Uncovering the proteogenomic landscape of head and neck squamous cell carcinoma through urine analysis

Uncovering the proteogenomic landscape of head and neck squamous cell carcinoma through urine analysis | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 20 March 2025 V1 Latest version Share on Uncovering the proteogenomic landscape of head and neck squamous cell carcinoma through urine analysis Authors : Oriana Barros , Joaquim Castro Silva , Eurico Monteiro , Susana Aveiro , Pedro Domingues , Pedro Valente Sousa , Carolina Castro , … Show All … , Catarina Rodrigues , António Barros , Francisco Amado , Saeid Ghavami , Vito Giuseppe D'Agostino , R Ferreira , and Lucio Lara Santos Show Fewer Authors Info & Affiliations https://doi.org/10.22541/au.174249381.13285606/v1 329 views 164 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Head and neck squamous cell carcinoma (HNSCC) is a major clinical challenge due to its aggressive nature and poor prognosis in advanced stages. Late detection, often due to delayed diagnosis, limits treatment success. This study investigates non-invasive diagnostic methods to identify early-stage molecular biomarkers using a proteogenomic approach. We analyzed urine samples from 19 male HNSCC patients and identified 1427 proteins by mass spectrometry. Of these, 730 overlapped with healthy controls, highlighting prognostic markers such as RNASE1, LRG1 and CD44. Machine learning techniques, including principal component analysis and partial least squares discriminant analysis, distinguished HNSCC patients from controls and revealed unique proteomic signatures. Pathogenic variants such as GAA p.(Trp746Cys) and SIAE p.(Pro210Leu) were found to be potential indicators of advanced disease. Functional analyzes linked the identified proteins to important tumor-related processes, including epithelial-mesenchymal transition and neutrophil degranulation. These results support urinary proteomics as a promising non-invasive diagnostic tool for early detection of HNSCC and disease monitoring. Future research should validate these biomarkers in larger, more diverse cohorts to improve clinical applicability. Uncovering the proteogenomic landscape of head and neck squamous cell carcinoma through urine analysis Running title: Proteogenomics in head and neck cancer Oriana Barros 1,2,3 , Joaquim Castro Silva 4 , Eurico Monteiro 4,5 , Susana Aveiro 6 , Pedro Domingues 6 , Pedro Valente Sousa 2 , Carolina Castro 2 , Catarina Rodrigues 2 , António Barros 7 , Francisco Amado 1,6 , Saeid Ghavami 9,10,11 , Vito G. D’Agostino 3 , Rita Ferreira 6 , Rui Vitorino 1,7 , Lúcio Lara Santos 2,8 1 iBiMED, Department of Medical Sciences, University of Aveiro, Aveiro, Portugal 2 Experimental Pathology and Therapeutics Group, Portuguese Institute of Oncology, Porto, Portugal 3 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy 4 Otorhinolaringology Service, Portuguese Institute for Oncology, Porto, Portugal 5 School of Medicine and Biomedical Sciences, University of Fernando Pessoa, Porto, Portugal 6 LAQV-REQUIMTE, Department of Chemistry, University of Aveiro, Aveiro, Portugal 7 RISE, Department of Surgery and Physiology, Faculty of Medicine, University of Porto, Porto, Portugal 8 Surgical Oncology Department, Portuguese Institute for Oncology, Porto, Portugal 9 Academy of Silesia, Faculty of Medicine, Rolna 43, 40-555 Katowice, Poland. 10 Paul Albrechtsen Research Institute, Cancer Care Manitoba, Winnipeg, MB R3E 0V9, Canada. 11 Department of Human Anatomy and Cell Science, Max Rady College of Medicine, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB R3E 0V9, Canada. Correspondence Rui Vitorino Department of Medical Sciences University of Aveiro 3810-193 Aveiro Portugal Email: [email protected] Lúcio Lara Santos Portuguese Institute of Oncology 4200-072 Porto Portugal Email: [email protected] Abstract Head and neck squamous cell carcinoma (HNSCC) is a major clinical challenge due to its aggressive nature and poor prognosis in advanced stages. Late detection, often due to delayed diagnosis, limits treatment success. This study investigates non-invasive diagnostic methods to identify early-stage molecular biomarkers using a proteogenomic approach. We analyzed urine samples from 19 male HNSCC patients and identified 1427 proteins by mass spectrometry. Of these, 730 overlapped with healthy controls, highlighting prognostic markers such as RNASE1, LRG1 and CD44. Machine learning techniques, including principal component analysis and partial least squares discriminant analysis, distinguished HNSCC patients from controls and revealed unique proteomic signatures. Pathogenic variants such as GAA p.(Trp746Cys) and SIAE p.(Pro210Leu) were found to be potential indicators of advanced disease. Functional analyzes linked the identified proteins to important tumor-related processes, including epithelial-mesenchymal transition and neutrophil degranulation. These results support urinary proteomics as a promising non-invasive diagnostic tool for early detection of HNSCC and disease monitoring. Future research should validate these biomarkers in larger, more diverse cohorts to improve clinical applicability. Keywords: head and neck cancer, urine proteome, variant, biomarker, early diagnosis Highlights • New HNSCC mutational signature composed of CP, GAA, SIAE, COL4A1, HSPG2, and EFEMP1 variants in urine • First identification of HNSCC-related pathogenic variants monitorable in urine • Identification of GAA p.(Thr806Met) variant in all cohort patients • SIAE p.(Pro210Leu) variant seems to be related to advanced disease stages Introduction Head and neck squamous cell carcinoma (HNSCC) is a malignant cancer whose incidence is increasing. This is a very heterogeneous cancer, mainly localized in the oral cavity, pharynx and larynx [1]. The main risk factors are alcohol consumption, smoking habits, infection with the human papillomavirus (HPV) or the Epstein-Barr virus (EBV). The treatment of HNSCC is very often multimodal, involving a combination of surgery, radiotherapy and/or chemotherapy. However, treatment of HNSCC in advanced stages of the disease is often ineffective. This reinforces the need to identify biomarkers to detect this disease as early as possible to increase treatment efficacy and improve survival outcomes in HNSCC [2]. Liquid biopsies are increasingly used to discover and validate HNSCC biomarkers. There are several biological fluids used as a source for liquid biopsies, the main ones being blood-derived fluids, saliva and urine [3]. In HNSCC, blood-derived fluids and saliva are the most commonly used fluids as a source of liquid biopsy. However, urine has several advantages, for example, it can be obtained in significant quantities by noninvasive approaches [4-6] and its contents are usually soluble and relatively stable. A significant percentage of the urinary content reflects what is found in plasma and can be used to investigate systemic disease [7-9]. Ferrari et al. performed a proteomics study on urine samples from HNSCC patients treated with boron neutron capture therapy (BCNT) [10]. The levels of GALS3BP, CD44 and OPN in the urine of HNSCC patients were significantly lower after BCNT treatment, suggesting their potential as biomarkers for monitoring treatment response [10]. Therefore, the urinary proteome can provide valuable information on the pathophysiology of HNSCC. More recently, proteogenomic approaches have been used in cancer studies. Proteogenomics allows the integration of proteomics data with mutation profiles, thus providing a comprehensive view of cancer-related molecular alterations. By combining proteomics with transcriptomics and genomics, this approach enables a more in-depth molecular characterization of HNSCC and provides a deeper assessment of how mutant proteins affect the phenotype and contribute to disease progression. However, to date, no study has investigated the urinary proteogenome in patients with HNSCC [11]. Therefore, this study seeks to perform a comprehensive characterization of the urinary proteogenome in HNSCC using a mass spectrometry (MS)-based proteogenomic approach. The aim is to identify HNSCC protein variants that are detectable in urine samples and may help in the early diagnosis of HNSCC. Methods 2.1 Recruitment of HNSCC patients and urine sample collection. Urine samples were collected from 19 male subjects aged between 50 and 71 years with HNSCC (Table S1) followed at the Instituto Português de Oncologia do Porto (IPO-Porto) between September 2018 and June 2019, and 10 healthy individuals aged between 51 and 60 years (PRIDE; PXD017902). Patients with the following eligibility criteria were included in the study: non-obese and non-diabetic; no history of cancer; with spinocellular tumors; no previous cancer treatment; able to be fully physically active. Urine samples were all collected in the morning. This study was approved by the Ethics Committee of the IPO-Porto. All experiments were conducted in accordance with the Declaration of Helsinki. All participants gave written informed consent before participating in the study. Data were anonymized for analysis. 2.2 GeLC-MS/MS analysis of urine samples The urine samples were centrifuged for 10 minutes at 4000x g at a temperature between 10 and 12 ºC. Approximately 800 µL of the supernatant was passed through a 10 KDa filter (Vivaspin, Sartorius) and the retentate was resuspended in resuspension buffer (10% SDS and 0.5 M Tris, pH 6.8). Subsequently, the protein content of the concentrated and desalted samples was determined with RC/DC TM Protein Assay Kit using bovine serum albumin as protein standard. SDS-PAGE was then performed with 50 µg of protein from each sample. In-gel digestion was performed, and tryptic peptides were analyzed using the QExactive Orbitrap (Thermo Fisher Scientific, Bremen, Germany) with the EASY-spray nano ESI source (Thermo Fisher Scientific, Bremen, Germany). The EASY-spray nano ESI source was coupled to an Ultimate 3000 high pressure liquid chromatography (HPLC) system (Dionex, Sunnyvale, CA). MaxQuant (version 1.6.9.0) and the Andromeda search engine were used for data analysis, applying standard search parameters, including a 1% false discovery rate (FDR) at the PSM, peptide, and protein levels. Spectra were aligned with Human protein sequences from the UniProt database (January 2023 release, containing 47,942 sequences). Mass tolerances for precursor and fragment ions were set to 4.5 ppm and 20 ppm, respectively. Enzyme specificity was set to cleavages at the C-terminus of arginine and lysine, with up to two failed cleavages possible. Oxidation of methionine and N-terminal acetylation were considered as variable modifications, while carbamidomethylation of cysteine was set as a fixed modification. A match between runs was enabled with a matching window of 0.7 minutes and an alignment window of 20 minutes. Reverse database hits and proteins with a Q-value above 0.01 were excluded. Proteins with less than three valid values in at least one group were also removed, and missing data were calculated using a normal distribution around the detection limit. Mass spectrometry proteomics data were deposited at the ProteomeXchange Consortium via the PRIDE repository with dataset identifier PXD057998. Exploratory analysis of the urinary proteome was performed using the “Statistic” module of MetaboAnalyst and a volcano plot was used to visualize proteins that were differentially expressed in HNSCC compared to healthy controls. Two dimensionality reduction techniques, Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA), were applied to reduce the model dimensions. Functional enrichment analysis was performed using a hypergeometric probability test to identify enriched functional categories. For these categories, proteins were subjected to multiple correction for p-values Further data processing was performed using PEAKS X PRO software (Bioinformatics Solutions Inc., Waterloo, Canada), with a precursor mass tolerance of 10 ppm and a fragment mass tolerance of 0.025 Da. Peptide FDR was controlled using the Decoy fusion method and limited to 1%. Peptides were considered differentially abundant if their expression ratio was ≥ 1.5 (overexpressed) or ≤ 0.67 (underexpressed), with refinement by a paired Student’s t-test (p < 0.05). 2.3 Characterization of urinary proteome in HNSCC The urinary proteome was analyzed to determine which proteins in the urine were associated with HNSCC. A Venn diagram was created with our urinary biomarkers and their encoding genes described in DisGeNET as correlated with HNSCC. Only the urinary biomarkers that were present in the DisGeNET HNSCC datasets were included in the urinary proteome analysis[12]. The differential expression of biomarkers of interest in the tumour was investigated in UALCAN[13]. UALCAN is a webtool that integrates level 3 RNA-seq data from TCGA on 33 different cancer types. This web tool displays data from various databases including TCGA, MET500, CPTAC and CBTTC. UALCAN is very useful for gene expression assessment and survival analysis. The encoding genes of the most abundant urinary proteins were characterized by univariate prognostic analyses using ToPP (http://www.biostatistics.online/topp/survival.php)15. ToPP is a computer tool that allows comprehensive analysis of prognosis by both univariate and multivariate analyses. For this study, The Cancer Genome Atlas (TCGA) dataset for HNSCC was selected and the ”best cutoff” was defined as the cutoff for the survival analyses. After performing univariate analyses, a multivariate analysis of these urinary biomarkers for HNSCC was performed to select the best combination of biomarkers for prognosis prediction and stratification of patients based on risk. A prognostic index (PI) was calculated for the selected biomarkers, PI= β1 × x1 + β2 × x2 + …… + βn × xn, where β is the multivariate regression coefficient of the corresponding mRNA and xn is the expression level of the corresponding mRNA. The Kaplan-Meyer method was used to plot the survival curve. The log-rank p-values and hazard ratios for each gene were observed to investigate independent prognostic factors. The statistical significance of prognostic prediction between two groups was evaluated using the log-rank test p < 0.05. Identification of Cancer-Associated Mutations Given the significant impact of protein variants on cancer progression, the proteogenomic profile of urine from HNSCC patients was analyzed. The mass spectra were compared with the Cancer Proteome Variation Database (CanProVar) 2.016. CanProvar integrates information from 8570 cancer-related variations in 2921 proteins and allows to characterize these variations from a functional point of view. To select urinary proteins that are highly likely to originate from HNSCC tissue, only protein variants present in at least four samples and known to be expressed in HNSCC were considered. VarSome (https://varsome.com) was used to investigate the impact of the variants identified with CanProVar17. This database integrates information from more than 140 public genomics-related databases and automatically classifies variants according to the American College of Medical Genetics and Genomics (ACMG) guidelines. Classification of variants according to the ACMG guidelines alone can be subjective. This tool can be used in complementarity to the ACMG guidelines to double-check the classification of variants or when the classification of variants according to the ACMG guidelines is unclear[14]. The DriverDBv4 database was used to identify which driver genes of HNSCC encode proteins that may be present in urine from a proteogenomic perspective. DriverDBv4 enables the investigation of driver genes associated with different cancer types and summarizes the major cancer driver databases, namely IntOGen, OncoVar and the previous version of this web server DriverDBv3[15]. Validation of HNC-associated proteins and protein variants To validate the tumor origin of the urinary proteins identified as the most promising HNSCC biomarker candidates, ProteomicsDB was used[16]. ProteomicsDB displays the expression of each protein in a variety of tissues, biological fluids, and cell lines. In this work, the median protein expression of our urinary protein candidates was evaluated in various normal tissues associated with this disease (salivary gland, tonsil, oral epithelium, tongue, and nasopharynx). The expression of the biomarkers selected in HNSCC in relation to healthy individuals was validated using UALCAN[13]. In the “Proteomics” module, it was possible to extract data from CPTAC samples of HNSCC. The spectral count ratio values obtained from CPTAC were normalized within each sample profile and across samples. Gene enrichment analysis The gene enrichment analysis was carried out with the BEST tool using the ”Gene Enrichment Analysis” module. This tool makes it possible to establish a correlation between the main biological processes associated with the biomarkers of interest, considering the different available datasets and a specific cancer type using Pearson correlation. The enriched genes evaluated encode the most abundant proteins and protein variants identified in urine proteome and proteogenome profiling of HNSCC patients. The relationship between the encoding genes of our urinary biomarkers of interest was analyzed using STRING[17] or STITCH[18]. Results 3.1 Baseline characteristics of our patient cohort Nineteen male patients with HNSCC were included in our exploratory study. The descriptive statistics for clinical variables are shown in Table 1. A total of 15 (73.7%) patients had construction-related professions, 13 (68.4%) were taking medications, 6 (31.6%) were smokers, 13 (68.4%) were former smokers, and 13 (68.4%) were heavy drinkers. The distribution of tumor sites showed that 2 (10.5%) were in the oral cavity, 3 (15.8%) in the larynx, 13 (68.4%) in the pharynx, and 1 (5.3%) in both the larynx and pharynx. The tumor stage (T) revealed that 4 (26.3%) were in stages 1-2 and 15 (73.7%) were in stages 3-4. As for tumor stage (N), 7 (36.8%) were in stages 0-1 and 12 (63.2%) were in stages 2-3. Tumor stage (M) data showed that 15 (78.9%) had no metastases (stage 0) and 3 (15.8%) had metastases. 3.2 Machine learning techniques showed a clear separation between urinary proteome from HNSCC patients in comparison to healthy individuals A total of 1427 proteins were identified (1260 proteins in HNSCC patients vs. 897 proteins in healthy individuals), using 2 peptides as criteria for ID (Table S2), of which 730 proteins were shared between HNSCC patients and healthy individuals, as shown in the Venn diagram in Figure 1A. The most differentially expressed proteins in HNSCC patients were RNASE1, LRG1, CD55, GM2A, FBN1, CD59, ORM1 and ORM2, APOD, and CD44, as shown in the volcano plot in Figure 1B. DisGeNET was used to verify which of these urinary proteins had been previously associated with HNSCC. CD44 was the only differentially overrepresented protein in the urine known to be associated with HNSCC and to have a significant impact on prognosis (Figure S1). PCA and PLS-DA were used to assess the differences in the proteome between the two groups. PCA and PLS analyses revealed a separation of clusters of HNSCC patients compared to healthy individuals, as shown in Figure 1C and Figure 1D. The clusters showed separation along the first component axis (PC1), which explained 34.1% of the variance in the PCA and 33.5% of the variance in the PLS. We then performed a functional analysis of the urinary proteins most differentially expressed in HNSCC patients compared to healthy individuals. The PPI network was organized into clusters, and three distinct clusters were identified. The urinary biomarkers forming each cluster were analyzed from a functional point of view using the bioinformatics tools STITCH and BEST. We found three different clusters (clusters A, B and C; Figure S2). Cluster A was formed by 80 proteins and was found to be enriched in the regulation of CDH11, the regulation of type II cadherins and the metabolism of hyaluronan. Cluster B consisted of 17 proteins related to the metabolism of angiotensinogen to angiotensins and the metabolism of peptide hormones. Both clusters A and B were enriched in proteins involved in focal adhesion, epithelial-mesenchymal transition and neutrophil degranulation. Cluster C contained only 2 proteins that were mainly enriched in cytokine signalling and MAPK1/MAPK3 activation. Eight different signalling pathways were common to the three clusters, namely KRAS upward signalling, TNF-α signalling via NFKB, IL2-STAT5 signaling, UV response dn, complement and coagulation cascades, myogenesis, inflammatory response, and viral myocarditis. 3.3 Three novel pathogenic variants were identified in HNSCC patients using urine as source of liquid biospy The urinary proteogenome of HNSCC patients was analyzed. Only protein variants listed in the CanProVar database that were observed in at least four patients and absent in all healthy participants were included in the study. After applying this first filter, 517 mutant peptides and 284 mutant protein isoforms were identified (Table S3 and Table S4). The 284 mutant protein isoforms identified in this study were associated with 168 encoding genes. To ensure that only mutant protein isoforms associated with HNSCC were included in the study, a Venn diagram was created by crossing our dataset with information from CanProVar and DriverDBv4 (Figure 2). Ten HNSCC driver genes encoding mutant protein isoforms present in urine were identified (CDH1, COL4A1, COL6A1, CUBN, FCGBP, FN1, HSPG2, PDGFRA, REG1A AND SIRPB1). In addition, 31 genes that have been shown to be associated with HNSCC and encode our mutant protein isoforms in urine (ANPEP, CNDP1, CP, DMAP1, EFEMP1, CP, EGF, ENDOD1, FGA, FGB, FGG, FOLR1, GAA, GC, GM2A, HEXA, ITIH4, KRT16, KRT9, LRG1, LRP2, LTF, PAPPA2, PIGR, PRSS8, QSOX1, SAMM50, SERPINB5, SIAE, SUSD2, TYRO3 and VTN) were identified. Data on the mutant protein isoforms associated with this panel of 41 genes were aligned with Varsome. After analyzing the variants associated with the mutant protein isoforms, nine variants were identified in the urinary proteogenome of HNSCC, which were classified as variants of uncertain significance (VUS) and pathogenic variants, described in Table 2. The VUS associated with HNSCC driver genes were COL4A p.(Leu1467Phe) and HSPG2 p.(Glu966Lys). The VUS identified in genes known to be associated with HNSCC were CP p.(Gly895Ala); EFEMP1 p.(Asp175Asn); GAA p.(Thr806Met) and GAA p.(Val19Met). The pathogenic variants identified in the urinary proteogenome were GAA p.(Trp746Cys); SIAE p.(Pro210Leu) and SIAE p.(Cys196Phe). The variants GAA p.(Val19Met) and SIAE p.(Pro210Leu) were present in patients with stage III or IV disease, as shown in Figure 3. These mutations were shown to be potential discriminators of disease stage. The COL4A1 p.(Leu146Phe) VUS was identified in only four patients. Of these four patients, three had advanced stage HNSCC (stage III or IV). The EFEMP1 p.(Asp175Asn) and GAA p.(Thr806Met) VUS were identified in all patients in our cohort. Analysis of clinical data revealed that all patients had or still had smoking and alcohol habits. By analyzing the proteomic profile of the proteins encoded by these genes, we observed that only the EFEMP1 protein was overexpressed in HNSCC compared to healthy patients (Figure S3). COL4A1, CP, HSPG2 and SIAE proteins were underexpressed in HNSCC. The results for GAA in CPTAC were not statistically significant. To better understand the molecular mechanisms involving the genes related to the most impactful variants in HNSCC, the subcellular location, protein function, and physiological effects of the protein variants were analyzed (Figure 4). CP, GAA and SIAE play a metabolic role. CP is a serum ferroxidase responsible for the peroxidation of Fe(II)-transferrin to Fe(III)-transferrin. GAA encodes for the enzyme alpha-glucosidase, which is important for the breakdown of glycogen in lysosomes. SIAE encodes for a serine esterase enzyme that removes 9-0-acetyl ester groups from sialic acids. COL4A1, HSPG2 and EFEMP1 have been shown to be critical for the regulation of extracellular matrix and focal adhesion. When the PPI network was analyzed, the most abundant proteins from urinary proteogenomic analysis were found to be highly linked to COL4A1, HSPG2 and CP. Gene enrichment analysis was performed for CD44 and the mutant protein isoforms in urine associated with the most promising variants in HNSCC (CP, SIAE, GAA, COL4A1, HSPG2 and EFEMP1) using the BEST tool (Figure S4). The main biological processes associated with this panel of urinary biomarkers were epithelial-mesenchymal transition, focal adhesion, external encapsulating structure organization, collagen fibril organization and collagen metabolic process. Discussion Here we report a proteogenomic characterization of HNSCC using urine as the source of liquid biopsy. We measured two levels of HNSCC biology, namely the proteome and the proteogenome. By analyzing the proteogenome landscape of HNSCC, we identified novel variants that may be related to HNSCC pathogenesis and provide a roadmap for future studies. In addition, the correlation between protein expression and survival was investigated. When analyzing the urinary proteome in the context of HNSCC, CD44 was found to be the most important protein in stratifying HNSCC prognosis, similar to other cancers [19]. CD44 has previously been associated with poorer T-category, N-category, tumor grade and prognosis in oral, pharyngeal, and laryngeal cancer [20-22]. Ferrary et al. applied a MS-based proteomic approach to urine samples from HNSCC [10]. Galectin-3 binding protein, CD44 and osteopontin were the urinary proteins most involved in inflammation. These glycoproteins have been shown to be related to inflammatory processes and probably to tumor progression. These findings consolidate the role of CD44 as a urinary prognostic biomarker for HNSCC. Analysis of the urinary proteogenome revealed that six genes (COL4A1, HSPG2, CP, EFEMP1, GAA and SIAE) are associated with nine variants that show a strong association with HNSCC. Of these nine variants, six were registered in the Varsome database as VUS variants and three as pathogenic variants. COL4A1 encodes the major type IV collagen chain, an essential component of the ECM and basement membrane. COL4A1 forms a strong network with several ECM elements, namely other collagens (COL4A), LAMC2 and TGFB1. Xiaobeitan et al. demonstrated that knockout of COL4A1 significantly reduced the processes of migration and tumor invasion in oral squamous cell carcinoma (OSCC) cell lines[23]. The COL4A1 p.(Leu1467Phe) VUS was identified by urinary proteogenome analysis. This analysis also identified a VUS of HSPG2 [HSPG2 p.(Glu966Lys)], also known as perlecan. This is one of the largest molecules in the ECM. Perlecan binds to a variety of molecules, namely collagen III, elastin, laminin, fibrillin-1, fibronectin, nidogen-1 and nidogen-2, α-dystroglycan, integrins, ECM1, ephrin B3. This molecule is involved in the interactions between cell-cell and cell-ECM adhesion molecules, which in turn are associated with carcinogenesis and tumor invasion. One of the histological changes of oral epithelial dysplasia or carcinoma in situ of the oral mucosa, as defined by the World Health Organization classification, is “loss of intercellular adherence”, which supports the presence of perlecan in the epithelial cells of the oral mucosa [24-26]. Kawahara et al. knocked down perlecan in OSCC and observed a very significant reduction in cell adhesion and migration processes as well as cell resistance to cisplatin [27]. We identified a VUS of EFEMP1 [EFEMP1 p.(Asp175Asn)] in urine samples from all HNSCC patients. EFEMP1, also known as fibullin-3, is a glycoprotein secreted by the ECM. EFEMP1 modulates the activity of various MMPs and tissue inhibitors. EFEMP1 exerts its role in OSCC by regulating Wnt/β-catenin signaling. Tiam1 interacts with EFEMP1 and promotes the activation of MMP7, which in turn leads to the induction of EMT in OSCC [28]. CP is a glycoprotein that acts as an endogenous antioxidant. It enables the incorporation of iron into transferrin without the formation of toxic products and regulates cell membrane oxidation [29-32]. Shah et al. have shown that serum CP levels can be a predictor of malignant transformation and thus has the potential to be used as a biomarker for potentially malignant oral lesions and OSCC [33]. A VUS of CP [CP p.(Gly895Ala)] was identified in urine samples from HNSCC patients. Proteogenomic analysis of urine also identified two pathogenic variants, SIAE p.(Pro210Leu) and SIAE p.(Cys196Phe). It is noteworthy that the SIAE p.(Pro210Leu) variant was associated with more advanced stages of the disease. SIAE is responsible for the deacetylation of 9-O-acetylated sialic acid residues and plays a key role in cell-cell and cell-ECM interactions. Vajaria et al. showed that SIAE could be a potential predictor for early OSCC diagnosis [34]. Finally, in our study, we identified two VUS [GAA p.(Thr806Met); GAA p.(Val19Met)] and a pathogenic variant [GAA p.(Trp746Cys)] in α-glucosidase detected in the urine of HNSCC. One of the VUS, GAA p.(Thr806Met), was identified in all patients in our cohort, while the other VUS, GAA p.(Val19Met), was associated with disease stage. However, the specific role of this enzyme in HNSCC is not yet known. So far, the contribution of urinary proteomics to the understanding of the pathophysiological mechanisms of HNSCC has been negligible. Our work has shown that it is possible to assess the main physiopathological mechanisms underlying HNSCC by analyzing the urinary proteome of patients. Moreover, this study is the first to identify pathogenic variants associated with HNSCC that can be monitored using urine samples from patients. However, there are some limitations that should be considered. One limitation is the sample size and gender of our study cohort. As this is a pilot study, the sample size allowed us to identify a panel of biomarkers of interest. Larger cohorts will also help to clarify the influence of different parameters (e.g. age, gender, medication, occupation, TNM stage) on the identified urinary biomarkers and to what extent. The urinary proteome is highly susceptible to variations caused by diet, water intake, metabolic processes and circadian rhythm. Inter-individual variability in the urinary proteome poses a challenge, making comparisons with a single, small-sized control group a limitation [35]. To address these limitations in future studies, several measures can be implemented. For example, standardizing the collection and processing of samples through established protocols is essential. Another effective approach could be the inclusion of a control group matched for dietary and lifestyle factors. However, pilot studies are crucial to make the necessary methodological adjustments to optimize large-scale studies. 5 Conclusion In conclusion, this study underscores the significant potential of urinary proteogenomics in the early detection and management of HNSCC. Our findings reveal a distinct proteomic signature in the urine of HNSCC patients, highlighting the feasibility of non-invasive liquid biopsies as a diagnostic tool for identifying disease biomarkers. Notably, CD44 emerged as a critical urinary protein with substantial prognostic value, reinforcing its role in the stratification of HNSCC prognosis, similar to its documented impact in other malignancies. The identification of a novel mutational signature involving variants in CP, GAA, SIAE, COL4A1, HSPG2, and EFEMP1 genes offers new insights into the molecular underpinnings of HNSCC pathogenesis. Particularly, the pathogenic variants in GAA and SIAE detected in the urine exemplify the potential for such biomarkers to not only enhance our understanding of the disease mechanisms but also to serve as indicators of disease severity. The association of the SIAE p.(Pro210Leu) variant with advanced stages of HNSCC is especially compelling, suggesting its utility in disease stage determination and possibly in guiding therapeutic decisions. Future directions of this research needs to be focused on the validation of these findings through larger, multi-center studies that encompass a broader demographic and genetic diversity to substantiate the clinical applicability of these biomarkers. Such studies will help to confirm the reliability and reproducibility of the urinary proteomic signatures and their correlation with disease progression and patient outcomes. In addition to validating this panel of biomarkers through larger-scale studies, there is potential to explore additional proteomic and genomic markers. Furthermore, integrating these biomarkers into existing clinical workflows could revolutionize the early detection and prognostic assessment of HNSCC. Developing standardized, cost-effective assays (e.g. point-of-care devices) for these urinary proteins and genetic variants could facilitate their adoption in routine clinical practice, offering a non-invasive alternative to current diagnostic methods. This would not only enhance patient compliance but also potentially improve survival rates by enabling earlier and more precise treatment interventions. Additionally, the translational application of this research holds promise for personalized medicine strategies in oncology. By tailoring treatment plans based on specific proteomic and genomic profiles identified through urine analysis, clinicians could better target the underlying mechanisms of the disease in individual patients, optimizing therapeutic efficacy and minimizing adverse effects. Considering that several molecular pathways are common to different types of cancer, in the future it would be worthwhile to understand the applicability of these urinary biomarkers of HNSCC in other types of cancer. Overall, the urinary proteome and proteogenome offer a rich resource for uncovering novel biomarkers and pathogenic drivers of HNSCC. Our study lays the groundwork for future research and clinical trials aimed at harnessing these findings to improve diagnostic accuracy, prognostic assessment, and personalized treatment approaches in HNSCC, ultimately enhancing patient care and outcomes in this challenging disease domain. CRediT authorship contribution statement Conceptualization, O.B., R.F., R.V., L.L.S.; formal analysis, O.B, R.F., R.V., L.L.S.; funding acquisition, R.V. and L.L.S.; investigation, O.B., J.C.S., E.M., A.B., F.A., S.G., R.F., R.V. and L.L.S.; methodology, O.B., S.A., P.D., P.V.S., C.C., C.R., R.V.; writing—original draft, O.B.; writing—review and editing, R.V., R.F., V.A. and L.L.S; supervision, R.V., R.F., V.A. and L.S. All authors have read and agreed to the published version of the manuscript. Acknowledgements This work was supported by the Portuguese Foundation for Science and Technology (FCT), European Union, QREN, FEDER, and COMPETE for funding Institute of Biomedicine (iBiMED; UIDB/04501/2020, POCI-01-0145-FEDER-007628), LAQV-REQUIMTE (UIDB/50006/2020 and UIDP/50006/2020), Instituto Português de Oncologia do Porto Francisco Gentil, EPE (CI-IPOP-134-2020), COST Action INTERCEPTOR and by an individual scholarship FCT-UA-ECIU from Oriana Barros (University of Aveiro). Declaration of Competing Interest The authors declare that there are no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Ethics statement The study was conducted following the Declaration of Helsinki. All procedures concerning the inclusion of patients were approved by the institutional Ethics Committee (Comissão de ética para a saúde—CES-IPOFG-EPE-CES-194/022) after patients’ informed consent. Data availability statement Mass spectrometry proteomics data were deposited at the ProteomeXchange Consortium via the PRIDE repository with dataset identifier PXD057998. Declaration of Generative AI and AI-assisted technologies in the writing process The authors state that they did not use AI or AI-assisted technology in this writing process. References [1] Johnson, D. E., Burtness, B., Leemans, C. R., Lui, V. W. Y. , et al. , Head and neck squamous cell carcinoma. Nature reviews. Disease primers 2020, 6 , 92.[2] González-Moles, M., Aguilar-Ruiz, M., Ramos-García, P., Challenges in the Early Diagnosis of Oral Cancer, Evidence Gaps and Strategies for Improvement: A Scoping Review of Systematic Reviews. Cancers 2022, 14 .[3] Thongboonkerd, V., Malasit, P., Renal and urinary proteomics: current applications and challenges. Proteomics 2005, 5 , 1033-1042.[4] Fan, G., Gong, T., Lin, Y., Wang, J. , et al. , Urine proteomics identifies biomarkers for diabetic kidney disease at different stages. Clinical proteomics 2021, 18 , 32.[5] Chen, Z., Kim, J., Urinary proteomics and metabolomics studies to monitor bladder health and urological diseases. BMC Urol 2016, 16 , 11.[6] Pejcic, M., Stojnev, S., Stefanovic, V., Urinary proteomics–a tool for biomarker discovery. Ren Fail 2010, 32 , 259-268.[7] Omenn, G. S., States, D. J., Adamski, M., Blackwell, T. W. , et al. , Overview of the HUPO Plasma Proteome Project: results from the pilot phase with 35 collaborating laboratories and multiple analytical groups, generating a core dataset of 3020 proteins and a publicly-available database. Proteomics 2005, 5 , 3226-3245.[8] Kolch, W., Neusüss, C., Pelzing, M., Mischak, H., Capillary electrophoresis-mass spectrometry as a powerful tool in clinical diagnosis and biomarker discovery. Mass spectrometry reviews 2005, 24 , 959-977.[9] Theodorescu, D., Wittke, S., Ross, M. M., Walden, M. , et al. , Discovery and validation of new protein biomarkers for urothelial cancer: a prospective analysis. Lancet Oncol 2006, 7 , 230-240.[10] Ferrari, E., Wittig, A., Basilico, F., Rossi, R. , et al. , Urinary Proteomics Profiles Are Useful for Detection of Cancer Biomarkers and Changes Induced by Therapeutic Procedures. Molecules 2019, 24 .[11] Rodriguez, H., Zenklusen, J. C., Staudt, L. M., Doroshow, J. H., Lowy, D. R., The next horizon in precision oncology: Proteogenomics to inform cancer diagnosis and treatment. Cell 2021, 184 , 1661-1670.[12] Piñero, J., Ramírez-Anguita, J. M., Saüch-Pitarch, J., Ronzano, F. , et al. , The DisGeNET knowledge platform for disease genomics: 2019 update. Nucleic acids research 2020, 48 , D845-d855.[13] Chandrashekar, D. S., Karthikeyan, S. K., Korla, P. K., Patel, H. , et al. , UALCAN: An update to the integrated cancer data analysis platform. Neoplasia 2022, 25 , 18-27.[14] Kopanos, C., Tsiolkas, V., Kouris, A., Chapple, C. E. , et al. , VarSome: the human genomic variant search engine. Bioinformatics 2019, 35 , 1978-1980.[15] Liu, C. H., Lai, Y. L., Shen, P. C., Liu, H. C. , et al. , DriverDBv4: a multi-omics integration database for cancer driver gene research. Nucleic acids research 2024, 52 , D1246-d1252.[16] Samaras, P., Schmidt, T., Frejno, M., Gessulat, S. , et al. , ProteomicsDB: a multi-omics and multi-organism resource for life science research. Nucleic acids research 2020, 48 , D1153-d1163.[17] Szklarczyk, D., Kirsch, R., Koutrouli, M., Nastou, K. , et al. , The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic acids research 2023, 51 , D638-d646.[18] Kuhn, M., von Mering, C., Campillos, M., Jensen, L. J., Bork, P., STITCH: interaction networks of chemicals and proteins. Nucleic acids research 2008, 36 , D684-688.[19] Gaiteiro, C., Soares, J., Relvas-Santos, M., Peixoto, A. , et al. , Glycoproteogenomics characterizes the CD44 splicing code associated with bladder cancer invasion. Theranostics 2022, 12 , 3150-3177.[20] Chen, J., Zhou, J., Lu, J., Xiong, H. , et al. , Significance of CD44 expression in head and neck cancer: a systemic review and meta-analysis. BMC cancer 2014, 14 , 15.[21] Mirhashemi, M., Sadeghi, M., Ghazi, N., Saghravanian, N. , et al. , Prognostic value of CD44 expression in oral squamous cell carcinoma: A meta-analysis. Ann Diagn Pathol 2023, 67 , 152213.[22] Chai, L., Liu, H., Zhang, Z., Wang, F. , et al. , CD44 expression is predictive of poor prognosis in pharyngolaryngeal cancer: systematic review and meta-analysis. Tohoku J Exp Med 2014, 232 , 9-19.[23] Tian, X., Sun, J., Li, C., Zhang, K., COL4A1 promotes the proliferation and migration of oral squamous cell carcinoma cells by binding to NID1. Exp Ther Med 2023, 25 , 176.[24] Ikarashi, T., Ida-Yonemochi, H., Ohshiro, K., Cheng, J., Saku, T., Intraepithelial expression of perlecan, a basement membrane-type heparan sulfate proteoglycan reflects dysplastic changes of the oral mucosal epithelium. Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology 2004, 33 , 87-95.[25] Maruyama, S., Shimazu, Y., Kudo, T., Sato, K. , et al. , Three-dimensional visualization of perlecan-rich neoplastic stroma induced concurrently with the invasion of oral squamous cell carcinoma. Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology 2014, 43 , 627-636.[26] Farach-Carson, M. C., Warren, C. R., Harrington, D. A., Carson, D. D., Border patrol: insights into the unique role of perlecan/heparan sulfate proteoglycan 2 at cell and tissue borders. Matrix biology : journal of the International Society for Matrix Biology 2014, 34 , 64-79.[27] Kawahara, R., Granato, D. C., Carnielli, C. M., Cervigne, N. K. , et al. , Agrin and perlecan mediate tumorigenic processes in oral squamous cell carcinoma. PloS one 2014, 9 , e115004.[28] Guo, H., Ji, Y., Zhang, B., Huang, X., Fibulin-3 sponges Tiam1 to manipulate MMP-7 activity through β-catenin signaling in oral squamous cell carcinoma. Medical oncology 2022, 39 , 154.[29] Patil, M. B., Lavanya, T., Kumari, C. M., Shetty, S. R. , et al. , Serum ceruloplasmin as cancer marker in oral pre-cancers and cancers. J Carcinog 2021, 20 , 15.[30] Mampilly, M. O., Ravindran, N., Parambil, M. S., Nilesh, K. , et al. , Assessment of Serum Selenium and Ceruloplasmin in Potentially Malignant Disorders and Oral Cancer. J Pharm Bioallied Sci 2021, 13 , S989-s992.[31] Taysi, S., Akcay, F., Uslu, C., Dogru, Y., Gulcin, I., Trace elements and some extracellular antioxidant protein levels in serum of patients with laryngeal cancer. Biol Trace Elem Res 2003, 91 , 11-18.[32] Rasheed, M. H., Beevi, S. S., Rajaraman, R., Bose, S. J., Alleviation of oxidative and nitrosative stress following curative resection in patient with oral cavity cancer. J Surg Oncol 2007, 96 , 194-199.[33] Shah, P. H., Venkatesh, R., More, C. B., Determination of role of ceruloplasmin in oral potentially malignant disorders and oral malignancy-A cross-sectional study. Oral diseases 2017, 23 , 1066-1071.[34] Vajaria, B. N., Patel, K. R., Begum, R., Shah, F. D. , et al. , Evaluation of serum and salivary total sialic acid and α-l-fucosidase in patients with oral precancerous conditions and oral cancer. Oral surgery, oral medicine, oral pathology and oral radiology 2013, 115 , 764-771.[35] Shao, C., Zhao, M., Chen, X., Sun, H. , et al. , Comprehensive Analysis of Individual Variation in the Urinary Proteome Revealed Significant Gender Differences. Molecular & cellular proteomics : MCP 2019, 18 , 1110-1122. Figure captions Figure 1: Proteomic analysis of urine samples from HNSCC patients and healthy individuals using MetaboAnalyst v6.0. Venn Diagram intersecting identified proteins between HNSCC patients and healthy individuals (A). Volcano Plot representing the most abundant urinary proteins in HSNCC patients in comparison to healthy individuals (B). Principal Component Analysis of the urinary proteome identified in HNSCC patients in comparison to healthy individuals (C). Partial Least Squares of the urinary proteome identified in HNSCC patients in comparison to healthy individuals (D). Figure 2: Integration of proteogenomic data extracted from urine samples of HNSCC patients. The expression profile in normal tissue (N) of the panel of genes encoding our urinary proteins was extracted from ProteomicsDB (boxes in orange). The expression of this panel of biomarkers in relation to healthy individuals and the relationship of their expression with HPV and TP53 mutational status was studied using UALCAN. Survival data, in particular overall survival (OS), was extracted from the ToPP database. The color gradient from least intense to most intense is proportional to the expression of our biomarkers. Only statistically significant values (p < 0.05) that are overexpressed are shown in the figure. The genes that appear in blue are the ones under-expressed in the tumor compared to healthy individuals. KRT16 is identified with an asterisk (*) to highlight the fact that its expression is increased in the tumor by more than 3000 transcripts per million compared to healthy individuals. Venn Diagram intersecting variants identified in urine samples of HNSCC (UrineProteo), driver genes (DriverDBv4) of HNSCC and genes with an established association with HNSCC (CanProVar) is presented in the top corner of the figure. Legend: VUS, variant of uncertain significance. Figure 3: Distribution of the selected mutant protein isoforms among our cohort. The numbers presented in the first row correspond to the patients (Table S1) and on the first column the mutations are described in HGVS nomenclature. Patients expressing a given mutation are marked in orange and patients not expressing a given mutation are marked in white. Figure 4: Characterization of the subcellular location, protein function and protein-protein interaction (PPI) network using Human Protein Atlas and STRING. The network include CD44 and the 6 urinary proteins with mutations related to variants of uncertain significance (VUS) and pathogenic variants (CP, SIAE, GAA, COL4A1, HSPG2 and EFEMP1). The network nodes are proteins. The edges correspond to predicted functional associations and the lines show the existence or not of interaction between the proteins of interest. Certain elements of this figure were designed with BioRender. Figure S1: Survival analysis of CD44 in HNSCC in ToPP. Overall survival of CD44 A). Progression-free survival of CD44 (B). Disease-specific survival of CD44 (C). Disease-free survival of CD44 (D). Relapse-free survival of CD44 (E). In survival analyses, all patients were divided into high and low risk groups according to the median of risk score. Black data points represent patients whose HNSCC tumors had a gene expression below the median level. Red data points represent patients whose HNSCC tumors had a gene expression above the median level. All results were statistically significant (log-rank test, p < 0.05). Figure S2: PPI network and gene enrichment analysis of the urinary biomarkers of HNSCC. PPI network and clustering of the urinary proteins identified in urine samples of HNSCC patients in STITCH. The detailed information of enriched signalling pathways extracted from BEST tool for each cluster is presented in colourful boxes. Green colour represents Cluster A, pink colour represents Cluster B, and blue colour represents Cluster C. Only statistically significant (p < 0.05) enriched signalling pathways were considered. Figure S3: Proteomic profile of our biomarkers in HNSCC using UALCAN. Protein expression of COL4A1 (A), CP (B) and EFEMP1 (C), GAA (D), HSPG2 (E) and SIAE (F) in HNSCC. Z-value represent the standard deviation from the median across samples of HNSCC. All results were statistically significant (p < 0.05), with exception of GAA. Figure S4: Gene enrichment analysis of our panel of interest using BEST tool. The panel of genes was composed by the genes encoding our most abundant urinary proteins and genes encoding the most promising variants in HNSCC. All results were statistically significant (p < 0.05). Tables Table 1: Descriptive statistics for clinical variables. Smoking habits Former smoker 13 (68.4%) Smoker 6 (31.6%) Drinking habits Light/ moderate 6 (31.6%) Heavy 13 (68.4%) Tumor site Oral cavity 2 (10.5%) Pharynx 13 (68.4%) Larynx 3 (15.8%) Larynx + pharynx 1 (5.3%) Tumor stage (T) 1-2 4 (26.3%) 3-4 15 (73.7%) Tumor stage (N) 0-1 7 (36.8%) 2-3 12 (63.2%) Tumor stage (M) 0 15 (78.9%) 1 3 (15.8%) No information 1 (3.3%) Medication No 6 (31.6%) Yes 13 (68.4%) Construction related professions No 4 (26.3%) Yes 15 (73.7%) Table 2: Most impactful protein variants found in urine samples from HNSCC patients. Driver genes of HNSCC COL4A1(NM_001845.6):c.4399C>T p.(Leu1467Phe) 13q34 chr13 110162293 VUS 5/19 T=0.000688 (182/264690) HSPG2(NM_005529.7):c.2896G>A p.(Glu966Lys) 1p36.12 chr1 21876336 VUS 15/19 T=0.000064 (17/264690) Genes related to HNSCC CP(NM_000096.4):c.2684G>C p.(Gly895Ala) 3q24 chr3 149178609 VUS 18/19 G=0.001908 (505/264690) EFEMP1(NM_001039348.3):c.523G>A p.(Asp175Asn) 2p16.1 chr2 55881729 VUS 19/19 T=0.000004 (1/264690) GAA(ENST00000302262.8):c.2417C>T p.(Thr806Met) 17q25.3 chr17 80117685 VUS 19/19 T=0.000688 (182/264690) GAA(ENST00000302262.8):c.55G>A p.(Val19Met) 17q25.3 chr17 80104641 VUS 4/19 A=0.000057 (8/140168) GAA(ENST00000302262.8):c.2238G>C p.(Trp746Cys) 17q25.3 chr17 80117016 Pathogenic 18/19 C=0.00023 (18/78700) SIAE(NM_170601.5):c.629C>T p.(Pro210Leu) 11q24.2 chr11 124649712 Pathogenic 6/19 A=0.000008 (2/264690) SIAE(NM_170601.5):c.587G>T p.(Cys196Phe) 11q24.2 chr11 124649754 Pathogenic 10/19 A=0.000767 (203/264690) Legend: Chr, chromosome; HGVS, Human Genome Variation Society nomenclature; N/A, no functional studies related to this variant, VUS, variant of uncertain significance. Urine frequency values for HNSCC patients refer to the number of patients in our cohort who had a given variant identified in urinary proteome. The allele frequency data refers to the frequency of the altered allele in the general population. Table S1: Characterization of HNSCC patients. Table S2: Normalized Log2 z-scores for proteins identified in urine from healthy subjects (Cont) and HNC patients. Table S3: List of proteins identified with single amino acid alterations according to CanProVar2.0. Table S4: List of all mutant protein isoforms identified in urine from HNSCC patients. Figure 1 Figure 2 Figure 3 Figure 4 Figure S1 Figure S2 Figure S3 Figure S4 Graphical Abstract Information & Authors Information Version history V1 Version 1 20 March 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords biomarker early diagnosis head and neck cancer urine variant Authors Affiliations Oriana Barros View all articles by this author Joaquim Castro Silva View all articles by this author Eurico Monteiro View all articles by this author Susana Aveiro View all articles by this author Pedro Domingues View all articles by this author Pedro Valente Sousa View all articles by this author Carolina Castro View all articles by this author Catarina Rodrigues View all articles by this author António Barros View all articles by this author Francisco Amado View all articles by this author Saeid Ghavami University of Manitoba Max Rady College of Medicine View all articles by this author Vito Giuseppe D'Agostino View all articles by this author R Ferreira View all articles by this author Lucio Lara Santos View all articles by this author Metrics & Citations Metrics Article Usage 329 views 164 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Oriana Barros, Joaquim Castro Silva, Eurico Monteiro, et al. Uncovering the proteogenomic landscape of head and neck squamous cell carcinoma through urine analysis. Authorea . 20 March 2025. DOI: https://doi.org/10.22541/au.174249381.13285606/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174249381.13285606/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe62477cdaeaa64',t:'MTc3OTIyNTU5Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();