rated them blindly. Cohen’s kappa test was also used to validate the agreement between both observers [27].

SPSS statistical software (IBM SPSS Statistic, Version 24.0, Chicago, IL, USA) was used to analyse and present the study data. To categorise mRNA and protein data, the median was used and the cut-off points were {METABRIC: 5.54; TCGA: 1; protein: (cytoplasmic 115 and membrane 160) H-score). Intra-class correlation coefficient (ICC) was performed to evaluate inter-observer agreement in KRT24 IHC scoring. A Pearson’s correlation test was also used to investigate the correlation between mRNA expression of KRT24 and epithelial-mesenchymal transition (EMT). A chi-square test was utilised to study the correlation between KRT24 expression and the other categorical variables at both transcriptomic and proteomic levels. A KM survival test was performed to assess the correlation with patient outcomes, including breast-cancer-specific survival (BCSS) and distant metastasis-free survival (DMFS). The Cox regression model was used for multivariate analysis. A p value of <0.05 was used to determine statistical significance. This study followed the reporting recommendations for tumour marker prognostic study (REMARK) criteria [28].

Among the transcriptomic cohorts, high KRT24 mRNA expression was significantly associated with clinicopathological features indicative of poor prognosis including higher tumour grade (p = 0.017), poor prognostic NPI groups (p = 0.002), hormone receptor (ER and PR) negativity (both p < 0.001), HER2 positivity (p < 0.001; Table 1). When we investigate the correlation with EMT-related genes in the METABRIC cohort, high KRT24 mRNA expression showed an association with adhesion molecules, specifically negative correlation with CDH1 (p = 0.003) and positive correlation with CDH2 (p < 0.001; Table 2). Additionally, high KRT24 mRNA expression correlated with EMT transcription regulators showing positive associations with TWIST1 (p < 0.001), LLGL2 (p = 0.013), and CRUMBS (p = 0.003), while negative correlations were observed with ZEB1 (p = 0.017), SNAIL (p = 0.010), and CTNNB1 (p = 0.003; Table 2). In the TCGA cohort, high KRT24 mRNA expression was correlated with higher expression of EMT transcription regulators, including NFKB1 (p < 0.001) and GSK3B (p = 0.007; Table 2).

Given the aforementioned finding that high KRT24 mRNA expression is associated with poor prognostic clinicopathological features, the TCGA-BRCA cohort was divided into low and high KRT24 expression groups following median mRNA expression categorisation, and DEGs were identified. A total of 1,620 DEGs were discovered, 1,133 with lower expression and 485 with higher expression in samples with high versus low KRT24 mRNA expression (Fig. 2). Downregulated genes in tumours with high KRT24 mRNA expression were highly enriched in pathways related to cytokine-cytokine receptor interaction, IL-17 signalling, and ER signalling. High DEGs in tumours with high KRT24 mRNA were enriched in pathways related to the PI3K-Akt signalling pathway, metabolic pathways, and metabolism of xenobiotics by cytochrome P450 (online suppl. Fig. 1a,b; for all online suppl. material, see https://doi.org/10.1159/000546022).

Before TMA staining and assessment, 14 full-face slides from BC patients, representing different molecular subtypes and tumour grades, were stained to evaluate the distribution of KRT24 protein expression across breast tissue. No heterogenous distribution of KRT24 protein expression was observed, confirming the suitability of TMA for assessing KRT24 protein expression. The KRT24 protein was predominantly detected in the cytoplasm and in the membranes of invasive BC tumour cells. Following the double-scoring of cases, an excellent concordance rate was observed between the two observers (ICC = 0.86, p < 0.001 for cytoplasmic expression; and ICC = 0.92, p < 0.001 for nuclear expression). To validate these results, the second observer scored cases were categorised into 4 categories and then compared those with the categories generated by the main scorer using Cohen’s kappa test and showed value of agreement = 0.83, p < 0.001 for cytoplasmic expression, and value of agreement = 0.88, p < 0.001 for membranous expression. Therefore, the main observer’s (YK) scoring was considered as the final decision analysis. After exclusion of uninformative cores, 832 cases were included in further analysis. The distribution of KRT24 protein expression showed a wide range from absent to high expression (cytoplasmic: 0–230; membranous: 0–250). For classification into low (negative) and high (positive) expression, the median H-score (cytoplasmic 115; membranous: 160) was used. High KRT24 protein expression was detected in 48% of the cases (cytoplasmic n = 402; membranous: n = 397).

High cytoplasmic (HC) expression of KRT24 was significantly associated with clinicopathological parameter characteristic of poor prognosis including younger patients’ age (p = 0.021; Table 3), negative hormone receptors (ER and PR; both p < 0.001), and HER2-positive tumours (p < 0.001; Table 3). When cases were stratified based on BC histological types, HC protein expression of KRT24 was significantly correlated with ductal NST BC tumours compared to other BC histological types (p < 0.001; Table 3). High KRT24 cytoplasmic protein expression was positively associated with aggressive BC biomarkers, including high levels of P53 (p = 0.020), N-cadherin (p = 0.026), basal cytokeratins (p < 0.001), and EGFR (p = 0.003), but with reduced expression of E-cadherin (p < 0.001; Table 3). Among BC IHC subtypes, the proportion of high KRT24 expression was more pronounced in the HER2-positive ER− subtype (p < 0.001; Table 3).

Similar association was observed between high membranous (HM) protein expression of KRT24 and ER (p = 0.007), HER2 (p = 0.003), N-cadherin (p = 0.013), basal cytokeratins (p = 0.016), EGFR (p = 0.047), whereas it associated with reduced E-cadherin (p = 0.047; Table 3). Upon the BC IHC subtypes, the HER2-positive ER− subtype had the highest percentage of KRT24 expression (p < 0.001; see Table 3).

Combined KRT24 membranous/cytoplasmic protein expression was evaluated in the invasive BC cohort, and it was categorised into four different groups based on their KRT24 protein expression (high/or low) on the cytoplasm and membranes. While 309 cases (37.4%) were detected as being in a low cytoplasmic/low membranous (LC/LM) group, 119 cases (14.4%) were in a high cytoplasmic/low membranous (HC/LM) group, 117 cases (14.1%) in a low cytoplasmic/high membranous (LC/HM) group, and 282 cases (34.1%) in a high cytoplasmic/high membranous (HC/HM) group. The HC/HM group was more likely to demonstrate invasive BC with aggressive behaviour: high histological grade (p = 0.018), ER and PR negativity (p < 0.001 and p = 0.006; respectively), HER2 negativity (p < 0.001), high level of P53 (p = 0.043), low level of E-cadherin (p = 0.004), high level of N-cadherin (p = 0.029), high level of basal phenotype (p = 0.007), and high level of EGFR (p = 0.010; online suppl. Table 1).

Among tumour histological types, HC/HM KRT24 protein expression was significantly associated with ductal NST BC tumours compared to other BC histological types (p < 0.001; online suppl. Table 1). When we stratified combined KRT24 protein expression according to BC IHC subtypes, the HC/HM KRT24 protein expression group was strongly correlated with HER2-positive ER− BC subtypes compared to other BC IHC subtypes (p < 0.001; online suppl. Table 1).

High KRT24 mRNA expression was significantly indicative of poor prognoses as cases with high KRT24 mRNA expression had shorter BCSS compared to BC patients with low KRT24 mRNA expression in the KM plotter and the bc-GenExMiner online data sets (p = 0.026; HR = 1.12; 95% CI 1.01–1.23, Fig.3a, and p = 0.009; HR = 1.16; 95% CI 1.04–1.30, Fig. 3b, respectively). Patients who were diagnosed with HC protein expression of KRT24 had shorter BCSS (p = 0.041, HR = 1.30, 95% CI 1.11–1.70; Fig. 3c) and DMFS (p = 0.039, HR = 1.37, 95% CI 1.06–1.76; Fig. 3d) compared to patients diagnosed with a low level of KRT24 cytoplasmic protein expression. Multivariate analysis revealed that cytoplasmic protein expression of KRT24 was associated with poor patient outcomes in terms of BCSS (p = 0.041, HR = 1.34, 95% CI 1.20–1.78), independent of other prognostic parameters, including tumour size, tumour grade, LN status, ER, and HER2 (Table 4).

Patients who had HM KRT24 protein expression revealed poor BCSS (p = 0.046, HR = 1.26, 95% CI 1.00–1.61; Fig. 3e) compared to patients who had LM KRT24 expression. In contrast, no significant association was observed between membranous KRT24 protein expression and patients’ survival in DMFS (p = 0.561, HR = 0.93, 95% CI 0.74–1.18; Fig. 3f). Multivariate analysis indicated that membranous KRT24 protein expression was an independent biomarker associated with worse patient outcomes in terms of BCSS (p = 0.043, HR = 1.60, 95% CI 1.01–1.70), regardless of other prognostic parameters, including tumour size, tumour grade, LN status, ER, and HER2 (Table 4).

When we examined the patient survival of the combined KRT24 protein expression groups, patients who had HC/HM KRT24 protein expression were indicated to have the lowest BCSS (p = 0.029, HR 1.70, 95% CI 1.85–2.41; Fig. 4a) compared to the other patient groups. While the HC/LM KRT24 protein expression patient group showed the second shortest BCSS, the LC/HM KRT24 protein expression patient group presented the second longest BCSS among all the patient groups. Notably, patients who had LC/LM KRT24 protein expression were shown to have the best BCSS outcomes compared to the other patient groups (Fig. 4a). However, in DMFS, no significant association was observed between combined KRT24 protein expression and patients’ survival (p = 0.308, HR 1.04, 95% CI 0.92–1.19; online suppl. Fig. 2a). Multivariate analysis showed that combined protein expression of KRT24 was associated with poor patient outcomes in terms of BCSS (p = 0.030, HR = 1.30, 95% CI 1.18–1.95), independent of other prognostic parameters, including tumour size, tumour grade, LN status, ER, and HER2 (Table 4).

Due to the fact that KRT24 overexpression in BC was substantially related to poor clinical outcomes independent of other prognostic parameters, this led us to test the ability of KRT24 expression to predict the outcome in patients treated with chemotherapy only. Thus, we investigated the proteomic cohort to understand KRT24 predictive value along with treatment response in BC patients. When patients were distributed based on their therapy, HC KRT24 protein expression in patients who received chemotherapy was associated with poor patient outcomes in BCSS (p = 0.047, HR = 1.28, 95% CI 1.04–1.72; Fig. 4b) and a higher risk of DMFS (p = 0.013, HR = 1.49, 95% CI 1.85–2.41; Fig. 4c). A similar association of survival was observed with combined KRT24 expression in BCSS (p = 0.035, HR = 1.47, 95% CI 1.05–2.08; Fig. 4d) but not in DMFS (p = 0.785, HR = 0.91, 95% CI 0.74–1.10; online suppl. Fig. 2b). Conversely, membranous KRT24 protein expression did not show any association with patients who received chemotherapy (p = 0.543, HR = 0.88, 95% CI 0.76–1.36; and p = 0.196, HR = 1.01, 95% CI 0.96–1.46: online suppl. Fig. 2c,d; respectively). Moreover, no such correlation was seen in those patients who did not receive chemotherapy.

To expand our understanding of KRT24 expression and its predictive value in treatment responses and patient outcomes, we used the ROC plotter data portal to estimate the predictive significance of KRT24 expression in BC patients’ responses to treatments. Our findings confirmed a strong statistical association with KRT24 expression being higher in individuals who did not respond to chemotherapy (p < 0.001), which implies that KRT24 could be employed as a prognostic marker for BC (online suppl. Fig. 3).

The findings of this study provide new insights into the role of KRT24 in BC. KRT24 is primarily expressed in the corneal epithelium and may contribute to epithelial cell differentiation-associated expression due to its similarity to other members of the type I KRT sub-cluster, including supra-basal epidermal KRT (KRT10), supra-basal corneal KRT (KRT12), and the inner-root sheath KRTs (KRT5, KRT26, KRT27, and KRT28) [29]. Consequently, KRT24 influences the epithelial cells, which are a crucial component of breast tissue. It is hypothesised that KRT24 may drive breast cells towards carcinogenic behaviour by activating EMT mechanisms [30, 31].

In this study, KRT24 was significantly associated with features indicative of aggressive behaviour and with poor patient outcomes, consistent with previous observations of KRT24 in colorectal cancer [6, 7]. Our findings demonstrate a correlation between high KRT24 expression and adverse clinicopathological features, including hormone receptor (ER, PR) negativity and HER2 positivity, positive expression of the tumour suppressor protein p53, E-cadherin negativity, N-cadherin positivity, high frequency of basal phenotype, and HER2-positive ER− BC subtype. These results suggest a potential oncogenic role for KRT24 in BC.

At both transcriptomic and proteomic levels, high KRT24 expression was strongly negatively associated with adhesion molecule-related biomarker (E-cadherin [CDH1]). Conversely, a significant positive association was observed with adhesion molecule biomarker (N-cadherin [CDH2]). This suggests that KRT24 plays a role in BC by contributing to the dysregulation of cell adhesion process at the primary site influencing epithelial cell migration and invasion during BC development. High KRT24 was speculated to be involved in BC tumour cell migration by controlling the EMT process by increasing the N-cadherin level and regulating the E-cadherin level, which allow BC cells to lose their adhesion to the primary site and increase migration and invasiveness behaviours [30, 32]. This series of actions aims to facilitate the migration process across lymphatic spaces and initiate secondary tumour sites by activating the Wnt and PI3K signalling pathways [33].

In this study, it was found that KRT24 expression is significantly associated with adhesion molecule (EMT-related) biomarkers at both the transcriptomic and proteomic levels. The data revealed a negative correlation between KRT24 gene expression and well-established EMT-related genes such as CDH1, ZEB1, SNAIL, and CTNNB1, at the transcriptomic level. Conversely, a positive association was observed between KRT24 gene expression and other EMT-related genes like CDH2, TWIST1, LLGL2, and CRUMBS. Similar associations were also evident between KRT24 protein expression (membranous, cytoplasmic, and combined), a low level of E-cadherin (CDH1), and a high level of N-cadherin (CDH2) at the proteomic level. These findings align with a previous in vitro study that demonstrated KRT24 overexpression boosted the cancer cell migration process by affecting the E-cadherin and N-cadherin levels [3]. This might explain the role of KRT24 expression during BC development by being involved in creating a suitable tumour microenvironment and invasiveness behaviour, including losing adhesion to the primary site and boosting the migration, invasion and metastases that occur due to controlling the EMT process.

Our study indicated that a high level of KRT24 protein is an independent predictive marker strongly associated with poor patient outcomes in terms of both shorter BCSS and DMFS. Moreover, patients who received chemotherapy and had high expression of KRT24 protein showed shorter survival compared to those with absence/low expression. High expression of KRT24 was also found to be associated with the HER2-positive ER− BC subtype, which is one of the most aggressive BC molecular subtypes [34, 35]. Likewise, KRT24 expression was significantly associated with worse clinicopathological features including hormone receptor negativity and high levels of P53 marker, basal phenotype, and EGFR. Together, these observations suggest the role of KRT24 in BC progression and development. Therefore, this opens up the possibility of gaining better biological insights into BC and utilising KRT24 as a therapeutic target for BC.

Nonetheless, our DEG analysis identified several important downregulated pathways associated with upregulation of KRT24 expression including cytokine-cytokine receptor interaction, IL-17 signalling, and ER signalling, highlighting the important role and mechanisms underlying KRT24 prognostic significance in BC progression and patient outcomes. For instance, ER signalling pathway downregulation has been associated with cisplatin resistance in BC cell lines, which is in agreement with our data showing poor chemotherapy response in patients harbouring high expression of KRT24 [36].

Our findings also showed that BC patients with high expression of KRT24 who were treated with chemotherapy had shorter outcome. This was demonstrated at both the transcriptome and protein levels, as well as an increased risk of distant metastases and death among chemotherapy-treated BC patients. Therefore, we suggest that KRT24 could play a role in BC aggressiveness that is not completely addressed by systemic chemotherapy. Thus, KRT24 expression could be utilised as an indicator of poor outcomes, even in chemotherapy-treated patients.

This study provides valuable insights into the role of KRT24 in BC. However, there are limitations that need to be addressed. It is necessary to gather data from different cohorts across multiple centres and conduct a uniform treatment in a randomised clinical study for an unbiased evaluation of KRT24 in BC. Another limitation of the study is that no anti-HER2 targeted therapy was utilised in this cohort, while the classical cyclophosphamide, methotrexate and 5-fluorouracil was used as chemotherapy. Moreover, further research is needed to uncover the exact molecular mechanisms underlying KRT24’s function in BC and to validate its potential as a predictive factor in the early stages of the disease. Future in vivo and in vitro investigations are also warranted.

In conclusion, this study revealed the prognostic and predictive significance of KRT24 in BC. KRT24 was significantly associated with low patients’ survival and poor clinicopathological features. High KRT24 is involved in the progression of advanced BC and is a predictive biomarker of patients’ resistance to chemotherapy. Hence, it might serve as both a therapeutic target and a prognostic marker. Therefore, further investigations aim to determine the exact role of KRT24 in BC through in vitro and in vivo studies that are necessary, especially in aggressive molecular subtypes.

The authors would like to thank the Deanship of Scientific Research at Shaqra University for supporting this work.

This study was approved by the Nottingham Research Ethics Committee 2 under the title “Development of a molecular genetic classification of breast cancer” and the North West – Greater Manchester Central Research Ethics Committee under the title “Nottingham Health Science Biobank (NHSB)” reference number 15/NW/0685. Informed written consent was obtained from all individuals prior to surgery to use their tissue materials in research, and samples were anonymised. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The release of data was also pseudonymised as per the UK Human Tissue Act regulations. This article does not contain any studies with animals performed by any of the authors.

All the authors declare that they have no conflict of interest.

There is no funding for this manuscript.

Conception and design: Yousif A. Kariri, Mansour Alasaleem, and Emad A. Rakha. Staining and scoring analysis: Yousif A. Kariri and Mohammed Alkharaiji. Data analysis and interpretation: Yousif A. Kariri, Mansour Alasaleem, Taher A. Kariri, Mohammed Asad, and Emad A. Rakha. Paper writing: Yousif A. Kariri, Mansour Alasaleem, Taher A. Kariri, and Emad A. Rakha. All authors contributed to revision and approval of the final version of the paper.

The data that support the findings of this study are not publicly available due to ethical reasons but are available from the corresponding author (E.A.R.) upon reasonable request.