ENDOU is potentially a robust diagnostic and prognostic biomarker in oral squamous cell carcinoma

ENDOU is potentially a robust diagnostic and prognostic biomarker in oral squamous cell carcinoma | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article ENDOU is potentially a robust diagnostic and prognostic biomarker in oral squamous cell carcinoma Karolin Dahlgren, Blanka Kolodziej, Helena Hermelin, Carolin Maria Dahlström, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6340150/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Oral squamous cell carcinoma (OSCC) is responsible for approximately 190,000 deaths each year worldwide, primarily due to late-stage diagnosis. The lack of clinically useful biomarkers has been identified as one of the crucial factors contributing to diagnostic delay in OSCC, limiting the impact of tumor biology on treatment decisions and leading to poor prognosis. We investigated the clinical utility of tissue staining for ENDOU protein, a recently identified tumor suppressor in oral mucosa, as a biomarker for early OSCC detection. Additionally, using publicly available datasets and versatile systems biology tools we dissected the molecular profiles of low ENDOU expressing tumors and precancerous lesions to elucidate the role of ENDOU in the disease progression. Results Our results demonstrated consistent downregulation of ENDOU mRNA in OSCC in 11 independent discovery cohorts. In the validation cohorts, semi-quantitative immunohistochemical assessment revealed complete loss of ENDOU protein in 93% of tumors (p<0.001), which was independent of clinicopathological parameters such as TNM stage, tumor grade, age, or gender. The confusion matrix demonstrated high diagnostic accuracy of ENDOU staining in OSCC detection (0.93, 95% CI = 0.85-0.97; AUC = 0.95, 95% CI = 0.88-1.00, p<0.0001), which was significantly higher (p<0.05) compared to PDPN staining (0.76, 95% CI = 0.63-0.86), used as a reference biomarker in this study. Multivariate analysis demonstrated that ENDOU loss could serve as an independent adverse prognostic marker for OSCC (HR = 2.21, 95% CI = 1.22-4.01, p<0.01). Molecular profiling of low ENDOU expressing tumors and oral potentially malignant disorders (OPMDs) revealed its association with biological processes such as keratinocyte differentiation and immune response. Furthermore, low ENDOU expressing tumors exhibited hyperactivation of c-myc, and inhibition of p53 signaling pathways while in the precancerous lesions, ENDOU loss was associated with hyperactivation of early oncogenic pathways including JAK-STAT, angiogenesis and EMT. Conclusions Our study highlights the potential of ENDOU as a diagnostic biomarker for early detection of OSCC and disease prognostication. Moreover, loss of ENDOU is associated with perturbation of vital biological processes and signaling pathways in the context of oral carcinogenesis. Further studies are warranted to validate these findings and explore their clinical implications. oral cancer oral potentially malignant disorders OSCC OPMDs biomarkers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Background Oral squamous cell carcinoma (OSCC) is recognized as one of the most aggressive cancer forms, affecting around 390,000 people worldwide each year (1). Accounting for some 190, 000 deaths and a staggering 1% increase in the incidence rates annually, OSCC is rapidly emerging as a global public health issue (1, 2). Due to frequent delays in diagnosis, only 50% of OSCC patients survive for five years or more, representing a survival rate that has been virtually unchanged since the 1970’s (3-5). Conversely, diagnosing OSCC at the early stage (T1) improves survival rates dramatically (6), which strongly supports the rationale of early detection to improve the disease outcome. Current OSCC prognostication relies almost exclusively on histopathology and the stage of the tumor, as for example classified according to the tumor-node-metastasis (TNM) system (7), which does not describe the tumor biology and therefore is not useful as a guide for an individual treatment plan. Identifying clinically useful biomarkers for early detection, and to institute precision medicine is critical to improve survival and quality of life in patients with OSCC. There are currently no established biomarkers clinically applicable for the early detection of OSCC and OPMDs which are at high risk to undergo malignant transformation, nor to identify the early signs of disease recurrence (8, 9). Therefore, an incisional biopsy followed by histopathological examination remains the gold standard and the only confirmatory test despite significant drawbacks such as invasiveness and inherent variability in inter-rater reliability (10, 11). Several studies investigated clinical utility of well-established biomarkers for malignant transformation of squamous epithelium, in detection of OSCC. For instance p16, surrogate marker for human papilloma virus (HPV) induced malignant transformation, demonstrated poor diagnostic (12, 13) and limited prognostic value in patients with oral lesions (12, 14). Others, including ki67, p63 and cyclin D1 aided exclusively to OSCC prognostication (15). Among emerging biomarkers, podoplanin (PDPN) is currently one of the most promising biomarkers for the early diagnosis and monitoring of oral cancer (16, 17). This is due to its capability to accurately predict the risk of malignant transformation in precancerous lesions (18, 19), its significant role in cancer progression (19) and its association with poor clinical outcomes (20, 21). Biomarkers are crucial for the advancement of precision medicine, allowing for personalized monitoring of disease progression and therapeutic response (22). Furthermore, they can pave the way towards the development of less invasive diagnostic modalities (23). Gene expression profiling, especially through the “omics approach” facilitates comprehensive understanding of malignant transformation and enables identification of molecular biomarkers that dissect complexity of tumor biology during disease progression (24). Furthermore, availability of high-throughput omics databases provides an unprecedented opportunity for discovery and pre-validation of novel cancer biomarkers with high potential for rapid translation to clinical settings (25). Using comprehensive bioinformatics analysis of the transcriptome from several hundred OSCC patients and matching controls available via The Cancer Genome Atlas project (TCGA-HNSC) (26), as well as individual datasets from several well-designed gene expression profiling studies (27-36), we have identified ENDOU as a potentially robust biomarker for early detection of OSCC and its prognostication. ENDOU encodes a family of enzymes with endoribonuclease activity which has recently been identified as a possible tumor suppressor in oral as well as several other carcinomas originating within squamous epithelium (37-39). In addition, several studies reported its prognostic significance in breast (40), ovarian (41) and head and neck carcinoma (42). To our knowledge this study is the first to examine the clinical utility of ENDOU as a diagnostic biomarker for OSCC. RESULTS ENDOU is downregulated in OSCC ENDOU mRNA expression in normal oral mucosa and OSCC tissues was mined using mRNA expression data for OSCC from The Cancer Genome Atlas – Head and neck squamous cell carcinoma project (TCGA-HNSC, https://cancergenome.nih.gov, last accessed April 18, 2024). Analysis revealed that ENDOU mRNA was significantly downregulated in OSCC tissues (n=309) compared to normal oral mucosa tissues (n=44), (log FC = -3.1, padj<0.0001, Figure 1A). HPV positive tumors tended to express lower ENDOU levels than HPV negative OSCC (p=0.053, Figure 1B). Consistent and robust ENDOU mRNA downregulation was confirmed by meta-analysis of 10 additional, publicly available datasets (Figure 1C) using random effects model (log FC = -3.14, p<0.0001) due to significant study heterogeneity (I 2 = 95.20 %, tau 2 = 2.72). ENDOU protein expression was further qualitatively and semi-quantitatively assessed by IHC analysis in 70 primary OSCC, 3 metastatic tumors and 23 nonmalignant oral tissues with squamous differentiation including 7 hyperplasia, 2 of chronically inflamed oral mucosa, 9 adjacent normal tissue and 5 of normal oral mucosa. IHC staining of malignant and nonmalignant tissues with non-squamous differentiation was characterized qualitatively (Supplementary Figure S1). Representative ENDOU immunostaining demonstrated a dominant cytoplasmic staining pattern localized to the suprabasal keratinocytes of the squamous epithelium, in both malignant and nonmalignant tissues (Figure 2A-F). Semi-quantitative IHC assessment of ENDOU expression showed that 91.4% of control tissues expressed ENDOU protein, while 93% of primary OSCC were negative (p<0.001, Figure 2K). Interestingly, high ENDOU expression was maintained in all premalignant oral lesions including chronic inflammatory lesions, hyperkeratotic and hyperplastic lesions, despite the presence of cellular atypia in some cases (Figure 2C-F). Additionally, ENDOU staining was negative in two of the three lymph node metastases that were examined (Figure 2I-J). Loss of ENDOU expression is an early event in the malignant transformation of oral mucosa Assessment of ENDOU protein expression in different TNM stages of OSCC using IHC revealed that 93% of the tumors were completely negative for ENDOU expression already at stage T1 (Figure 2L). IHC scores were not found to be confounded by TNM stage or other vital clinicopathological variables examined in this study ( Table 1 ). ENDOU downregulation was observed already in dysplastic lesions, but not prior to oral dysplasia, in inflammatory and hyperplastic/hyperkeratotic lesions (Figure 2M-N). Concordantly, all OPMDs available in the validation cohort including hyperplastic mucosa with some cellular atypia maintained stable ENDOU protein expression (Figure 2C-F, Supplementary Figure S2). ENDOU is potentially a robust diagnostic biomarker for OSCC The constructed ROC curve demonstrated an AUC of 0.946 (95% CI = 0.88-1.00, p<0.0001, Supplementary Figure S3A). Meanwhile, the confusion matrix, with the selected cut-off score of 4, demonstrated high diagnostic accuracy of IHC staining, even with the inclusion of premalignant lesions in the matrix ( Table 2 ). Next, we compared diagnostic performance of ENDOU staining to the diagnostic performance of PDPN staining in OSCC detection. We chose PDPN as the reference biomarker because previous reports highlighted its high performance in OSCC detection and disease prognostication (17, 60). In line with previous reports (20, 60), we observed significant overexpression of PDPN in OSCC, which had four distinct patterns: patchy, peripheral, diffused and mixed (p<0.0001, Supplementary Figure S4A-F). When compared to PDPN (AUC = 0.90, 95% CI = 0.83-0.98, p<0.0001, Supplementary Figure S3B), ENDOU staining demonstrated significantly higher sensitivity (Δ sensitivity = 21%, p<0.05) and slightly lower specificity = -9 %, p<0.05) in OSCC detection, making it more accurate than PDPN (Δ accuracy = 17%, p<0.05, Table 2 ). Comparison between ROC curves revealed increased area under the curve for ENDOU staining, which was however not statistically significant (ΔAUC = 0.05, p=0.40, Figure 3A). When correlating ENDOU expression with PDPN using GSE227919 dataset, we observed a significant negative correlation between these two biomarkers during the progression of oral carcinogenesis (r = -0.6, p<0.0001, Figure 3B). Low ENDOU expression is an independent risk factor associated with worse prognosis in OSCC In agreement with the previous findings by Xu et al. (42), ENDOU downregulation showed a significant prognostic value in the subgroup of head and neck carcinomas confined exclusively to the oral cavity. Here, we confirmed its prognostic value in patients with malignancies confined exclusively to the oral cavity. In a univariate analysis we observed significantly shorter overall survival (p<0.01, Figure 3C), disease specific survival (p<0.01, Figure 3D) and progression free survival (p<0.05, Figure 3E) in OSCC patients whose tumors expressed low ENDOU levels. In addition, patients with progressive disease expressed significantly lower ENDOU levels compared to those in remission (p<0.05, Figure 3F). This was further corroborated in a multivariate analysis after correcting confounding effects of tumor stage, tumor grade, radiation therapy, age and gender which demonstrated that low ENDOU expression in tumors is an independent adverse prognostic factor (HR=2.2, 95% CI =1.22-4.01, p=0<0.01, Figure 4). Low ENDOU expressing OSCC are poorly differentiated and exhibit high MYC and E2F target scores Tumors dichotomized into low and high ENDOU expressing groups were subsequently subjected to a comprehensive systems biology analysis to further elucidate the role of ENDOU in OSCC tumor biology. The analysis indicated ENDOU’s involvement in biological processes such as epithelial cell differentiation, keratinization, cytoskeleton organization, and humoral immune response (Figure 5A). Interestingly, most downregulated genes were enriched in terms related to keratinocyte differentiation (NES = -3.61, p<0.0001, Figure 5C), while the upregulated genes were enriched in processes such as embryonic organ development (NES = 2.41, p<0.001, Figure 5D). Moreover, low ENDOU expressing tumors showed an embryonic stem cell-like gene expression signature also reported in poorly differentiated, aggressive human head and neck carcinomas by Rickman et al. (61) (Figure 5E) Ben-Porath et al. (62) (Figure 5F-H). Conversely, ENDOU-positive tumors in the validation cohort maintained a high level of differentiation, showing weak to moderate focal positivity confined to the central areas of the tumor nests (Figure 6B). These tumors exhibited also a distinct PDPN protein expression localized to the proliferating periphery of the tumor cell nests (Figure 6F). In contrast, poorly differentiated OSCC exhibited diffused PDPN staining and were ENDOU negative without exception (Figure 6C, 6F, 6H). To gain a deeper understanding of the molecular characteristics associated with the loss of ENDOU, we conducted GSVA using a meticulously curated list of HALLMARK gene sets (63). We calculated pathway scores for each sample and compared them collectively across the two conditions. Our findings revealed that putative oncogenic pathways, including c-myc (log FC = 0.2, p<0.01) and E2F (log FC = 0.24, p<0.001), were significantly upregulated (Figure 5B), while pivotal tumor suppressive, p53 pathway was significantly downregulated (log FC = -0.13, p<0.001) in patients with low ENDOU expressing tumors (Figure 5B). Immune deconvolution of low ENDOU expressing tumors Given that tumors expressing low levels of ENDOU demonstrated suppression of immune response gene signatures (Figure 7A-B), we performed an immune deconvolution analysis on these tumors using CIBERSORT algorithm, as previously described (64). Our aim was to investigate the composition of immune cells and understand their potential role in the progression and prognosis of these tumors. Compared to high ENDOU expressing tumors, those expressing low ENDOU levels were significantly less infiltrated by plasma cells, neutrophils, and regulatory T cells (Tregs) (Figure 7D). In contrast, more M2 macrophages were observed within the tumor stroma of low ENDOU expressing OSCC (Figure 7D). There was no statistically significant difference between immune score, stromal score or ESTIMATE score between two tumor groups (Figure 7C). Molecular profiling of low ENDOU expressing precancerous lesions Most OPMDs do not show any evidence of dysplasia, and yet they frequently undergo malignant transformation (65). This is why we opted to include hyperkeratosis/hyperplasia lesions without dysplasia in the subsequent analysis, together with dysplastic lesions. Differentially expressed genes in lesions expressing low ENDOU levels were functionally enriched in biological processes such as keratinocyte differentiation, cytoskeleton organization and humoral immune response (Figure 8A). GSVA analysis demonstrated hyperactivation of hallmark pathways including JAK-STAT (log FC = 0.25, padj<0.05), EMT (log FC = 0.33, padj<0.01) and angiogenesis (log FC = 0.34, padj<0.01) in the low ENDOU expressing OPMDs lesions (Figure 8B). In contrast, early estrogen response (log FC = -0.23, padj<0.001) and late estrogen response (log FC = -0.17, padj<0.01) hallmark pathways were the most downregulated pathways in the same tissues (Figure 8B). Lesions expressing low ENDOU levels were significantly less infiltrated by naïve B cells, while a higher number of plasma cells (p<0.05), M1 macrophages (p<0.05), and M2 macrophages (p<0.01) were observed (Figure 8C). These lesions exhibited enhanced IL-17 pathway signaling activity (NES = 2.54, padj<0.001, Figure 8D) compared to those expressing high levels of ENDOU. The gene-disease network association analysis revealed significant association between ENDOU downregulation and pathological conditions affecting the oral cavity such as inflammation, periodontal disease, aggressive periodontitis and acute periodontitis (padj<0.001, Figure 8E). DISCUSSION Detecting oral cancer at an early stage, such as T1, or earlier is crucial for improving treatment outcomes and survival rates (8, 66). Nevertheless, biomarkers aiding early detection of OSCC remain scarce (8). In the current study, we report on the potential clinical applicability of ENDOU protein expression measurement in oral mucosa as an accurate biomarker for the early detection of OSCC. In addition to its high diagnostic accuracy and established prognostic value, we demonstrated that its performance was not significantly confounded by traditional clinicopathological parameters such as tumor stage, grade, gender, or age, making it diagnostically relevant in broader patient populations. Moreover, we found that ENDOU downregulation might reflect the perturbation of vital biological processes and signaling pathways in the early stages of oral carcinogenesis. Widespread remodeling of the transcriptome is a hallmark of malignant transformation and endoribonucleases such as ENDOU play crucial roles in its early stages through their involvement in RNA processing, regulation of gene expression, and maintenance of cellular homeostasis (37, 42, 67-69). Corroborated by a complete loss of ENDOU protein expression in 93% of T1 stage OSCC examined in this study and its significant downregulation in the dysplastic lesions, it is plausible that dysregulation of ENDOU expression is indicative of early events in oral carcinogenesis, which could be exploited diagnostically. The inclusion of ENDOU protein loss detection in the confusion matrices demonstrated its high diagnostic accuracy, especially when compared to PDPN, a previously established biomarker for early OSCC detection and disease prognostication (20, 21). In contrast to PDPN (19, 20, 70), its mRNA and protein levels remained stable in non-dysplastic, hyperkeratotic/hyperplastic oral lesions. This suggests that loss of ENDOU expression might identify clinically more relevant lesions, warranting further investigation in the group of patients with oral dysplasia. ENDOU is recently characterized as a tumor suppressor in squamous epithelium (37, 42) that negatively regulates somatic mRNA abundance of several putative oncogenes including c-myc (67, 71). In agreement, we observed significant upregulation of c-myc target genes in the OSCC tumors expressing low ENDOU mRNA levels, indicating high c-myc pathway activity. Furthermore, low ENDOU expressing tumors exhibited molecular profile associated with several phenotypic features characteristic for activated c-myc, including more pronounced M2 macrophage polarization (72), immune evasion, enhanced E2F activity (73), poor differentiation (74) and epigenetic reprogramming enhancing cancer stem cell phenotype (75). Additionally, malign tumors expressing low ENDOU levels exhibited suppression of p53 activity and a higher G2M score, which might have further contributed to shorter overall survival and poorer treatment response in these patients, which is in agreement with several previous reports (75-77). Corroborating ENDOU’s role as a critical regulator of peripheral B cell survival and antigen responsiveness (67) we observed expression profile associated with both, lower B cell infiltration and suppressed humoral response in low ENDOU expressing tumors. This might further explain shorter overall survival in this patient group, due to the crucial role B cells have in enhancing the density and anti-tumoral response of T cells in the tumor microenvironment (78). Even though evidence is still considerably limited, ENDOU appears to be significantly involved in the squamous cell differentiation and epithelial cell homeostasis (38, 42). In line with this, we observed that ENDOU downregulation in OSCC as well as its precursor lesions is strongly associated with inhibition of biological processes such as keratinocyte differentiation and cytoskeletal organization. Moreover, OSCC tissues that maintained some extent of ENDOU protein expression, retained a high degree of differentiation closely resembling hyperplastic and low dysplastic oral mucosa. Our findings show that ENDOU-positive tumors are not only well-differentiated but also exhibit a distinct PDPN expression pattern confined to the proliferating periphery of the tumor cell nests. This expression pattern was previously established in a group of low-aggressive, well-differentiated head and neck carcinomas that exhibit morphology mimicking the pattern seen in the hyperplastic and low dysplastic epithelium (20, 21, 79). Contrary to OSCC, lesions with low ENDOU expression were more immunologically active than those with high ENDOU expression. We observed molecular profiles associated with higher infiltration of plasma cells and macrophages, particularly those with M2 polarization. Together, these cells can create a highly immunosuppressive, inflammatory microenvironment conducive to tumor formation (80, 81). Moreover, lesions with low ENDOU expression exhibited enhanced IL-17 signaling, which is known to play a crucial role during early malignant transformation of the oral epithelium. (82). IL-17 promotes cell proliferation, survival, angiogenesis and EMT in the microenvironment increasing the risk of malignant transformation (83). Concordantly, ENDOU downregulation in precancerous lesions was associated with the activation of oncogenic pathways that interact with IL-17 during early tumorigenesis, including the JAK-STAT signaling pathway (84), EMT (85) and angiogenesis (86). Consistent with these findings, ENDOU expression in OPMDs demonstrated a moderate negative correlation with PDPN, an established biomarker for predicting malignant transformation in oral mucosa (16, 18, 60, 87) which modulates immune response, drives EMT and promotes angiogenesis during early steps of oral carcinogenesis (17). CONCLUSION In conclusion, the current study is, to our knowledge, the first study investigating clinical applicability of ENDOU protein expression measurement in oral mucosa as a biomarker for OSCC detection. Moreover, loss of ENDOU expression reflects perturbation of vital biological processes and signaling pathways in the context of early oral carcinogenesis. The observation that ENDOU expression is completely lost as early as the T1 stage of tumor development, along with its downregulation in dysplastic lesions, could have implications for early detection and intervention strategies in OSCC. Further studies are warranted to validate these findings and explore their clinical utility. Methods and materials Patient cohorts Demographics, survival, and gene expression profile data for 309 OSCC patients and 44 normal oral mucosa samples were accessed and downloaded from the TCGA data portal. Next, gene expression data was obtained from another 460 OSCC patients and 219 normal oral mucosa samples from published publicly available datasets (Table S1). Additionally, we downloaded gene expression profiles from 54 precancerous oral lesions including 35 leukoplakias with dysplasia and 16 leukoplakias without dysplasia (Table S1) (27, 28). Protein expressions were assessed in a tissue microarray (TMA) slide containing 50 cases of OSCC and 10 cases of adjacent normal mucosa (OR601c, BioCat GMBH, Germany) as well as in oral cavity disease spectrum (oral cavity carcinoma progression) TMA slide(s?), containing 28 cases of squamous cell carcinoma, 4 adenocarcinoma, 8 mucoepidermoid carcinoma, 2 basal cell carcinoma, 4 metastatic carcinoma, 8 adamantinoma, 7 hyperplasia, 5 each of adjacent tissue with inflammation, adjacent normal tissue and normal tissue (OR802, BioCat GMBH, Germany). TMA slides included information regarding age, sex, anatomic site, TNM and pathology grade. Sample sizes were determined by data availability. Selection of GEO datasets for meta-analysis To investigate whether ENDOU expression significantly differs between OSCC and normal mucosa in a wider patient population, we performed a meta-analysis of mRNA profiling data available on the GEO Datasets Platform as of November 2024. The search strings used were: (oral [All Fields] AND ("Cancer"[Organism] OR cancer [All Fields])) AND ("transcriptome gene expression"[Filter] AND "bioproject gds"[Filter] AND "org human"[Filter]). All the eligible studies met the following inclusion criteria: 1) Organism: we selected only studies on Homo sapiens; 2) Biological matrices: we choose only studies where the mRNA evaluation was performed on solid biopsies; 3) Type of analysis: we included datasets of mRNA profiling by means of high-throughput methods. 4) Study design: we considered only studies comparing normal oral mucosa with cancerous tissue and tissue from OPMDs. Studies with no publicly available datasets, duplicate studies, as well as those performed in tumor explants, cell lines, swabs and brush biopsies were excluded (Table S2). In total, 10 studies were deemed as eligible and included in the subsequent meta-analysis (Table S1). Heterogeneity was tested using I 2 statistic with values over 50% and Chi-squared test with P ≤ 0.05 indicating strong heterogeneity between the studies. Tau-squared (τ 2 ) was used to determine how much heterogeneity was explained by subgroup differences. The data was pooled using The Random Effect Maximum Likelihood (REML) method, implemented in the R package Metafor version 4.6.0 (43). Effect sizes and confidence intervals were obtained using Limma and DESeq2 bioconductor packages for R statistical software as described previously (44, 45). Differential gene expression analysis Level III mRNA-sequencing data (raw counts) belonging to OSCC patients and normal controls from The Cancer Genome Atlas (TCGA) HNSC project (26) were downloaded using TCGAbiolinks package for Bioconductor, version 2.33.0 (46). Differential gene expression between normal oral mucosa and OSCC has been performed using DESeq2 package for Bioconductor, version 1.44.0 (44). Normalized and log2 transformed ENDOU and PDPN gene expression (log2 (n + 1)) produced by variance stabilizing transformations (VST) of the raw expression matrix were visualized using packages, ggplot2 version 3.5.1 (47), ggpubr version 0.6.0 (48), and ggsignif, version 0.64.9000 (49). In addition, differential gene expression analysis was performed between top quartile and bottom quartile ENDOU expressing OSCC from the HNSC-TCGA project. Thresholds for all differential expression analyses were set at an absolute fold change cut-off of 1.5 and false discovery rate (FDR) of 5% (adjusted p < 0.05). Correlation between ENDOU and PDPN gene expression was calculated and illustrated using ggplot2 version 3.5.1 using ggcorrplot extension (47). Gene expression profiles from individual OSCC and leukoplakia GEO datasets (Table S1) were assessed and downloaded from Gene Expression Omnibus database using GEOquery package for Bioconductor, version 2.72.0 (50). Differential gene expressions between normal oral mucosa, OPMDs and OSCC has been performed using DESeq2 package for Bioconductor, version 1.44.0 (44) for data obtained by RNA sequencing or Limma package (45) for data obtained using microarray platforms. Additionally, differential gene expression analysis was performed between top quartile and bottom quartile ENDOU expressing precursor lesions using GSE227919 gene expression data. Thresholds for all differential expression analyses were set at an absolute fold change cut-off of 1.5 and FDR of 5% (padj < 0.05). Pathway Enrichment, Gene-disease Association, Gene Ontology and Gene Set Variation Analysis Functional enrichment in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, gene-disease network analysis (DisGeNET) and gene ontology (GO) were performed using clusterProfiler package for Bioconductor, version 4.0 (51) and visualized using packages enrichplot, version 1.24 (52) and GOPlot, version 1.0.2 (53). Gene set variation analysis (GSVA) at the pathway level was performed using GSVA package version 1.5.2 (54). We extracted a total of 50 hallmark gene sets from the molecular signature database (MSigDB, http://www.gsea-msigdb.org/gsea ) and used the enrichment score (ES) generated from the gene set variation analysis (GSVA) to quantify the activity of hallmark pathways in the ENDOU low and ENDOU high expressing leukoplakia and OSCC tissues. The higher the value of ES per individual sample was, the stronger the activity of the perturbed pathway was assumed to be. A volcano plot was constructed to visualize the differential expression analysis at a pathway level between low and high ENDOU expressing tissues. The infiltration levels of immune cells in high and low ENDOU expressing tumors were inferred by using the R package immunedeconv v.2.13 (55). The R package ESTIMATE was used to estimate the immune cell and stromal cell content in each tumor sample (56). Immunohistochemistry Surgically removed specimens from the lining mucosa from different locations within the oral cavity including Alveolar mucosa, the lining between the buccal and labial mucosae. Buccal mucosa, the inside lining of the cheeks. Labial mucosa, the inside lining of the lips. Masticatory mucosa keratinized squamous found on the dorsum of the tongue and hard palate and attached gingiva. were preserved in 10% neutral, phosphate buffered formalin for 24 h at room temperature before they were dehydrated with gradient ethanol, cleared with xylene, and embedded in paraffin with the aid of a tissue processor (Leica Biosystems). Prior to staining, TMA slides were heated in an oven at 60°C for 2 h. Immunohistochemical analysis was performed on 5 µm thick TMA sections that were deparaffinized using EZ prep solution. The deparaffinization step was done for 8 min at 75°C in the Ventana Discovery XT platform (Ventana Medical System, Inc.). Antigen retrieval was performed with Tris-EDTA buffer pH 7.8 (Cell Conditioner #1; Ventana Medical System, Inc.) at 95°C for 44 min followed by blocking of endogenous peroxides and proteins with inhibitor CM (Ventana Medical System, Inc.; cat. no. 760-159) at 37°C for 4 min. Immunohistochemical staining for ENDOU was performed as a fully automated assay in the BenchMark ULTRA automated slide stainer (Ventana Medical System, Inc.) using rabbit polyclonal antibody against ENDOU (500 ng/ml; Atlas antibodies, cat. no. HPA012388; http://www.atlasantibodies.com ) and UltraView DAB IHC Detection kit (cat. no. 760-500; Ventana Medical System, Inc.). The slides were counterstained with hematoxylin at room temperature for 8 min and post counterstained with Bluing Reagent (cat. no. 760-2037; Ventana Medical System, Inc.) at room temperature for 8 min. Immunohistochemical staining for PDPN was performed as a fully automated assay in the Dako Omnis automated slide stainer (Agilent Technologies Inc., CA, USA) using mouse monoclonal antibody against human Podoplanin (D2-40 clone, Code IR072, Agilent Technologies Inc. , CA, USA) and EnVision FLEX visualization kit (Code K8000, Agilent Technologies Inc., CA, USA). Analysis was performed on 5 μm thick TMA sections (Or601c) that were deparaffinized using Clearify (Code Gc810, Agilent Technologies Inc., CA, USA). The deparaffinization step was done for 1 minute at 25°C in Dako Omnis automated slide stainer (Agilent Technologies Inc., CA, USA). Antigen retrieval was performed with EnVision Flex TRS high pH (Code K8000, Agilent Technologies Inc., CA, USA) 97°C for 30 minutes followed by blocking of endogenous peroxides and proteins with EnVision FLEX peroxidase-Blocking Reagent (Agilent Technologies Inc., CA, USA) for 3 minutes. The slides were counterstained with hematoxylin at room temperature for 3 minutes. The negative controls were obtained by omitting the primary antibody, whereas TMA slides were complemented by an external positive control. Positive tissue controls for ENDOU (placenta) and PDPN (tonsil) were utilized in accordance with manufacturer’s recommendations. Evaluation of immunohistochemical staining Immunohistochemical staining for ENDOU and PDPN was evaluated and manually scored by a senior pathologist using a light microscope (BX45, Olympus, Sweden) taking into account the mean intensity of the tissue biopsy staining categorized into: 0 = no staining; 1 = weak staining; 2 = moderate staining; 3 = intense staining, and the mean proportion of stained tumor cells categorized into: 0 = negative; 1 = 50% = strongly positive specimens. Obtained results were documented in an EXCEL sheet (Microsoft Inc. CA, USA) and a comprehensive qualitative and semi-quantitative score was calculated by multiplying the corresponding intensity and proportion category values for each tissue core (scores 0–3, negative expression; score 4, weak positive, score 6, moderately positive and score 9, strongly positive). For ENDOU staining in the OR802 array, 12 cases of SCC and 1 normal mucosa were excluded from further analysis due to tissue loss and suboptimal staining. The association between ENDOU and PDPN expression and the clinicopathological characteristics of patients with OSCC was assessed using Fisher’s exact test. Diagnostic accuracy of ENDOU and PDPN staining in detection of OSCC The diagnostic accuracy for each protein biomarker was assessed individually by constructing confusion matrices as well as receiver operating characteristic (ROC) curves and calculating the area under the curve (AUC) as described previously (38). During this examination, individuals without the disease were regarded as controls, whereas patients with OSCC were classified as positive. ROC analysis was performed using IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Armonk, New York, USA) using an extended trapezoidal rule and a nonparametric method analogous to the Wilcoxon/Mann-Whitney test. Confidence intervals were determined using DeLong's variance estimate. Comparison between ROC curves for candidate biomarkers was performed with bootstrap method with D statistics and the curves were plotted using pROC package for Bioconductor, version 1.18.5 (57). Survival analysis Univariate survival analysis in OSCC-TCGA was performed using survival package version 3.6–4 (58). Univariate Cox proportional hazards regression was used to determine the association between ENDOU mRNA, clinical covariates, and overall survival. The log-rank test was plotted to compare survival curves in two groups divided by the high and low quartiles of mRNA expression as previously described (38). A log-rank P<0.05 (Kaplan-Meier) was considered as the threshold for significance. For the progression free survival analysis, a positive event was defined as a new tumor event after initial treatment. The hazards in two groups were compared by calculating the hazard ratios and confidence intervals according to Altman and De Stavola (59). Multivariate analysis including following variables: age, clinical stage, and histological grade in relation to ENDOU expression was performed using IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Chicago, Michigan, USA). Statistical analysis Statistical analyses were performed using R version 4.23 and Bioconductor 3.19 statistical software (R Foundation for Statistical Computing) or IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Chicago, Michigan, USA). Comparisons between groups were performed by applying either Willcoxon sum-rank test (two groups) or Kruskal-Wallis (three or more than three groups) with Dunn's post hoc test. The correlation of PDPN with ENDOU was analyzed using the Pearson correlation coefficient. Univariate and multivariate Cox proportional hazards regression was used to determine the association between ENDOU mRNA, clinical covariates, and overall survival. The association between ENDOU and PDPN expression and the clinicopathological characteristics of patients with OSCC was assessed using the Fischer’s exact test. Confusion matrices were compared using McNamar’s test. Statistical significance was set at P < 0.05. Availability of data and materials Gene expression profiling was performed entirely on genomic data deposited in publicly available dataset repositories, including Gene Expression Omnibus (GEO) ( https://www.ncbi.nlm.nih.gov/geo ) and The Cancer Genome Atlas (TCGA) ( https://portal.gdc.cancer.gov ). Abbreviations OSCC : Oral squamous cell carcinoma ENDOU : Endonuclease, Poly(U) Specific TNM : The TNM Classification of Malignant Tumors PDPN : Podoplanin ROC : Receiver operating characteristic AUC : Area under curve OPMDs : Oral potentially malignant disorders HNSC : Head and neck squamous cell carcinoma GEO : Gene Expression Omnibus TCGA : The Cancer Genome Atlas HR : Hazard ratio P53 : Protein 53 EMT : Epithelial to mesenchymal transition JAK : Janus kinases STAT : Signal transducer and activator of transcription proteins HPV : Human papilloma virus GSEA : Gene set enrichment analysis GSVA : Gene set variation analysis OS : Overall survival DSS : Disease specific survival PFS : Progression free survival BP : Biological processes KEGG : Kyoto Encyclopedia of Genes and Genomes HALLMARK : Hallmark gene sets PRC2 : Polycom repressive complex 2 SUZ12 : SUZ12 Polycomb Repressive Complex 2 Subunit NES : Normalized enrichment score IHC : Immunohistochemistry c-myc : MYC Proto-Oncogene, BHLH Transcription Factor ES : Embryonal stem cell Declarations Acknowledgements We thank Department of Laboratory Medicine and Pathology, Region Jönköping County, Jönköping, Sweden for skillful PDPN IHC staining. Funding The present study was supported by the Dalarna County Council (grant no. CKFUU-936869 to VB) and Analytic Imaging Diagnostics Arena (AIDA) (VINNOVA grant no. 2021-01420, AIDA 2218 to JL). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study. Author information Karolin Dahlgren and Blanka Kolodziej contributed equally to this work. Authors and Affiliations Department of Laboratory Medicine and Pathology, Region Jönköping County, Jönköping, Sweden. Karolin Dahlgren and Blanka Kolodziej Department of Pathology and Cytology, Falun County Hospital, Sweden Helena Hermelin Department of Immunology, Genetics & Pathology, Science for Life Laboratory, Uppsala University, Uppsala, Sweden. Caroline Maria Dahlström and Liza Löf Department of Surgical Sciences, Oral and Maxillofacial Surgery, Medical Faculty, Uppsala University, Uppsala, Sweden. Jan-Michael Hirsch Department of Research, Development and Education, Public Dental Services, Stockholm, Sweden Jan-Michael Hirsch Centre for Image Analysis, Department of Information Technology, Uppsala University, Uppsala, Sweden. Joakim Lindblad Clinical Research Center Dalarna, Uppsala University, Falun, Sweden Ulrika Pellas and Vladimir Basic Department of Clinical Diagnostics, School of Health and Welfare, Jönköping University, Jönköping, Sweden Vladimir Basic Contributions KD, BK, UP, JMH, CRS, JL and VB conceived the study and designed the workflow method. KD, HH and CMD performed different laboratory analyses. BK and KD performed IHC scoring. VB designed and performed bioinformatics analysis. All authors were involved in writing and editing the manuscript. All authors read and approved of the final manuscript. Ethics declarations Immunohistochemical staining was performed on commercially available tissue arrays, for which the manufacturer collected the tissues after obtaining written consent, and according to the regulations protecting the privacy and security of health information prescribed by The Health Insurance Portability and Accountability Act of 1996 (HIPAA) issued by the Secretary of the U.S. Department of Health and Human Services. The present study was performed entirely on de-identified clinical material; thus, it was not subjected to the Swedish Ethical Review Act (SFS 2003:460). References Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians.n/a(n/a). Damgacioglu H, Sonawane K, Zhu Y, Li R, Balasubramanian BA, Lairson DR, et al. Oropharyngeal Cancer Incidence and Mortality Trends in All 50 States in the US, 2001-2017. JAMA Otolaryngology–Head & Neck Surgery. 2022;148(2):155-65. Gigliotti J, Madathil S, Makhoul N. Delays in oral cavity cancer. International Journal of Oral and Maxillofacial Surgery. 2019;48(9):1131-7. Kim Y-J, Kim JH. Increasing incidence and improving survival of oral tongue squamous cell carcinoma. Scientific Reports. 2020;10(1):7877. Zanoni DK, Montero PH, Migliacci JC, Shah JP, Wong RJ, Ganly I, et al. Survival outcomes after treatment of cancer of the oral cavity (1985-2015). Oral Oncol. 2019;90:115-21. Stathopoulos P, Smith WP. Analysis of Survival Rates Following Primary Surgery of 178 Consecutive Patients with Oral Cancer in a Large District General Hospital. J Maxillofac Oral Surg. 2017;16(2):158-63. Almangush A, Mäkitie AA, Triantafyllou A, de Bree R, Strojan P, Rinaldo A, et al. Staging and grading of oral squamous cell carcinoma: An update. Oral Oncol. 2020;107:104799. Haj-Hosseini N, Lindblad J, Hasséus B, Kumar VV, Subramaniam N, Hirsch J-M. Early Detection of Oral Potentially Malignant Disorders: A Review on Prospective Screening Methods with Regard to Global Challenges. Journal of Maxillofacial and Oral Surgery. 2024;23(1):23-32. Walsh T, Macey R, Kerr AR, Lingen MW, Ogden GR, Warnakulasuriya S. Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident lesions. Cochrane Database of Systematic Reviews. 2021(7). Fischer DJ, Epstein JB, Morton Jr TH, Schwartz SM. Interobserver reliability in the histopathologic diagnosis of oral pre-malignant and malignant lesions. Journal of Oral Pathology & Medicine. 2004;33(2):65-70. Shah JP, Gil Z. Current concepts in management of oral cancer--surgery. Oral Oncol. 2009;45(4-5):394-401. DOLL C, STEFFEN C, BECK-BROICHSITTER B, RICHTER M, NEUMANN K, POHRT A, et al. The Prognostic Significance of p16 and its Role as a Surrogate Marker for Human Papilloma Virus in Oral Squamous Cell Carcinoma: An Analysis of 281 Cases. Anticancer Research. 2022;42(5):2405-13. Sundberg J, Korytowska M, Burgos PM, Blomgren J, Blomstrand L, S DEL, et al. Combined Testing of p16 Tumour-suppressor Protein and Human Papillomavirus in Patients With Oral Leukoplakia and Oral Squamous Cell Carcinoma. Anticancer Res. 2019;39(3):1293-300. Palve V, Bagwan J, Krishnan NM, Pareek M, Chandola U, Suresh A, et al. Detection of High-Risk Human Papillomavirus in Oral Cavity Squamous Cell Carcinoma Using Multiple Analytes and Their Role in Patient Survival. J Glob Oncol. 2018;4:1-33. Oliveira LR, Ribeiro-Silva A. Prognostic significance of immunohistochemical biomarkers in oral squamous cell carcinoma. International Journal of Oral and Maxillofacial Surgery. 2011;40(3):298-307. Monteiro L, Mariano LC, Warnakulasuriya S. Podoplanin could be a predictive biomarker of the risk of patients with oral leukoplakia to develop oral cancer: A systematic review and meta-analysis. Oral Diseases. 2024;30(2):207-15. Retzbach EP, Sheehan SA, Nevel EM, Batra A, Phi T, Nguyen ATP, et al. Podoplanin emerges as a functionally relevant oral cancer biomarker and therapeutic target. Oral Oncology. 2018;78:126-36. Swain N, Kumar SV, Routray S, Pathak J, Patel S. Podoplanin—a novel marker in oral carcinogenesis. Tumor Biology. 2014;35(9):8407-13. Monteiro L, do Amaral B, Delgado L, Garcês F, Salazar F, Pacheco JJ, et al. Podoplanin Expression Independently and Jointly with Oral Epithelial Dysplasia Grade Acts as a Potential Biomarker of Malignant Transformation in Oral Leukoplakia. Biomolecules. 2022;12(5):606. Yuan P, Temam S, El-Naggar A, Zhou X, Liu DD, Lee JJ, et al. Overexpression of podoplanin in oral cancer and its association with poor clinical outcome. Cancer. 2006;107(3):563-9. CÎRLIGERIU L, CIMPEAN AM, RAICA M, DOROŞ CI. Dual Role of Podoplanin in Oral Cancer Development. In Vivo. 2014;28(3):341-7. Vargas AJ, Harris CC. Biomarker development in the precision medicine era: lung cancer as a case study. Nature Reviews Cancer. 2016;16(8):525-37. Anstee QM, Castera L, Loomba R. Impact of non-invasive biomarkers on hepatology practice: Past, present and future. Journal of Hepatology. 2022;76(6):1362-78. Kamel HFM, Al-Amodi H. Exploitation of Gene Expression and Cancer Biomarkers in Paving the Path to Era of Personalized Medicine. Genomics Proteomics Bioinformatics. 2017;15(4):220-35. Smith JC, Sheltzer JM. Genome-wide identification and analysis of prognostic features in human cancers. Cell Reports. 2022;38(13):110569. Lawrence MS, Sougnez C, Lichtenstein L, Cibulskis K, Lander E, Gabriel SB, et al. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature. 2015;517(7536):576-82. Khan MM, Frustino J, Villa A, Nguyen BC, Woo SB, Johnson WE, et al. Total RNA sequencing reveals gene expression and microbial alterations shared by oral pre-malignant lesions and cancer. Hum Genomics. 2023;17(1):72. Chen C, Méndez E, Houck J, Fan W, Lohavanichbutr P, Doody D, et al. Gene expression profiling identifies genes predictive of oral squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 2008;17(8):2152-62. Sheu JJ, Lee CC, Hua CH, Li CI, Lai MT, Lee SC, et al. LRIG1 modulates aggressiveness of head and neck cancers by regulating EGFR-MAPK-SPHK1 signaling and extracellular matrix remodeling. Oncogene. 2014;33(11):1375-84. Verduci L, Ferraiuolo M, Sacconi A, Ganci F, Vitale J, Colombo T, et al. The oncogenic role of circPVT1 in head and neck squamous cell carcinoma is mediated through the mutant p53/YAP/TEAD transcription-competent complex. Genome Biol. 2017;18(1):237. Ye H, Yu T, Temam S, Ziober BL, Wang J, Schwartz JL, et al. Transcriptomic dissection of tongue squamous cell carcinoma. BMC Genomics. 2008;9:69. Oghumu S, Knobloch TJ, Terrazas C, Varikuti S, Ahn-Jarvis J, Bollinger CE, et al. Deletion of macrophage migration inhibitory factor inhibits murine oral carcinogenesis: Potential role for chronic pro-inflammatory immune mediators. Int J Cancer. 2016;139(6):1379-90. Rentoft M, Coates PJ, Laurell G, Nylander K. Transcriptional profiling of formalin fixed paraffin embedded tissue: pitfalls and recommendations for identifying biologically relevant changes. PLoS One. 2012;7(4):e35276. Reis PP, Waldron L, Perez-Ordonez B, Pintilie M, Galloni NN, Xuan Y, et al. A gene signature in histologically normal surgical margins is predictive of oral carcinoma recurrence. BMC Cancer. 2011;11:437. Ambatipudi S, Gerstung M, Pandey M, Samant T, Patil A, Kane S, et al. Genome-wide expression and copy number analysis identifies driver genes in gingivobuccal cancers. Genes Chromosomes Cancer. 2012;51(2):161-73. Peng CH, Liao CT, Peng SC, Chen YJ, Cheng AJ, Juang JL, et al. A novel molecular signature identified by systems genetics approach predicts prognosis in oral squamous cell carcinoma. PLoS One. 2011;6(8):e23452. Lou Y, Ren L, Wang L. Hsa_circ_0049396 inhibited oral squamous cell carcinoma progression by regulating the miR-663b/ENDOU axis. Genomics. 2023;115(1):110553. Basic V, Zhang B, Domert J, Pellas U, Tot T. Integrative meta-analysis of gene expression profiles identifies FEN1 and ENDOU as potential diagnostic biomarkers for cervical squamous cell carcinoma. Oncol Lett. 2021;22(6):840. Alotaibi A, Gadekar VP, Gundla PS, Mandarthi S, Jayendra N, Tungekar A, et al. Global comparative transcriptomes uncover novel and population-specific gene expression in esophageal squamous cell carcinoma. Infect Agent Cancer. 2023;18(1):47. Inaba N, Renk T, Daume E, Bohn H. Ectopic production of placenta-"specific" tissue proteins (PP5 and PP11) by malignant breast tumors. Arch Gynecol. 1981;231(1):87-90. Inaba N, Ishige H, Ijichi M, Satoh N, Ohkawa R, Sekiya S, et al. Immunohistochemical detection of pregnancy-specific protein (SP1) and placenta-specific tissue proteins (PP5, PP10, PP11 and PP12) in ovarian adenocarcinomas. Oncodev Biol Med. 1982;3(5-6):379-89. Xu C, Zhang Y, Shen Y, Shi Y, Zhang M, Zhou L. Integrated Analysis Reveals ENDOU as a Biomarker in Head and Neck Squamous Cell Carcinoma Progression. Front Oncol. 2020;10:522332. Viechtbauer W. Conducting Meta-Analyses in R with the metafor Package. Journal of Statistical Software. 2010;36(3):1 - 48. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research. 2015;43(7):e47-e. Colaprico A, Silva TC, Olsen C, Garofano L, Cava C, Garolini D, et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Res. 2016;44(8):e71. Wilkinson L. ggplot2: Elegant Graphics for Data Analysis by H. WICKHAM. Biometrics. 2011;67:678-9. Kassambara A. ggpubr: “ggplot2” Based Publication Ready Plots. R Package Version 0.6.0. 0.6.0 ed2023. Ahlmann-Eltze C, Patil I. ggsignif: R Package for Displaying Significance Brackets for 'ggplot2'. PsyArXiv; 2021. Davis S, Meltzer PS. GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics. 2007;23(14):1846-7. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation. 2021;2(3). G Y. Enrichplot: Visualization of Functional Enrichment Result. R package version 1.24.0. 2024. Walter W, Sánchez-Cabo F, Ricote M. GOplot: an R package for visually combining expression data with functional analysis. Bioinformatics. 2015;31(17):2912-4. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics. 2013;14(1):7. Sturm G, Finotello F, Petitprez F, Zhang JD, Baumbach J, Fridman WH, et al. Comprehensive evaluation of transcriptome-based cell-type quantification methods for immuno-oncology. Bioinformatics. 2019;35(14):i436-i45. Yoshihara K, Shahmoradgoli M, Martínez E, Vegesna R, Kim H, Torres-Garcia W, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun. 2013;4:2612. Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez J-C, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12(1):77. Borgan Ø. Modeling Survival Data: Extending the Cox Model. Terry M. Therneau and Patricia M. Grambsch, Springer‐Verlag, New York, 2000. No. of pages: xiii + 350. Price: $69.95. ISBN 0‐387‐98784‐3. Statistics in Medicine. 2001;20(13):2053-4. Altman DG, De Stavola BL. Practical problems in fitting a proportional hazards model to data with updated measurements of the covariates. Stat Med. 1994;13(4):301-41. Rai K, Prasad K, Sagar P, Rao RS, Vanishri CH, Gowri MS. Expression of podoplanin in predicting the biological behavior of oral squamous cell carcinoma – A clinicopathological correlation. Journal of Advanced Clinical and Research Insights. 2019;6(6):163-70. Rickman DS, Beltran H, Demichelis F, Rubin MA. Biology and evolution of poorly differentiated neuroendocrine tumors. Nat Med. 2017;23(6):1-10. Ben-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, et al. An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet. 2008;40(5):499-507. Liberzon A, Birger C, Thorvaldsdóttir H, Ghandi M, Mesirov JP, Tamayo P. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst. 2015;1(6):417-25. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nature Methods. 2015;12(5):453-7. Stojanov IJ, Woo S-B. Malignant Transformation Rate of Non-reactive Oral Hyperkeratoses Suggests an Early Dysplastic Phenotype. Head and Neck Pathology. 2022;16(2):366-74. Lin N-C, Hsu J-T, Tsai K-Y. Survival and clinicopathological characteristics of different histological grades of oral cavity squamous cell carcinoma: A single-center retrospective study. PLOS ONE. 2020;15(8):e0238103. Poe JC, Kountikov EI, Lykken JM, Natarajan A, Marchuk DA, Tedder TF. EndoU is a novel regulator of AICD during peripheral B cell selection. Journal of Experimental Medicine. 2013;211(1):57-69. Jin Y, Qin X. Comprehensive Analysis of the Roles and Prognostic Value of RNA-Binding Proteins in Head and Neck Squamous Cell Carcinoma. DNA Cell Biol. 2020;39(10):1789-98. Liu H, Ye Z, Wang X, Wu Y, Deng C. Comprehensive analysis of the functions, prognostic and diagnostic values of RNA binding proteins in head and neck squamous cell carcinoma. J Stomatol Oral Maxillofac Surg. 2024;125(4s):101937. Kreppel M, Kreppel B, Drebber U, Wedemayer I, Rothamel D, Zöller J, et al. Podoplanin expression in oral leukoplakia: prognostic value and clinicopathological implications. Oral Diseases. 2012;18(7):692-9. Qi W, Gromoff EDV, Xu F, Zhao Q, Yang W, Pfeifer D, et al. The secreted endoribonuclease ENDU-2 from the soma protects germline immortality in C. elegans. Nature Communications. 2021;12(1):1262. Pello OM, De Pizzol M, Mirolo M, Soucek L, Zammataro L, Amabile A, et al. Role of c-MYC in alternative activation of human macrophages and tumor-associated macrophage biology. Blood. 2012;119(2):411-21. Leone G, Sears R, Huang E, Rempel R, Nuckolls F, Park CH, et al. Myc requires distinct E2F activities to induce S phase and apoptosis. Mol Cell. 2001;8(1):105-13. Wang C, Zhang J, Yin J, Gan Y, Xu S, Gu Y, et al. Alternative approaches to target Myc for cancer treatment. Signal Transduction and Targeted Therapy. 2021;6(1):117. Yoshida GJ. Emerging roles of Myc in stem cell biology and novel tumor therapies. Journal of Experimental & Clinical Cancer Research. 2018;37(1):173. Oshi M, Takahashi H, Tokumaru Y, Yan L, Rashid OM, Matsuyama R, et al. G2M Cell Cycle Pathway Score as a Prognostic Biomarker of Metastasis in Estrogen Receptor (ER)-Positive Breast Cancer. Int J Mol Sci. 2020;21(8). Amit M, Takahashi H, Dragomir MP, Lindemann A, Gleber-Netto FO, Pickering CR, et al. Loss of p53 drives neuron reprogramming in head and neck cancer. Nature. 2020;578(7795):449-54. Bruno TC, Ebner PJ, Moore BL, Squalls OG, Waugh KA, Eruslanov EB, et al. Antigen-Presenting Intratumoral B Cells Affect CD4+ TIL Phenotypes in Non–Small Cell Lung Cancer Patients. Cancer Immunology Research. 2017;5(10):898-907. Rodrigo JP, García-Carracedo D, González MV, Mancebo G, Fresno MF, García-Pedrero J. Podoplanin expression in the development and progression of laryngeal squamous cell carcinomas. Mol Cancer. 2010;9:48. Li J-R, Shaw V, Lin Y, Wang X, Aminu M, Li Y, et al. The prognostic effect of infiltrating immune cells is shaped by proximal M2 macrophages in lung adenocarcinoma. Molecular Cancer. 2024;23(1):185. Kuang D-M. Positive Feedback Loop between B cells/plasma cells and Macrophages Promote M2 Macrophage-Elicited Hepatoma Progression. The Journal of Immunology. 2016;196(1_Supplement):211.6-.6. Ladjevac N, Milovanovic M, Jevtovic A, Arsenijevic D, Stojanovic B, Dimitrijevic Stojanovic M, et al. The Role of IL-17 in the Pathogenesis of Oral Squamous Cell Carcinoma. International Journal of Molecular Sciences. 2023;24(12):9874. Zhao J, Chen X, Herjan T, Li X. The role of interleukin-17 in tumor development and progression. Journal of Experimental Medicine. 2019;217(1):e20190297. Grandis JR, Drenning SD, Zeng Q, Watkins SC, Melhem MF, Endo S, et al. Constitutive activation of Stat3 signaling abrogates apoptosis in squamous cell carcinogenesis in vivo. Proc Natl Acad Sci U S A. 2000;97(8):4227-32. Tang HM, Kuay KT, Koh PF, Asad M, Tan TZ, Chung VY, et al. An epithelial marker promoter induction screen identifies histone deacetylase inhibitors to restore epithelial differentiation and abolishes anchorage independence growth in cancers. Cell Death Discovery. 2016;2(1):16041. Hasina R, Lingen MW. Angiogenesis in oral cancer. J Dent Educ. 2001;65(11):1282-90. de Vicente JC, Rodrigo JP, Rodriguez-Santamarta T, Lequerica-Fernández P, Allonca E, García-Pedrero JM. Podoplanin expression in oral leukoplakia: Tumorigenic role. Oral Oncology. 2013;49(6):598-603. Tables Tables are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table2.docx Table1.docx TableS1.docx TableS2.docx SupplementaryFigures.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6340150","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":438284831,"identity":"33d38f20-7e92-4c2e-9bfa-a52cb0d72486","order_by":0,"name":"Karolin Dahlgren","email":"","orcid":"","institution":"Region Jönköping County","correspondingAuthor":false,"prefix":"","firstName":"Karolin","middleName":"","lastName":"Dahlgren","suffix":""},{"id":438284832,"identity":"5e2b74a4-a8d7-4a8e-b171-e03cb5b1fd5e","order_by":1,"name":"Blanka Kolodziej","email":"","orcid":"","institution":"Region Jönköping County","correspondingAuthor":false,"prefix":"","firstName":"Blanka","middleName":"","lastName":"Kolodziej","suffix":""},{"id":438284833,"identity":"eb795343-cda5-452c-9ee0-fa1b29d63f56","order_by":2,"name":"Helena Hermelin","email":"","orcid":"","institution":"Falun County Hospital","correspondingAuthor":false,"prefix":"","firstName":"Helena","middleName":"","lastName":"Hermelin","suffix":""},{"id":438284834,"identity":"05f55a33-5a3b-4e0c-a754-eab955d13758","order_by":3,"name":"Carolin Maria Dahlström","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Carolin","middleName":"Maria","lastName":"Dahlström","suffix":""},{"id":438284835,"identity":"14c02cf4-a5d3-4287-a9d7-1fba9cc3f4e2","order_by":4,"name":"Liza Löf","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Liza","middleName":"","lastName":"Löf","suffix":""},{"id":438284836,"identity":"2e3dde6c-1270-462c-bad5-859ca6d6392a","order_by":5,"name":"Christina Runow Stark","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"Runow","lastName":"Stark","suffix":""},{"id":438284837,"identity":"5618cc02-53f7-4318-9f1d-e369f7b4a62e","order_by":6,"name":"Jan-Michael Hirsch","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Jan-Michael","middleName":"","lastName":"Hirsch","suffix":""},{"id":438284838,"identity":"d1f3a8e5-813d-4ee8-9e22-d61dda58c847","order_by":7,"name":"Joakim Lindblad","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Joakim","middleName":"","lastName":"Lindblad","suffix":""},{"id":438284839,"identity":"438c9187-853f-417e-bcae-72bbe10c2615","order_by":8,"name":"Ulrika Pellas","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Ulrika","middleName":"","lastName":"Pellas","suffix":""},{"id":438284840,"identity":"a323502e-fdf9-4a2c-8059-0058922d9a8a","order_by":9,"name":"Vladimir Basic","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIiWNgGAWjYDACdgYGiQQGCQYQYkhgsGEmrIUZVUsakVrADAh5mLAOg8PMD288zLHIZ5BufvzhQc15doMD7A8f4NfCZmyRuE3CskHmmJlEwrHbzAYHeIwN8GmRbGYwkwBqMQD6yIwhsQGshU0Cvxb2b1At6Z8/JDacA2phf/4DnxZ+Zh6YLTkGEokNB4BaGMzw6QBpKQb5xYBN5kwZ0C/JzJKHeYzxOoyNvX3jzZ/b6gz4pds3f/xRY5fMd7z94Qe81sD1QulkBiIiExXYkaphFIyCUTAKhj8AAPi6P+DynDolAAAAAElFTkSuQmCC","orcid":"","institution":"Uppsala University","correspondingAuthor":true,"prefix":"","firstName":"Vladimir","middleName":"","lastName":"Basic","suffix":""}],"badges":[],"createdAt":"2025-03-30 20:23:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6340150/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6340150/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95207100,"identity":"8a3ece21-0c98-473e-a79a-d7e7a9dc4c6d","added_by":"auto","created_at":"2025-11-05 13:24:24","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":269740,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eENDOU is downregulated in OSCC.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA) Differential ENDOU expression between OSCC and normal oral mucosa. Raw mRNA expression data obtained from the HNSC-TCGA project. B) Differential ENDOU expression between HPV negative and HPV positive OSCC. Raw mRNA expression data obtained from the HNSC-TCGA project. C)\u003cstrong\u003e \u003c/strong\u003eMeta-analysis of differential ENDOU gene expression between OSCC and normal oral mucosa across 10 independent microarray studies from GEO repository.\u003c/p\u003e","description":"","filename":"F1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/fb9b20796bae293cf2f680e2.jpg"},{"id":95207091,"identity":"d7e83671-337d-466b-8408-3ce1be60b50f","added_by":"auto","created_at":"2025-11-05 13:24:18","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":585915,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eENDOU downregulation is an early event in oral carcinogenesis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIHC assessment of ENDOU expression in: A) normal oral mucosa at 10X magnification.\u003c/p\u003e\n\u003cp\u003eB) normal oral mucosa at 40X magnification. C) chronic inflammatory lesion at 10X magnification. D) chronic inflammatory lesion at 40X magnification. E) oral hyperplasia at 10X magnification. F) oral hyperplasia at 40X magnification. G) OSCC at 10X magnification. H) OSCC at 40X magnification. I) lymph node metastasis at 10X magnification. J) lymph node metastasis at 40X magnification. K) IHC staining proportion of ENDOU in normal oral mucosa and OSCC. L) IHC scores for ENDOU staining in different T stages of OSCC. M) ENDOU mRNA expression is significantly downloaded in oral dysplasia. Raw mRNA expression data obtained from GSE30784. M) ENDOU mRNA expression is significantly downloaded in oral dysplasia but not in pre-dysplastic lesions including hyperkeratosis/hyperplasia. Raw mRNA expression data obtained from GSE227919. *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001, ****p\u0026lt;0.0001. HkNR = hyperkeratosis, non-reactive.\u003c/p\u003e","description":"","filename":"F2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/a8b9b50ca9e49d255ad02569.jpg"},{"id":95207085,"identity":"ed61191f-792d-4ce2-970c-baf364f5110c","added_by":"auto","created_at":"2025-11-05 13:24:17","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":257261,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eENDOU is potentially a robust biomarker for OSCC\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA) Statistical comparison between diagnostic ROC curves for ENDOU and PDPN staining B) Correlation analysis between ENDOU and PDPN expression. Raw mRNA expression data obtained from GSE227919. C) Overall survival analysis. D) Disease specific survival analysis. E) Progression free survival analysis. F) ENDOU is expressed at lower levels in patients with the progressive disease. Raw mRNA expression data obtained from the HNSC-TCGA project.\u003c/p\u003e","description":"","filename":"F3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/e8e41317089544279cad5258.jpg"},{"id":95207090,"identity":"b17cb05a-6d10-4d9e-9e88-19e8704b72f8","added_by":"auto","created_at":"2025-11-05 13:24:18","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":123833,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMultivariate survival analysis\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"F4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/90bd19869f31556ed56abde7.jpg"},{"id":95207088,"identity":"0d83118c-4383-4f5a-af66-906ebd9f660c","added_by":"auto","created_at":"2025-11-05 13:24:17","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":461496,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMolecular profiling av low ENDOU expressing OSCC\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA) Gene ontology analysis of biological processes perturbated in OSCC expressing low ENDOU levels. B) GSVA of differentially expressed pathways in tumors expressing low levels of ENDOU. Green square represents HALLMARK_E2F_TARGETS pathway. Blue square represents HALLMARK_MYC_TARGETS_V2 pathway. C-H) GSEA of differentially expressed genes in low ENDOU expressing OSCC\u003c/p\u003e","description":"","filename":"F5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/8cec3084491fc01514ea43cf.jpg"},{"id":95228719,"identity":"bce17b79-0baf-4a4a-8ce0-e825cfabb566","added_by":"auto","created_at":"2025-11-05 16:34:05","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":352636,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eENDOU positive tumors are well differentiated and show distinct PDPN expression pattern\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIHC analysis of ENDOU expression in: A) normal mucosa. B) Grade 1 OSCC. C) Grade 2 OSCC. D) ENDOU IHC-score in different grades of OSCC. IHC analysis of PDPN expression in: A) normal mucosa. B) Grade 1 OSCC. C) Grade 2 OSCC. D) ENDOU IHC-score in tumors with different PDPN staining patterns. Red arrows point to the tumor areas demonstrating positive biomarker expression.\u003c/p\u003e","description":"","filename":"F6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/6f8b0277f2cddc7f8e444294.jpg"},{"id":95227772,"identity":"414c192c-99a8-4725-acc5-2eb137c9b5e4","added_by":"auto","created_at":"2025-11-05 16:32:56","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":135795,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImmune infiltration analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene Set Enrichment Analysis (GSEA) of significantly enriched pathways; A) Immune response. B) Immune system process. C) Stromal, immune and ESTIMATE scores. D) Boxplot for assessing the degree of immune cell infiltration in low and high ENDOU expressing tumors\u003c/p\u003e","description":"","filename":"F7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/995800f6d19fef4f6b24497c.jpg"},{"id":95206921,"identity":"5abbb253-e983-49df-8e3d-de1372730b44","added_by":"auto","created_at":"2025-11-05 13:24:14","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":370620,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMolecular profiling of low ENDOU expressing OPDMs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene ontology analysis of biological processes perturbated in OPDMs expressing low ENDOU levels. B) GSVA of differentially expressed pathways in lesions expressing low levels of ENDOU. Blue square represents HALLMARK_ANGIOGENESIS pathway. Pink square represents HALLMARK_EPITHELIAL_MESENCHYMAL_TRANSITION pathway. Green square represents HALLMARK_IL6_JAK_STAT3_SIGNALING pathway. C) Immune infiltration analysis. D) Gene-disease association analysis. E) GSEA KEGG pathway analysis. *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001, ****p\u0026lt;0.0001.\u003c/p\u003e","description":"","filename":"F8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/c3993b21b3b76f7c5022e65d.jpg"},{"id":95230689,"identity":"0e5f2863-1c42-46e1-b01f-97cbf22cf211","added_by":"auto","created_at":"2025-11-05 16:38:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3305941,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/382b4f15-69eb-4110-a1c6-e505fe2d3eed.pdf"},{"id":95207083,"identity":"94629e4d-0f7a-4cda-85f5-7ade895d66a6","added_by":"auto","created_at":"2025-11-05 13:24:16","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":23503,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/b08a957e7de3ccccaef7374b.docx"},{"id":95207009,"identity":"81840d21-6cb6-4c59-a33f-c0616e71163b","added_by":"auto","created_at":"2025-11-05 13:24:16","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27100,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/82166e5b6e301d3868e35355.docx"},{"id":95228049,"identity":"01ae67aa-4387-4e2a-8157-03788515bbaa","added_by":"auto","created_at":"2025-11-05 16:33:19","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":24513,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/8ec2be39cad3e9769cc0af61.docx"},{"id":95207092,"identity":"b869236b-d2a4-49ce-8222-dfbb56ba39f3","added_by":"auto","created_at":"2025-11-05 13:24:18","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":45106,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/b369bb90696b56862cedb075.docx"},{"id":95207086,"identity":"a3c221b6-51c9-48a4-a127-d52539082f56","added_by":"auto","created_at":"2025-11-05 13:24:17","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2440704,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-6340150/v1/4654ce009680d5211dd798c0.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"ENDOU is potentially a robust diagnostic and prognostic biomarker in oral squamous cell carcinoma","fulltext":[{"header":"Background","content":"\u003cp\u003eOral squamous cell carcinoma (OSCC) is recognized as one of the most aggressive cancer forms, affecting around 390,000 people worldwide each year (1). Accounting for some 190,\u0026nbsp; \u0026nbsp; \u0026nbsp;000 deaths and a staggering 1% increase in the incidence rates annually, OSCC is rapidly emerging as a global public health issue (1, 2). Due to frequent delays in diagnosis, only 50% of OSCC patients survive for five years or more, representing a survival rate that has been virtually unchanged since the 1970’s (3-5). Conversely, diagnosing OSCC at the early stage (T1) improves survival rates dramatically (6), which strongly supports the rationale of early detection to improve the disease outcome. Current OSCC prognostication relies almost exclusively on histopathology and the stage of the tumor, as for example classified according to the tumor-node-metastasis (TNM) system (7), which does not describe the tumor biology and therefore is not useful as a guide for an individual treatment plan. Identifying clinically useful biomarkers for early detection, and to institute precision medicine is critical to improve survival and quality of life in patients with OSCC.\u003c/p\u003e\n\u003cp\u003eThere are currently no established biomarkers clinically applicable for the early detection of OSCC and OPMDs which are at high risk to undergo malignant transformation,\u0026nbsp;nor\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;to identify the early signs of disease recurrence (8, 9). Therefore, an incisional biopsy followed by histopathological examination remains the gold standard and the only confirmatory test despite significant drawbacks such as invasiveness and inherent variability in inter-rater reliability (10, 11). Several studies investigated clinical utility of well-established biomarkers for malignant transformation of squamous epithelium, in detection of OSCC. For instance p16, surrogate marker for human papilloma virus (HPV) induced malignant transformation, demonstrated poor diagnostic (12, 13) and limited prognostic value in patients with oral lesions (12, 14). Others, including ki67, p63 and cyclin D1 aided exclusively to OSCC prognostication (15). Among emerging biomarkers, podoplanin (PDPN) is currently one of the most promising biomarkers for the early diagnosis and monitoring of oral cancer (16, 17). This is due to its capability to accurately predict the risk of malignant transformation in precancerous lesions (18, 19), its significant role in cancer progression (19) and its association with poor clinical outcomes (20, 21).\u003c/p\u003e\n\u003cp\u003eBiomarkers are crucial for the advancement of precision medicine, allowing for personalized monitoring of disease progression and therapeutic response (22). \u0026nbsp;Furthermore, they can pave the way towards the development of less invasive diagnostic modalities (23). Gene expression profiling, especially through the “omics approach” facilitates comprehensive understanding of malignant transformation and enables identification of molecular biomarkers that dissect complexity of tumor biology during disease progression (24). Furthermore, availability of high-throughput omics databases provides an unprecedented opportunity for discovery and pre-validation of novel cancer biomarkers with high potential for rapid translation to clinical settings (25). Using comprehensive bioinformatics analysis of the transcriptome from several hundred OSCC patients and matching controls available via The Cancer Genome Atlas project (TCGA-HNSC) (26), as well as individual datasets from several well-designed gene expression profiling studies (27-36), we have identified ENDOU as a potentially robust biomarker for early detection of OSCC and its prognostication. ENDOU encodes a family of enzymes with endoribonuclease activity which has recently been identified as a possible tumor suppressor in oral as well as several other carcinomas originating within squamous epithelium (37-39). In addition, several studies reported its prognostic significance in breast (40), ovarian (41) and head and neck carcinoma (42). To our knowledge this study is the first to examine the clinical utility of ENDOU as a diagnostic biomarker for OSCC.\u0026nbsp;\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cem\u003eENDOU is downregulated in OSCC\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eENDOU mRNA expression in normal oral mucosa and OSCC tissues was mined using mRNA expression data for OSCC from The Cancer Genome Atlas – Head and neck squamous cell carcinoma project (TCGA-HNSC, https://cancergenome.nih.gov, last accessed April 18, 2024). Analysis revealed that ENDOU mRNA was significantly downregulated in OSCC tissues (n=309) compared to normal oral mucosa tissues (n=44),\u0026nbsp;(log FC = -3.1, padj\u0026lt;0.0001, Figure 1A). HPV positive tumors tended to express lower ENDOU levels than HPV negative OSCC (p=0.053, Figure 1B). Consistent and robust ENDOU mRNA downregulation was confirmed by meta-analysis of 10 additional, publicly available datasets (Figure 1C) using random effects model (log FC = -3.14, p\u0026lt;0.0001) due to significant study heterogeneity (I\u003csup\u003e2\u003c/sup\u003e = 95.20 %, tau\u003csup\u003e2\u003c/sup\u003e = 2.72). ENDOU protein expression was further qualitatively and semi-quantitatively assessed by IHC analysis in 70 primary OSCC, 3 metastatic tumors and 23 nonmalignant oral tissues with squamous differentiation including 7 hyperplasia, 2 of chronically inflamed oral mucosa, 9 adjacent normal tissue and 5 of normal oral mucosa. IHC staining of malignant and nonmalignant tissues with non-squamous differentiation was characterized qualitatively (Supplementary Figure S1). Representative ENDOU immunostaining demonstrated a dominant cytoplasmic staining pattern localized to the suprabasal keratinocytes of the squamous epithelium, in both malignant and nonmalignant tissues (Figure 2A-F). Semi-quantitative IHC assessment of ENDOU expression showed that 91.4% of control tissues expressed ENDOU protein, while 93% of primary OSCC were negative (p\u0026lt;0.001, Figure 2K). \u0026nbsp;Interestingly, high ENDOU expression was maintained in all premalignant oral lesions including chronic inflammatory lesions, hyperkeratotic and hyperplastic lesions, despite the presence of cellular atypia in some cases (Figure 2C-F). Additionally, ENDOU staining was negative in two of the three lymph node metastases that were examined (Figure 2I-J).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLoss of ENDOU expression is an early event in the malignant transformation of oral mucosa\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAssessment of ENDOU protein expression in different TNM stages of OSCC using IHC revealed that 93% of the tumors were completely negative for ENDOU expression already at stage T1 (Figure 2L). IHC scores were not found to be confounded by TNM stage or other vital clinicopathological variables examined in this study (\u003cem\u003eTable 1\u003c/em\u003e). ENDOU downregulation \u0026nbsp;was observed already in dysplastic lesions, but not prior to oral dysplasia, in inflammatory and hyperplastic/hyperkeratotic lesions (Figure 2M-N). Concordantly, all OPMDs available in the validation cohort including hyperplastic mucosa with some cellular atypia maintained stable ENDOU protein expression (Figure 2C-F, Supplementary Figure S2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eENDOU is potentially a robust diagnostic biomarker for OSCC\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe constructed ROC curve demonstrated an AUC of 0.946 (95% CI = 0.88-1.00, p\u0026lt;0.0001, Supplementary Figure S3A). Meanwhile, the confusion matrix, with the selected cut-off score of 4, demonstrated high diagnostic accuracy of IHC staining, even with the inclusion of premalignant lesions in the matrix (\u003cem\u003eTable 2\u003c/em\u003e). \u0026nbsp;Next, we compared diagnostic performance of ENDOU staining to the diagnostic performance of PDPN staining in OSCC detection. We chose PDPN as the reference biomarker because previous reports highlighted its high performance in OSCC detection and disease prognostication (17, 60). In line with previous reports (20, 60), we observed significant overexpression of PDPN in OSCC, which had four distinct patterns: patchy, peripheral, diffused and mixed (p\u0026lt;0.0001, Supplementary Figure S4A-F). When compared to PDPN (AUC = 0.90, 95% CI = 0.83-0.98, p\u0026lt;0.0001, Supplementary Figure S3B), ENDOU staining demonstrated significantly higher sensitivity (Δ\u0026nbsp;sensitivity = \u0026nbsp;21%, p\u0026lt;0.05) and slightly lower specificity = \u0026nbsp;-9\u0026nbsp;%, p\u0026lt;0.05) in OSCC detection, making it more accurate than PDPN (Δ\u0026nbsp;accuracy = 17%, p\u0026lt;0.05, \u003cem\u003eTable 2\u003c/em\u003e). Comparison between ROC curves revealed increased area under the curve for ENDOU staining, which was however not statistically significant (ΔAUC = 0.05, p=0.40, Figure 3A). When correlating ENDOU expression with PDPN using GSE227919 dataset, we observed a significant negative correlation between these two biomarkers during the progression of oral carcinogenesis (r = \u0026nbsp;-0.6, p\u0026lt;0.0001, Figure 3B).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLow ENDOU expression is an independent risk factor associated with worse prognosis in OSCC\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn agreement with the previous findings by Xu et al. (42), ENDOU downregulation showed a significant prognostic value in the subgroup of head and neck carcinomas confined exclusively to the oral cavity. Here, we confirmed its prognostic value in patients with malignancies confined exclusively to the oral cavity. In a univariate analysis we observed significantly shorter overall survival (p\u0026lt;0.01, Figure 3C), disease specific survival (p\u0026lt;0.01, Figure 3D) and progression free survival (p\u0026lt;0.05, Figure 3E) in OSCC patients whose tumors expressed low ENDOU levels. In addition, patients with progressive disease expressed significantly lower ENDOU levels compared to those in remission (p\u0026lt;0.05, Figure 3F). This was further corroborated in a multivariate analysis after correcting confounding effects of tumor stage, tumor grade, radiation therapy, age and gender which demonstrated that low ENDOU expression in tumors is an independent adverse prognostic factor (HR=2.2,\u0026nbsp;95% CI\u0026nbsp;=1.22-4.01, p=0\u0026lt;0.01, Figure 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLow ENDOU expressing OSCC are poorly differentiated and exhibit high MYC and E2F target scores\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTumors dichotomized into low and high ENDOU expressing groups were subsequently subjected to a comprehensive systems biology analysis to further elucidate the role of ENDOU in OSCC tumor biology. The analysis indicated ENDOU’s involvement in biological processes such as epithelial cell differentiation, keratinization, cytoskeleton organization, and humoral immune response (Figure 5A). Interestingly, most downregulated genes were enriched in terms related to keratinocyte differentiation (NES = -3.61, p\u0026lt;0.0001, Figure 5C), while the upregulated genes were enriched in processes such as embryonic organ development (NES = 2.41, p\u0026lt;0.001, Figure 5D). Moreover, low ENDOU expressing tumors showed an embryonic stem cell-like gene expression signature also reported in poorly differentiated, aggressive human head and neck carcinomas by Rickman \u003cem\u003eet al.\u003c/em\u003e (61) (Figure 5E) Ben-Porath \u003cem\u003eet al.\u003c/em\u003e (62) (Figure 5F-H). Conversely, ENDOU-positive tumors in the validation cohort maintained a high level of differentiation, showing weak to moderate focal positivity confined to the central areas of the tumor nests (Figure 6B). These tumors exhibited also a distinct PDPN protein expression localized to the proliferating periphery of the tumor cell nests (Figure 6F). In contrast, poorly differentiated OSCC exhibited diffused PDPN staining and were ENDOU negative without exception (Figure 6C, 6F, 6H).\u003c/p\u003e\n\u003cp\u003eTo gain a deeper understanding of the molecular characteristics associated with the loss of ENDOU, we conducted GSVA using a meticulously curated list of HALLMARK gene sets (63). We calculated pathway scores for each sample and compared them collectively across the two conditions. Our findings revealed that putative oncogenic pathways, including c-myc (log FC = 0.2, p\u0026lt;0.01) and E2F (log FC = 0.24, p\u0026lt;0.001), were significantly upregulated (Figure 5B), while pivotal tumor suppressive, p53 pathway was significantly downregulated (log FC = -0.13, p\u0026lt;0.001) in patients with low ENDOU expressing tumors (Figure 5B).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImmune deconvolution of low ENDOU expressing tumors\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eGiven that tumors expressing low levels of ENDOU demonstrated suppression of immune response gene signatures (Figure 7A-B), we performed an immune deconvolution analysis on these tumors using CIBERSORT algorithm, as previously described (64). Our aim was to investigate the composition of immune cells and understand their potential role in the progression and prognosis of these tumors. Compared to high ENDOU expressing tumors, those expressing low ENDOU levels were significantly less infiltrated by plasma cells, neutrophils, and regulatory T cells (Tregs) (Figure 7D). In contrast, more M2 macrophages were observed within the tumor stroma of low ENDOU expressing OSCC (Figure 7D). There was no statistically significant difference between immune score, stromal score or ESTIMATE score between two tumor groups (Figure 7C).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMolecular profiling of low ENDOU expressing precancerous lesions\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMost OPMDs do not show any evidence of dysplasia, and yet they frequently undergo malignant transformation (65). \u0026nbsp;This is why we opted to include hyperkeratosis/hyperplasia lesions without dysplasia in the subsequent analysis, together with dysplastic lesions. Differentially expressed genes in lesions expressing low ENDOU levels were functionally enriched in biological processes such as keratinocyte differentiation, cytoskeleton organization and humoral immune response (Figure 8A). GSVA analysis demonstrated hyperactivation of hallmark pathways including JAK-STAT (log FC = 0.25, padj\u0026lt;0.05), EMT (log FC = 0.33, padj\u0026lt;0.01) and angiogenesis (log FC = 0.34, padj\u0026lt;0.01) in the low ENDOU expressing OPMDs lesions (Figure 8B). In contrast, early estrogen response (log FC = -0.23, padj\u0026lt;0.001) and late estrogen response (log FC = -0.17, padj\u0026lt;0.01) hallmark pathways were the most downregulated pathways in the same tissues (Figure 8B). \u0026nbsp;Lesions expressing low ENDOU levels were significantly less infiltrated by naïve B cells, while a higher number of plasma cells (p\u0026lt;0.05), M1 macrophages (p\u0026lt;0.05), and M2 macrophages (p\u0026lt;0.01) were observed (Figure 8C). These lesions exhibited enhanced IL-17 pathway signaling activity (NES = 2.54, padj\u0026lt;0.001, Figure 8D) compared to those expressing high levels of ENDOU. The gene-disease network association analysis revealed significant association between ENDOU downregulation and pathological conditions affecting the oral cavity such as inflammation, periodontal disease, aggressive periodontitis and acute periodontitis (padj\u0026lt;0.001, Figure 8E).\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eDetecting oral cancer at an early stage, such as T1, or earlier is crucial for improving treatment outcomes and survival rates (8, 66). Nevertheless, biomarkers aiding early detection of OSCC remain scarce (8). In the current study, we report on the potential clinical applicability of ENDOU protein expression measurement in oral mucosa as an accurate biomarker for the early detection of OSCC. In addition to its high diagnostic accuracy and established prognostic value, we demonstrated that its performance was not significantly confounded by traditional clinicopathological parameters such as tumor stage, grade, gender, or age, making it diagnostically relevant in broader patient populations. Moreover, we found that ENDOU downregulation might reflect the perturbation of vital biological processes and signaling pathways in the early stages of oral carcinogenesis.\u003c/p\u003e\n\u003cp\u003eWidespread remodeling of the transcriptome is a hallmark of malignant transformation and endoribonucleases such as ENDOU play crucial roles in its early stages through their involvement in RNA processing, regulation of gene expression, and maintenance of cellular homeostasis (37, 42, 67-69). Corroborated by a complete loss of ENDOU protein expression in 93% of T1 stage OSCC examined in this study and its significant downregulation in the dysplastic lesions, it is plausible that dysregulation of ENDOU expression is indicative of early events in oral carcinogenesis, which could be exploited diagnostically. The inclusion of ENDOU protein loss detection in the confusion matrices demonstrated its high diagnostic accuracy, especially when compared to PDPN, a previously established biomarker for early OSCC detection and disease prognostication (20, 21). In contrast to PDPN (19, 20, 70), its mRNA and protein levels remained stable in non-dysplastic, hyperkeratotic/hyperplastic oral lesions. This suggests that loss of ENDOU expression might identify clinically more relevant lesions, warranting further investigation in the group of patients with oral dysplasia.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eENDOU is recently characterized as a tumor suppressor in squamous epithelium (37, 42) that negatively regulates somatic mRNA abundance of several putative oncogenes including c-myc (67, 71). In agreement, we observed significant upregulation of c-myc target genes in the OSCC tumors expressing low ENDOU mRNA levels, indicating high c-myc pathway activity. Furthermore, low ENDOU expressing tumors exhibited molecular profile associated with several phenotypic features characteristic for activated c-myc, including more pronounced M2 macrophage polarization (72), immune evasion, enhanced E2F activity (73), poor differentiation (74) and epigenetic reprogramming enhancing cancer stem cell phenotype (75). Additionally, malign tumors expressing low ENDOU levels exhibited suppression of p53 activity and a higher G2M score, which might have further contributed to shorter overall survival and poorer treatment response in these patients, which is in agreement with several previous reports (75-77). Corroborating ENDOU’s role as a critical regulator of peripheral B cell survival and antigen responsiveness (67) we observed expression profile associated with both, lower B cell infiltration and suppressed humoral response in low ENDOU expressing tumors. This might further explain shorter overall survival in this patient group, due to the crucial role B cells have in enhancing the density and anti-tumoral response of T cells in the tumor microenvironment (78).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEven though evidence is still considerably limited, ENDOU appears to be significantly involved in the squamous cell differentiation and epithelial cell homeostasis (38, 42). In line with this, we observed that ENDOU downregulation in OSCC as well as its precursor lesions is strongly associated with inhibition of biological processes such as keratinocyte differentiation and cytoskeletal organization. Moreover, OSCC tissues that maintained some extent of ENDOU protein expression, retained a high degree of differentiation closely resembling hyperplastic and low dysplastic oral mucosa. Our findings show that ENDOU-positive tumors are not only well-differentiated but also exhibit a distinct PDPN expression pattern confined to the proliferating periphery of the tumor cell nests. This expression pattern was previously established in a group of low-aggressive, well-differentiated head and neck carcinomas that exhibit morphology mimicking the pattern seen in the hyperplastic and low dysplastic epithelium (20, 21, 79).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eContrary to OSCC, lesions with low ENDOU expression were more immunologically active than those with high ENDOU expression. We observed molecular profiles associated with higher infiltration of plasma cells and macrophages, particularly those with M2 polarization. Together, these cells can create a highly immunosuppressive, inflammatory microenvironment conducive to tumor formation (80, 81). \u0026nbsp; Moreover, lesions with low ENDOU expression exhibited enhanced IL-17 signaling, which is known to play a crucial role during early malignant transformation of the oral epithelium. (82). IL-17 promotes cell proliferation, survival, angiogenesis and EMT in the microenvironment increasing the risk of malignant transformation (83). Concordantly, ENDOU downregulation in precancerous lesions was associated with the activation of oncogenic pathways that interact with IL-17 during early tumorigenesis, including the JAK-STAT signaling pathway (84), EMT (85) and angiogenesis (86). Consistent with these findings, ENDOU expression in OPMDs demonstrated a moderate negative correlation with PDPN, an established biomarker for predicting malignant transformation in oral mucosa (16, 18, 60, 87) which modulates immune response, drives EMT and promotes angiogenesis during early steps of oral carcinogenesis (17).\u0026nbsp;\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eIn conclusion, the current study is, to our knowledge, the first study investigating clinical applicability of ENDOU protein expression measurement in oral mucosa as a biomarker for OSCC detection. Moreover, loss of ENDOU expression reflects perturbation of vital biological processes and signaling pathways in the context of early oral carcinogenesis. The observation that ENDOU expression is completely lost as early as the T1 stage of tumor development, along with its downregulation in dysplastic lesions, could have implications for early detection and intervention strategies in OSCC. Further studies are warranted to validate these findings and explore their clinical utility.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods and materials","content":"\u003cp\u003e\u003cem\u003ePatient cohorts\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDemographics, survival, and gene expression profile data for 309 OSCC patients and 44 normal oral mucosa samples were accessed and downloaded from the TCGA data portal. \u0026nbsp;Next, gene expression data was obtained from another 460 OSCC patients and 219 normal oral mucosa samples from published publicly available datasets (Table S1). Additionally, we downloaded gene expression profiles\u0026nbsp;from 54 precancerous oral lesions including 35 leukoplakias with dysplasia and 16 leukoplakias without dysplasia (Table S1) \u0026nbsp;(27, 28). Protein expressions were assessed in a tissue microarray (TMA) slide containing 50 cases of OSCC and 10 cases of adjacent normal mucosa (OR601c, BioCat GMBH, Germany) as well as in oral cavity disease spectrum (oral cavity carcinoma progression) TMA slide(s?), containing 28 cases of squamous cell carcinoma, 4 adenocarcinoma, 8 mucoepidermoid carcinoma, 2 basal cell carcinoma, 4 metastatic carcinoma, 8 adamantinoma, 7 hyperplasia, 5 each of adjacent tissue with inflammation, adjacent normal tissue and normal tissue (OR802, BioCat GMBH, Germany). TMA slides included information regarding age, sex, anatomic site, TNM and pathology grade. Sample sizes were determined by data availability.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSelection of GEO datasets for meta-analysis\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate whether ENDOU expression significantly differs between\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;OSCC and normal mucosa in a wider patient population, we performed a meta-analysis of mRNA profiling data available on the GEO Datasets Platform as of November 2024. The search strings used were: (oral [All Fields] AND (\u0026quot;Cancer\u0026quot;[Organism] OR cancer [All Fields])) AND (\u0026quot;transcriptome gene expression\u0026quot;[Filter] AND \u0026quot;bioproject gds\u0026quot;[Filter] AND \u0026quot;org human\u0026quot;[Filter]). All the eligible studies met the following inclusion criteria: 1) Organism: we selected only studies on Homo sapiens; 2) Biological matrices: we choose only studies where the mRNA evaluation was performed on solid biopsies; 3) Type of analysis: we included datasets of mRNA profiling by means of high-throughput methods. 4) Study design: we considered only studies comparing normal oral mucosa with cancerous tissue and tissue from OPMDs. Studies with no publicly available datasets, duplicate studies, as well as those performed in tumor explants, cell lines, swabs and brush biopsies were excluded (Table S2). In total, 10 studies were deemed as eligible and included in the subsequent meta-analysis (Table S1). Heterogeneity was tested using I\u003csup\u003e2\u003c/sup\u003e statistic with values over 50% and Chi-squared test with P \u0026le; 0.05 indicating strong heterogeneity between the studies. Tau-squared (\u0026tau;\u003csup\u003e2\u003c/sup\u003e) was used to determine how much heterogeneity was explained by subgroup differences. The data was pooled using The Random Effect Maximum Likelihood (REML) method, implemented in the R package Metafor version 4.6.0 (43). Effect sizes and confidence intervals were obtained using Limma and DESeq2 bioconductor packages for R statistical software as described previously (44, 45).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDifferential gene expression analysis\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eLevel III mRNA-sequencing data (raw counts) belonging to OSCC patients and normal controls from The Cancer Genome Atlas (TCGA) HNSC project (26) were downloaded using TCGAbiolinks package for Bioconductor, version 2.33.0 (46). Differential gene expression between normal oral mucosa and OSCC has been performed using DESeq2 package for Bioconductor, version 1.44.0 (44). Normalized and log2 transformed ENDOU and PDPN gene expression (log2 (n\u0026thinsp;+\u0026thinsp;1)) produced by variance stabilizing transformations (VST) of the raw expression matrix were visualized using packages, ggplot2 version 3.5.1 (47), ggpubr version 0.6.0 (48), and ggsignif, version 0.64.9000 (49). In addition, differential gene expression analysis was performed between top quartile and bottom quartile ENDOU expressing OSCC from the HNSC-TCGA project. Thresholds for all differential expression analyses were set at an absolute fold change cut-off of 1.5 and false discovery rate (FDR) of 5% (adjusted p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Correlation between ENDOU and PDPN gene expression was calculated and illustrated using ggplot2 version 3.5.1 using ggcorrplot extension (47).\u003c/p\u003e\n\u003cp\u003eGene expression profiles from individual OSCC and leukoplakia GEO datasets (Table S1) were assessed and downloaded from Gene Expression Omnibus database using GEOquery package for Bioconductor, version 2.72.0 (50). \u0026nbsp;Differential gene expressions between normal oral mucosa, OPMDs \u0026nbsp; and OSCC has been performed using DESeq2 package for Bioconductor, version 1.44.0 (44) for data obtained by RNA sequencing or Limma package (45) for data obtained using microarray platforms. Additionally, differential gene expression analysis was performed between top quartile and bottom quartile ENDOU expressing precursor lesions using GSE227919 gene expression data. Thresholds for all differential expression analyses were set at an absolute fold change cut-off of 1.5 and FDR of 5% (padj\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePathway Enrichment, Gene-disease Association, Gene Ontology and Gene Set Variation Analysis\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFunctional enrichment in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, gene-disease network analysis (DisGeNET) and gene ontology (GO) were performed using clusterProfiler package for Bioconductor, version 4.0 (51) and visualized using packages enrichplot, version 1.24 (52) and GOPlot, version 1.0.2 (53). Gene set variation analysis (GSVA) at the pathway level was performed using GSVA package version 1.5.2 (54). We extracted a total of 50 hallmark gene sets from the molecular signature database (MSigDB, \u003cu\u003ehttp://www.gsea-msigdb.org/gsea\u003c/u\u003e) and used the enrichment score (ES) generated from the gene set variation analysis (GSVA) to quantify the activity of hallmark pathways in the ENDOU low and ENDOU high expressing leukoplakia and OSCC tissues. The higher the value of ES per individual sample was, the stronger the activity of the perturbed pathway was assumed to be. A volcano plot was constructed to visualize the differential expression analysis at a pathway level between low and high ENDOU expressing tissues. The infiltration levels of immune cells in high and low ENDOU expressing tumors were inferred by using the R package immunedeconv v.2.13 (55). The R package ESTIMATE was used to estimate the immune cell and stromal cell content in each tumor sample (56).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImmunohistochemistry\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSurgically removed specimens from the lining mucosa from different locations within the oral cavity including \u003cem\u003eAlveolar mucosa,\u0026nbsp;\u003c/em\u003ethe lining between the buccal and labial\u003cem\u003e\u0026nbsp;mucosae. Buccal mucosa,\u0026nbsp;\u003c/em\u003ethe inside lining of the cheeks.\u003cem\u003e\u0026nbsp;Labial mucosa,\u0026nbsp;\u003c/em\u003ethe inside lining of the lips.\u003cem\u003e\u0026nbsp;Masticatory mucosa\u0026nbsp;\u003c/em\u003ekeratinized squamous found on the dorsum of the tongue and hard palate and attached \u003cem\u003egingiva.\u003c/em\u003e were preserved in 10% neutral, phosphate buffered formalin for 24 h at room temperature before they were dehydrated with gradient ethanol, cleared with xylene, and embedded in paraffin with the aid of a tissue processor (Leica Biosystems). Prior to staining, TMA slides were heated in an oven at 60\u0026deg;C for 2 h. Immunohistochemical analysis was performed on 5 \u0026micro;m thick TMA sections that were deparaffinized using EZ prep solution. The deparaffinization step was done for 8 min at 75\u0026deg;C in the Ventana Discovery XT platform (Ventana Medical System, Inc.). Antigen retrieval was performed with Tris-EDTA buffer pH 7.8 (Cell Conditioner #1; Ventana Medical System, Inc.) at 95\u0026deg;C for 44 min followed by blocking of endogenous peroxides and proteins with inhibitor CM (Ventana Medical System, Inc.; cat. no. 760-159) at 37\u0026deg;C for 4 min. Immunohistochemical staining for ENDOU was performed as a fully automated assay in the BenchMark ULTRA automated slide stainer (Ventana Medical System, Inc.) using rabbit polyclonal antibody against ENDOU (500 ng/ml; Atlas antibodies, cat. no. HPA012388; \u003cu\u003ehttp://www.atlasantibodies.com\u003c/u\u003e) and UltraView DAB IHC Detection kit (cat. no. 760-500; Ventana Medical System, Inc.). The slides were counterstained with hematoxylin at room temperature for 8 min and post counterstained with Bluing Reagent (cat. no. 760-2037; Ventana Medical System, Inc.) at room temperature for 8 min. Immunohistochemical staining for PDPN was performed as a fully automated assay in the Dako Omnis automated slide stainer (Agilent Technologies Inc., CA, USA) using mouse monoclonal antibody against human Podoplanin (D2-40 clone, Code IR072, Agilent Technologies Inc.\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;, CA, USA) and EnVision FLEX visualization kit (Code K8000, Agilent Technologies Inc., CA, USA). Analysis was performed on 5 \u0026mu;m thick TMA sections (Or601c) that were deparaffinized using Clearify (Code Gc810, Agilent Technologies Inc., CA, USA). The deparaffinization step was done for 1 minute at 25\u0026deg;C in Dako Omnis automated slide stainer (Agilent Technologies Inc., CA, USA). Antigen retrieval was performed with EnVision Flex TRS high pH (Code K8000, Agilent Technologies Inc., CA, USA) 97\u0026deg;C for 30 minutes followed by blocking of endogenous peroxides and proteins with EnVision FLEX peroxidase-Blocking Reagent (Agilent Technologies Inc., CA, USA) for 3 minutes. The slides were counterstained with hematoxylin at room temperature for 3 minutes. The negative controls were obtained by omitting the primary antibody, whereas TMA slides were complemented by an external positive control. Positive tissue controls for ENDOU (placenta) and PDPN (tonsil) were utilized in accordance with manufacturer\u0026rsquo;s recommendations.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEvaluation of immunohistochemical staining\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eImmunohistochemical staining for ENDOU and PDPN was evaluated and manually scored by a senior pathologist using a light microscope (BX45, \u0026nbsp; Olympus, Sweden) taking into account the mean intensity of the tissue biopsy staining categorized into: 0\u0026thinsp; =\u0026thinsp; no staining; 1\u0026thinsp;=\u0026thinsp; weak staining; 2\u0026thinsp; = moderate staining; 3\u0026thinsp; =\u0026thinsp; intense staining, and the mean proportion of stained tumor cells categorized into: 0\u0026thinsp; =\u0026thinsp; negative; 1\u0026thinsp;=\u0026thinsp; \u0026lt;10% =\u0026thinsp; negative; 2 =10-50% =\u0026thinsp; positive; 3\u0026thinsp; =\u0026thinsp;\u0026gt; \u0026nbsp;50% =\u0026thinsp; strongly positive specimens. Obtained results were documented in an EXCEL sheet (Microsoft Inc. CA, USA) and a comprehensive qualitative and semi-quantitative score was calculated by multiplying the corresponding intensity and proportion category values for each tissue core (scores 0\u0026ndash;3, negative expression; score 4, weak positive, score 6, moderately positive and score 9, strongly positive). For ENDOU staining in the OR802 array, 12 cases of SCC and 1 normal mucosa were excluded from further analysis due to tissue loss and suboptimal staining. The association between ENDOU and PDPN expression and the clinicopathological characteristics of patients with OSCC was assessed using Fisher\u0026rsquo;s exact test.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDiagnostic accuracy of ENDOU and PDPN staining in detection of OSCC\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe diagnostic accuracy for each protein biomarker was assessed individually by constructing confusion matrices as well as receiver operating characteristic (ROC) curves and calculating the area under the curve (AUC) as described previously (38). During this examination, individuals without the disease were regarded as controls, whereas patients with OSCC were classified as positive. ROC analysis was performed using IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Armonk, New York, USA) using an extended trapezoidal rule and a nonparametric method analogous to the Wilcoxon/Mann-Whitney test. Confidence intervals were determined using DeLong\u0026apos;s variance estimate. Comparison between ROC curves for candidate biomarkers was performed with bootstrap method with D statistics and the curves were plotted using pROC package for Bioconductor, version 1.18.5 (57).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSurvival analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eUnivariate survival analysis in OSCC-TCGA was performed using survival package version 3.6\u0026ndash;4 (58). Univariate Cox proportional hazards regression was used to determine the association between ENDOU mRNA, clinical covariates, and overall survival. \u0026nbsp;The log-rank test was plotted to compare survival curves in two groups divided by the high and low quartiles of mRNA expression as previously described (38). A log-rank P\u0026lt;0.05 (Kaplan-Meier) was considered as the threshold for significance. For the progression free survival analysis, a\u0026nbsp;positive event was defined as\u0026nbsp;a\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;new tumor event after initial treatment. The hazards in two groups were compared by calculating the hazard ratios and confidence intervals according to Altman and De Stavola (59). Multivariate analysis including following variables: age, clinical stage, and histological grade in relation to ENDOU expression was performed using IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Chicago, Michigan, USA).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStatistical analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed using R version 4.23 and Bioconductor 3.19 statistical software (R Foundation for Statistical Computing) or IBM SPSS statistical Software, version 29.0.1 (IBM Inc., Chicago, Michigan, USA). Comparisons between groups were performed by applying either Willcoxon sum-rank test (two groups) or Kruskal-Wallis (three or more than three groups) with Dunn\u0026apos;s post hoc test. The correlation of PDPN with ENDOU was analyzed using the Pearson correlation coefficient. Univariate and multivariate Cox proportional hazards regression was used to determine the association between ENDOU mRNA, clinical covariates, and overall survival. The association between ENDOU and PDPN expression and the clinicopathological characteristics of patients with OSCC was assessed using the Fischer\u0026rsquo;s exact test. Confusion matrices were compared using McNamar\u0026rsquo;s test. Statistical significance was set at P \u0026lt; 0.05.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene expression profiling was performed entirely on genomic data deposited in\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;publicly available dataset repositories,\u0026nbsp;including Gene Expression Omnibus (GEO) (\u003cu\u003ehttps://www.ncbi.nlm.nih.gov/geo\u003c/u\u003e) and The Cancer Genome Atlas (TCGA) (\u003cu\u003ehttps://portal.gdc.cancer.gov\u003c/u\u003e).\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cem\u003eOSCC\u003c/em\u003e: Oral squamous cell carcinoma\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eENDOU\u003c/em\u003e: Endonuclease, Poly(U) Specific\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTNM\u003c/em\u003e: The TNM Classification of Malignant Tumors\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePDPN\u003c/em\u003e: Podoplanin\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eROC\u003c/em\u003e: Receiver operating characteristic\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAUC\u003c/em\u003e: Area under curve\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOPMDs\u003c/em\u003e: Oral potentially malignant disorders\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHNSC\u003c/em\u003e: Head and neck squamous cell carcinoma\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGEO\u003c/em\u003e: Gene Expression Omnibus \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTCGA\u003c/em\u003e: The Cancer Genome Atlas\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHR\u003c/em\u003e: Hazard ratio\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eP53\u003c/em\u003e: Protein 53\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEMT\u003c/em\u003e: Epithelial to mesenchymal transition\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eJAK\u003c/em\u003e: Janus kinases\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSTAT\u003c/em\u003e: Signal transducer and activator of transcription proteins\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHPV\u003c/em\u003e: Human papilloma virus\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGSEA\u003c/em\u003e: Gene set enrichment analysis\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGSVA\u003c/em\u003e: Gene set variation analysis\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOS\u003c/em\u003e: Overall survival\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDSS\u003c/em\u003e: Disease specific survival\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePFS\u003c/em\u003e: Progression free survival\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBP\u003c/em\u003e: Biological processes\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eKEGG\u003c/em\u003e: Kyoto Encyclopedia of Genes and Genomes\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHALLMARK\u003c/em\u003e: Hallmark gene sets\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePRC2\u003c/em\u003e: Polycom repressive complex 2\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSUZ12\u003c/em\u003e: SUZ12 Polycomb Repressive Complex 2 Subunit\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNES\u003c/em\u003e: Normalized enrichment score\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIHC\u003c/em\u003e: Immunohistochemistry\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ec-myc\u003c/em\u003e: MYC Proto-Oncogene, BHLH Transcription Factor\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eES\u003c/em\u003e: Embryonal stem cell\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Department of Laboratory Medicine and Pathology, Region Jönköping County, Jönköping, Sweden for skillful PDPN IHC staining.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe present study was supported by the Dalarna County Council (grant no. CKFUU-936869 to VB) and Analytic Imaging Diagnostics Arena (AIDA) (VINNOVA grant no. 2021-01420, \u0026nbsp; \u0026nbsp; \u0026nbsp;AIDA 2218 to JL). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKarolin Dahlgren and Blanka Kolodziej contributed equally to this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors and Affiliations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDepartment of Laboratory Medicine and Pathology, Region Jönköping County, Jönköping, Sweden.\u003c/p\u003e\n\u003cp\u003eKarolin Dahlgren and Blanka Kolodziej\u003c/p\u003e\n\u003cp\u003eDepartment of Pathology and Cytology, Falun County Hospital, Sweden\u003c/p\u003e\n\u003cp\u003eHelena Hermelin\u003c/p\u003e\n\u003cp\u003eDepartment of Immunology, Genetics \u0026amp; Pathology, Science for Life Laboratory, Uppsala University, Uppsala, Sweden.\u003c/p\u003e\n\u003cp\u003eCaroline Maria Dahlström and Liza Löf\u003c/p\u003e\n\u003cp\u003eDepartment of Surgical Sciences, Oral and Maxillofacial Surgery, Medical Faculty, Uppsala University, Uppsala, Sweden.\u003c/p\u003e\n\u003cp\u003eJan-Michael Hirsch\u003c/p\u003e\n\u003cp\u003eDepartment of Research, Development and Education, Public Dental Services, Stockholm, Sweden\u003c/p\u003e\n\u003cp\u003eJan-Michael Hirsch\u003c/p\u003e\n\u003cp\u003eCentre for Image Analysis, Department of Information Technology, Uppsala University, Uppsala, Sweden.\u003c/p\u003e\n\u003cp\u003eJoakim Lindblad\u003c/p\u003e\n\u003cp\u003eClinical Research Center Dalarna, Uppsala University, Falun, Sweden\u003c/p\u003e\n\u003cp\u003eUlrika Pellas and Vladimir Basic\u003c/p\u003e\n\u003cp\u003eDepartment of Clinical Diagnostics, School of Health and Welfare, Jönköping University, Jönköping, Sweden\u003c/p\u003e\n\u003cp\u003eVladimir Basic\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKD, BK, UP, JMH, CRS, JL and VB conceived the study and designed the workflow method. KD, HH and CMD performed different laboratory analyses. BK and KD performed IHC scoring. VB designed and performed bioinformatics analysis. All authors were involved in writing and editing the manuscript. All authors read and approved of the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunohistochemical staining was performed on commercially available tissue arrays, for which the manufacturer collected the tissues after obtaining written consent, and according to the regulations protecting the privacy and security of health information prescribed by The Health Insurance Portability and Accountability Act of 1996 (HIPAA) issued by the Secretary of the U.S. Department of Health and Human Services. The present study was performed entirely on de-identified clinical material; thus, it was not subjected to the Swedish Ethical Review Act (SFS 2003:460).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians.n/a(n/a).\u003c/li\u003e\n\u003cli\u003eDamgacioglu H, Sonawane K, Zhu Y, Li R, Balasubramanian BA, Lairson DR, et al. Oropharyngeal Cancer Incidence and Mortality Trends in All 50 States in the US, 2001-2017. JAMA Otolaryngology\u0026ndash;Head \u0026amp; Neck Surgery. 2022;148(2):155-65.\u003c/li\u003e\n\u003cli\u003eGigliotti J, Madathil S, Makhoul N. Delays in oral cavity cancer. International Journal of Oral and Maxillofacial Surgery. 2019;48(9):1131-7.\u003c/li\u003e\n\u003cli\u003eKim Y-J, Kim JH. Increasing incidence and improving survival of oral tongue squamous cell carcinoma. Scientific Reports. 2020;10(1):7877.\u003c/li\u003e\n\u003cli\u003eZanoni DK, Montero PH, Migliacci JC, Shah JP, Wong RJ, Ganly I, et al. Survival outcomes after treatment of cancer of the oral cavity (1985-2015). Oral Oncol. 2019;90:115-21.\u003c/li\u003e\n\u003cli\u003eStathopoulos P, Smith WP. Analysis of Survival Rates Following Primary Surgery of 178 Consecutive Patients with Oral Cancer in a Large District General Hospital. J Maxillofac Oral Surg. 2017;16(2):158-63.\u003c/li\u003e\n\u003cli\u003eAlmangush A, M\u0026auml;kitie AA, Triantafyllou A, de Bree R, Strojan P, Rinaldo A, et al. Staging and grading of oral squamous cell carcinoma: An update. Oral Oncol. 2020;107:104799.\u003c/li\u003e\n\u003cli\u003eHaj-Hosseini N, Lindblad J, Hass\u0026eacute;us B, Kumar VV, Subramaniam N, Hirsch J-M. Early Detection of Oral Potentially Malignant Disorders: A Review on Prospective Screening Methods with Regard to Global Challenges. Journal of Maxillofacial and Oral Surgery. 2024;23(1):23-32.\u003c/li\u003e\n\u003cli\u003eWalsh T, Macey R, Kerr AR, Lingen MW, Ogden GR, Warnakulasuriya S. Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident lesions. Cochrane Database of Systematic Reviews. 2021(7).\u003c/li\u003e\n\u003cli\u003eFischer DJ, Epstein JB, Morton Jr TH, Schwartz SM. Interobserver reliability in the histopathologic diagnosis of oral pre-malignant and malignant lesions. Journal of Oral Pathology \u0026amp; Medicine. 2004;33(2):65-70.\u003c/li\u003e\n\u003cli\u003eShah JP, Gil Z. Current concepts in management of oral cancer--surgery. Oral Oncol. 2009;45(4-5):394-401.\u003c/li\u003e\n\u003cli\u003eDOLL C, STEFFEN C, BECK-BROICHSITTER B, RICHTER M, NEUMANN K, POHRT A, et al. The Prognostic Significance of p16 and its Role as a Surrogate Marker for Human Papilloma Virus in Oral Squamous Cell Carcinoma: An Analysis of 281 Cases. Anticancer Research. 2022;42(5):2405-13.\u003c/li\u003e\n\u003cli\u003eSundberg J, Korytowska M, Burgos PM, Blomgren J, Blomstrand L, S DEL, et al. Combined Testing of p16 Tumour-suppressor Protein and Human Papillomavirus in Patients With Oral Leukoplakia and Oral Squamous Cell Carcinoma. Anticancer Res. 2019;39(3):1293-300.\u003c/li\u003e\n\u003cli\u003ePalve V, Bagwan J, Krishnan NM, Pareek M, Chandola U, Suresh A, et al. Detection of High-Risk Human Papillomavirus in Oral Cavity Squamous Cell Carcinoma Using Multiple Analytes and Their Role in Patient Survival. J Glob Oncol. 2018;4:1-33.\u003c/li\u003e\n\u003cli\u003eOliveira LR, Ribeiro-Silva A. Prognostic significance of immunohistochemical biomarkers in oral squamous cell carcinoma. International Journal of Oral and Maxillofacial Surgery. 2011;40(3):298-307.\u003c/li\u003e\n\u003cli\u003eMonteiro L, Mariano LC, Warnakulasuriya S. Podoplanin could be a predictive biomarker of the risk of patients with oral leukoplakia to develop oral cancer: A\u0026nbsp;systematic review and meta-analysis. Oral Diseases. 2024;30(2):207-15.\u003c/li\u003e\n\u003cli\u003eRetzbach EP, Sheehan SA, Nevel EM, Batra A, Phi T, Nguyen ATP, et al. Podoplanin emerges as a functionally relevant oral cancer biomarker and therapeutic target. Oral Oncology. 2018;78:126-36.\u003c/li\u003e\n\u003cli\u003eSwain N, Kumar SV, Routray S, Pathak J, Patel S. Podoplanin\u0026mdash;a novel marker in oral carcinogenesis. Tumor Biology. 2014;35(9):8407-13.\u003c/li\u003e\n\u003cli\u003eMonteiro L, do Amaral B, Delgado L, Garc\u0026ecirc;s F, Salazar F, Pacheco JJ, et al. Podoplanin Expression Independently and Jointly with Oral Epithelial Dysplasia Grade Acts as a Potential Biomarker of Malignant Transformation in Oral Leukoplakia. Biomolecules. 2022;12(5):606.\u003c/li\u003e\n\u003cli\u003eYuan P, Temam S, El-Naggar A, Zhou X, Liu DD, Lee JJ, et al. Overexpression of podoplanin in oral cancer and its association with poor clinical outcome. Cancer. 2006;107(3):563-9.\u003c/li\u003e\n\u003cli\u003eC\u0026Icirc;RLIGERIU L, CIMPEAN AM, RAICA M, DOROŞ CI. Dual Role of Podoplanin in Oral Cancer Development. In Vivo. 2014;28(3):341-7.\u003c/li\u003e\n\u003cli\u003eVargas AJ, Harris CC. Biomarker development in the precision medicine era: lung cancer as a case study. Nature Reviews Cancer. 2016;16(8):525-37.\u003c/li\u003e\n\u003cli\u003eAnstee QM, Castera L, Loomba R. Impact of non-invasive biomarkers on hepatology practice: Past, present and future. Journal of Hepatology. 2022;76(6):1362-78.\u003c/li\u003e\n\u003cli\u003eKamel HFM, Al-Amodi H. Exploitation of Gene Expression and Cancer Biomarkers in Paving the Path to Era of Personalized Medicine. Genomics Proteomics Bioinformatics. 2017;15(4):220-35.\u003c/li\u003e\n\u003cli\u003eSmith JC, Sheltzer JM. Genome-wide identification and analysis of prognostic features in human cancers. Cell Reports. 2022;38(13):110569.\u003c/li\u003e\n\u003cli\u003eLawrence MS, Sougnez C, Lichtenstein L, Cibulskis K, Lander E, Gabriel SB, et al. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature. 2015;517(7536):576-82.\u003c/li\u003e\n\u003cli\u003eKhan MM, Frustino J, Villa A, Nguyen BC, Woo SB, Johnson WE, et al. Total RNA sequencing reveals gene expression and microbial alterations shared by oral pre-malignant lesions and cancer. Hum Genomics. 2023;17(1):72.\u003c/li\u003e\n\u003cli\u003eChen C, M\u0026eacute;ndez E, Houck J, Fan W, Lohavanichbutr P, Doody D, et al. Gene expression profiling identifies genes predictive of oral squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 2008;17(8):2152-62.\u003c/li\u003e\n\u003cli\u003eSheu JJ, Lee CC, Hua CH, Li CI, Lai MT, Lee SC, et al. LRIG1 modulates aggressiveness of head and neck cancers by regulating EGFR-MAPK-SPHK1 signaling and extracellular matrix remodeling. Oncogene. 2014;33(11):1375-84.\u003c/li\u003e\n\u003cli\u003eVerduci L, Ferraiuolo M, Sacconi A, Ganci F, Vitale J, Colombo T, et al. The oncogenic role of circPVT1 in head and neck squamous cell carcinoma is mediated through the mutant p53/YAP/TEAD transcription-competent complex. Genome Biol. 2017;18(1):237.\u003c/li\u003e\n\u003cli\u003eYe H, Yu T, Temam S, Ziober BL, Wang J, Schwartz JL, et al. Transcriptomic dissection of tongue squamous cell carcinoma. BMC Genomics. 2008;9:69.\u003c/li\u003e\n\u003cli\u003eOghumu S, Knobloch TJ, Terrazas C, Varikuti S, Ahn-Jarvis J, Bollinger CE, et al. Deletion of macrophage migration inhibitory factor inhibits murine oral carcinogenesis: Potential role for chronic pro-inflammatory immune mediators. Int J Cancer. 2016;139(6):1379-90.\u003c/li\u003e\n\u003cli\u003eRentoft M, Coates PJ, Laurell G, Nylander K. Transcriptional profiling of formalin fixed paraffin embedded tissue: pitfalls and recommendations for identifying biologically relevant changes. PLoS One. 2012;7(4):e35276.\u003c/li\u003e\n\u003cli\u003eReis PP, Waldron L, Perez-Ordonez B, Pintilie M, Galloni NN, Xuan Y, et al. A gene signature in histologically normal surgical margins is predictive of oral carcinoma recurrence. BMC Cancer. 2011;11:437.\u003c/li\u003e\n\u003cli\u003eAmbatipudi S, Gerstung M, Pandey M, Samant T, Patil A, Kane S, et al. Genome-wide expression and copy number analysis identifies driver genes in gingivobuccal cancers. Genes Chromosomes Cancer. 2012;51(2):161-73.\u003c/li\u003e\n\u003cli\u003ePeng CH, Liao CT, Peng SC, Chen YJ, Cheng AJ, Juang JL, et al. A novel molecular signature identified by systems genetics approach predicts prognosis in oral squamous cell carcinoma. PLoS One. 2011;6(8):e23452.\u003c/li\u003e\n\u003cli\u003eLou Y, Ren L, Wang L. Hsa_circ_0049396 inhibited oral squamous cell carcinoma progression by regulating the miR-663b/ENDOU axis. Genomics. 2023;115(1):110553.\u003c/li\u003e\n\u003cli\u003eBasic V, Zhang B, Domert J, Pellas U, Tot T. Integrative meta-analysis of gene expression profiles identifies FEN1 and ENDOU as potential diagnostic biomarkers for cervical squamous cell carcinoma. Oncol Lett. 2021;22(6):840.\u003c/li\u003e\n\u003cli\u003eAlotaibi A, Gadekar VP, Gundla PS, Mandarthi S, Jayendra N, Tungekar A, et al. Global comparative transcriptomes uncover novel and population-specific gene expression in esophageal squamous cell carcinoma. Infect Agent Cancer. 2023;18(1):47.\u003c/li\u003e\n\u003cli\u003eInaba N, Renk T, Daume E, Bohn H. Ectopic production of placenta-\"specific\" tissue proteins (PP5 and PP11) by malignant breast tumors. Arch Gynecol. 1981;231(1):87-90.\u003c/li\u003e\n\u003cli\u003eInaba N, Ishige H, Ijichi M, Satoh N, Ohkawa R, Sekiya S, et al. Immunohistochemical detection of pregnancy-specific protein (SP1) and placenta-specific tissue proteins (PP5, PP10, PP11 and PP12) in ovarian adenocarcinomas. Oncodev Biol Med. 1982;3(5-6):379-89.\u003c/li\u003e\n\u003cli\u003eXu C, Zhang Y, Shen Y, Shi Y, Zhang M, Zhou L. Integrated Analysis Reveals ENDOU as a Biomarker in Head and Neck Squamous Cell Carcinoma Progression. Front Oncol. 2020;10:522332.\u003c/li\u003e\n\u003cli\u003eViechtbauer W. Conducting Meta-Analyses in R with the metafor Package. Journal of Statistical Software. 2010;36(3):1 - 48.\u003c/li\u003e\n\u003cli\u003eLove MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550.\u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research. 2015;43(7):e47-e.\u003c/li\u003e\n\u003cli\u003eColaprico A, Silva TC, Olsen C, Garofano L, Cava C, Garolini D, et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Res. 2016;44(8):e71.\u003c/li\u003e\n\u003cli\u003eWilkinson L. ggplot2: Elegant Graphics for Data Analysis by H. WICKHAM. Biometrics. 2011;67:678-9.\u003c/li\u003e\n\u003cli\u003eKassambara A. ggpubr: \u0026ldquo;ggplot2\u0026rdquo; Based Publication Ready Plots. R Package Version 0.6.0. 0.6.0 ed2023.\u003c/li\u003e\n\u003cli\u003eAhlmann-Eltze C, Patil I. ggsignif: R Package for Displaying Significance Brackets for 'ggplot2'. PsyArXiv; 2021.\u003c/li\u003e\n\u003cli\u003eDavis S, Meltzer PS. GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics. 2007;23(14):1846-7.\u003c/li\u003e\n\u003cli\u003eWu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation. 2021;2(3).\u003c/li\u003e\n\u003cli\u003eG Y. Enrichplot: Visualization of Functional Enrichment Result. R package version 1.24.0. 2024.\u003c/li\u003e\n\u003cli\u003eWalter W, S\u0026aacute;nchez-Cabo F, Ricote M. GOplot: an R package for visually combining expression data with functional analysis. Bioinformatics. 2015;31(17):2912-4.\u003c/li\u003e\n\u003cli\u003eH\u0026auml;nzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics. 2013;14(1):7.\u003c/li\u003e\n\u003cli\u003eSturm G, Finotello F, Petitprez F, Zhang JD, Baumbach J, Fridman WH, et al. Comprehensive evaluation of transcriptome-based cell-type quantification methods for immuno-oncology. Bioinformatics. 2019;35(14):i436-i45.\u003c/li\u003e\n\u003cli\u003eYoshihara K, Shahmoradgoli M, Mart\u0026iacute;nez E, Vegesna R, Kim H, Torres-Garcia W, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun. 2013;4:2612.\u003c/li\u003e\n\u003cli\u003eRobin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez J-C, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12(1):77.\u003c/li\u003e\n\u003cli\u003eBorgan \u0026Oslash;. Modeling Survival Data: Extending the Cox Model. Terry M. Therneau and Patricia M. Grambsch, Springer‐Verlag, New York, 2000. No. of pages: xiii + 350. Price: $69.95. ISBN 0‐387‐98784‐3. Statistics in Medicine. 2001;20(13):2053-4.\u003c/li\u003e\n\u003cli\u003eAltman DG, De Stavola BL. Practical problems in fitting a proportional hazards model to data with updated measurements of the covariates. Stat Med. 1994;13(4):301-41.\u003c/li\u003e\n\u003cli\u003eRai K, Prasad K, Sagar P, Rao RS, Vanishri CH, Gowri MS. Expression of podoplanin in predicting the biological behavior of oral squamous cell carcinoma \u0026ndash; A clinicopathological correlation. Journal of Advanced Clinical and Research Insights. 2019;6(6):163-70.\u003c/li\u003e\n\u003cli\u003eRickman DS, Beltran H, Demichelis F, Rubin MA. Biology and evolution of poorly differentiated neuroendocrine tumors. Nat Med. 2017;23(6):1-10.\u003c/li\u003e\n\u003cli\u003eBen-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, et al. An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet. 2008;40(5):499-507.\u003c/li\u003e\n\u003cli\u003eLiberzon A, Birger C, Thorvaldsd\u0026oacute;ttir H, Ghandi M, Mesirov JP, Tamayo P. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst. 2015;1(6):417-25.\u003c/li\u003e\n\u003cli\u003eNewman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nature Methods. 2015;12(5):453-7.\u003c/li\u003e\n\u003cli\u003eStojanov IJ, Woo S-B. Malignant Transformation Rate of Non-reactive Oral Hyperkeratoses Suggests an Early Dysplastic Phenotype. Head and Neck Pathology. 2022;16(2):366-74.\u003c/li\u003e\n\u003cli\u003eLin N-C, Hsu J-T, Tsai K-Y. Survival and clinicopathological characteristics of different histological grades of oral cavity squamous cell carcinoma: A single-center retrospective study. PLOS ONE. 2020;15(8):e0238103.\u003c/li\u003e\n\u003cli\u003ePoe JC, Kountikov EI, Lykken JM, Natarajan A, Marchuk DA, Tedder TF. EndoU is a novel regulator of AICD during peripheral B cell selection. Journal of Experimental Medicine. 2013;211(1):57-69.\u003c/li\u003e\n\u003cli\u003eJin Y, Qin X. Comprehensive Analysis of the Roles and Prognostic Value of RNA-Binding Proteins in Head and Neck Squamous Cell Carcinoma. DNA Cell Biol. 2020;39(10):1789-98.\u003c/li\u003e\n\u003cli\u003eLiu H, Ye Z, Wang X, Wu Y, Deng C. Comprehensive analysis of the functions, prognostic and diagnostic values of RNA binding proteins in head and neck squamous cell carcinoma. J Stomatol Oral Maxillofac Surg. 2024;125(4s):101937.\u003c/li\u003e\n\u003cli\u003eKreppel M, Kreppel B, Drebber U, Wedemayer I, Rothamel D, Z\u0026ouml;ller J, et al. Podoplanin expression in oral leukoplakia: prognostic value and clinicopathological implications. Oral Diseases. 2012;18(7):692-9.\u003c/li\u003e\n\u003cli\u003eQi W, Gromoff EDV, Xu F, Zhao Q, Yang W, Pfeifer D, et al. The secreted endoribonuclease ENDU-2 from the soma protects germline immortality in C. elegans. Nature Communications. 2021;12(1):1262.\u003c/li\u003e\n\u003cli\u003ePello OM, De Pizzol M, Mirolo M, Soucek L, Zammataro L, Amabile A, et al. Role of c-MYC in alternative activation of human macrophages and tumor-associated macrophage biology. Blood. 2012;119(2):411-21.\u003c/li\u003e\n\u003cli\u003eLeone G, Sears R, Huang E, Rempel R, Nuckolls F, Park CH, et al. Myc requires distinct E2F activities to induce S phase and apoptosis. Mol Cell. 2001;8(1):105-13.\u003c/li\u003e\n\u003cli\u003eWang C, Zhang J, Yin J, Gan Y, Xu S, Gu Y, et al. Alternative approaches to target Myc for cancer treatment. Signal Transduction and Targeted Therapy. 2021;6(1):117.\u003c/li\u003e\n\u003cli\u003eYoshida GJ. Emerging roles of Myc in stem cell biology and novel tumor therapies. Journal of Experimental \u0026amp; Clinical Cancer Research. 2018;37(1):173.\u003c/li\u003e\n\u003cli\u003eOshi M, Takahashi H, Tokumaru Y, Yan L, Rashid OM, Matsuyama R, et al. G2M Cell Cycle Pathway Score as a Prognostic Biomarker of Metastasis in Estrogen Receptor (ER)-Positive Breast Cancer. Int J Mol Sci. 2020;21(8).\u003c/li\u003e\n\u003cli\u003eAmit M, Takahashi H, Dragomir MP, Lindemann A, Gleber-Netto FO, Pickering CR, et al. Loss of p53 drives neuron reprogramming in head and neck cancer. Nature. 2020;578(7795):449-54.\u003c/li\u003e\n\u003cli\u003eBruno TC, Ebner PJ, Moore BL, Squalls OG, Waugh KA, Eruslanov EB, et al. Antigen-Presenting Intratumoral B Cells Affect CD4+ TIL Phenotypes in Non\u0026ndash;Small Cell Lung Cancer Patients. Cancer Immunology Research. 2017;5(10):898-907.\u003c/li\u003e\n\u003cli\u003eRodrigo JP, Garc\u0026iacute;a-Carracedo D, Gonz\u0026aacute;lez MV, Mancebo G, Fresno MF, Garc\u0026iacute;a-Pedrero J. Podoplanin expression in the development and progression of laryngeal squamous cell carcinomas. Mol Cancer. 2010;9:48.\u003c/li\u003e\n\u003cli\u003eLi J-R, Shaw V, Lin Y, Wang X, Aminu M, Li Y, et al. The prognostic effect of infiltrating immune cells is shaped by proximal M2 macrophages in lung adenocarcinoma. Molecular Cancer. 2024;23(1):185.\u003c/li\u003e\n\u003cli\u003eKuang D-M. Positive Feedback Loop between B cells/plasma cells and Macrophages Promote M2 Macrophage-Elicited Hepatoma Progression. The Journal of Immunology. 2016;196(1_Supplement):211.6-.6.\u003c/li\u003e\n\u003cli\u003eLadjevac N, Milovanovic M, Jevtovic A, Arsenijevic D, Stojanovic B, Dimitrijevic Stojanovic M, et al. The Role of IL-17 in the Pathogenesis of Oral Squamous Cell Carcinoma. International Journal of Molecular Sciences. 2023;24(12):9874.\u003c/li\u003e\n\u003cli\u003eZhao J, Chen X, Herjan T, Li X. The role of interleukin-17 in tumor development and progression. Journal of Experimental Medicine. 2019;217(1):e20190297.\u003c/li\u003e\n\u003cli\u003eGrandis JR, Drenning SD, Zeng Q, Watkins SC, Melhem MF, Endo S, et al. Constitutive activation of Stat3 signaling abrogates apoptosis in squamous cell carcinogenesis in vivo. Proc Natl Acad Sci U S A. 2000;97(8):4227-32.\u003c/li\u003e\n\u003cli\u003eTang HM, Kuay KT, Koh PF, Asad M, Tan TZ, Chung VY, et al. An epithelial marker promoter induction screen identifies histone deacetylase inhibitors to restore epithelial differentiation and abolishes anchorage independence growth in cancers. Cell Death Discovery. 2016;2(1):16041.\u003c/li\u003e\n\u003cli\u003eHasina R, Lingen MW. Angiogenesis in oral cancer. J Dent Educ. 2001;65(11):1282-90.\u003c/li\u003e\n\u003cli\u003ede Vicente JC, Rodrigo JP, Rodriguez-Santamarta T, Lequerica-Fern\u0026aacute;ndez P, Allonca E, Garc\u0026iacute;a-Pedrero JM. Podoplanin expression in oral leukoplakia: Tumorigenic role. Oral Oncology. 2013;49(6):598-603.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"oral cancer, oral potentially malignant disorders, OSCC, OPMDs, biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-6340150/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6340150/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOral squamous cell carcinoma (OSCC) is responsible for approximately 190,000 deaths each year worldwide, primarily due to late-stage diagnosis. The lack of clinically useful biomarkers has been identified as one of the crucial factors contributing to diagnostic delay in OSCC, limiting the impact of tumor biology on treatment decisions and leading to poor prognosis. We investigated the clinical utility of tissue staining for ENDOU protein, a recently identified tumor suppressor in oral mucosa, as a biomarker for early OSCC detection. Additionally, using publicly available datasets and versatile systems biology tools we dissected the molecular profiles of low ENDOU expressing tumors and precancerous lesions to elucidate the role of ENDOU in the disease progression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur results demonstrated consistent downregulation of ENDOU mRNA in OSCC in 11 independent discovery cohorts. In the validation cohorts, semi-quantitative immunohistochemical assessment revealed complete loss of ENDOU protein in 93% of tumors (p\u0026lt;0.001), which was independent of clinicopathological parameters such as TNM stage, tumor grade, age, or gender. The confusion matrix demonstrated high diagnostic accuracy of ENDOU staining in OSCC detection (0.93, 95% CI = 0.85-0.97; AUC = 0.95, 95% CI = 0.88-1.00, p\u0026lt;0.0001), which was significantly higher (p\u0026lt;0.05) compared to PDPN staining (0.76, 95% CI = 0.63-0.86), used as a reference biomarker in this study. Multivariate analysis demonstrated that ENDOU loss could serve as an independent adverse prognostic marker for OSCC (HR = 2.21, 95% CI = 1.22-4.01, p\u0026lt;0.01). Molecular profiling of low ENDOU expressing tumors and oral potentially malignant disorders (OPMDs) revealed its association with biological processes such as keratinocyte differentiation and immune response. Furthermore, low ENDOU expressing tumors exhibited hyperactivation of c-myc, and inhibition of p53 signaling pathways while in the precancerous lesions, ENDOU loss was associated with hyperactivation of early oncogenic pathways including JAK-STAT, angiogenesis and EMT.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur study highlights the potential of ENDOU as a diagnostic biomarker for early detection of OSCC and disease prognostication. Moreover, loss of ENDOU is associated with perturbation of vital biological processes and signaling pathways in the context of oral carcinogenesis. Further studies are warranted to validate these findings and explore their clinical implications.\u003c/p\u003e","manuscriptTitle":"ENDOU is potentially a robust diagnostic and prognostic biomarker in oral squamous cell carcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-05 13:23:47","doi":"10.21203/rs.3.rs-6340150/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"23b2e1a6-50d1-436b-9709-d0c4c01da306","owner":[],"postedDate":"November 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-05T13:23:47+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-05 13:23:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6340150","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6340150","identity":"rs-6340150","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}