Identification of Novel Hub Genes and Potential Signaling Pathways with the Pathogenesis of Oral Cavity Squamous Cell Carcinoma Based on Bioinformatics Analysis

Abstract

Background & aim:

Oral squamous cell carcinoma (OSCC) is a devastating disease with poor prognosis and low survival rates, despite advancements in diagnosis and treatment. Early detection and identification of molecular targets are crucial for improving patient outcomes. This study aims to identify differentially expressed genes (DEGs) and key molecular pathways involved in the OSCC. This study’s findings will contribute to the development of effective targeted therapies, ultimately improving the prognosis and survival rates of OSCC patients.

Materials & methods:

Three gene expression profiles (GSE37991, GSE30784, and GSE107591) from the GEO database were analyzed for differentially expressed genes using EnrichR. Subsequent downstream analyses of the selected module genes were conducted using various bioinformatics tools including STRING, Cytoscape, GEPIA, cBioPortal, NetworkAnalyst, MirWalk, and a bipartite miRNA-mRNA correlation network.

Result:

The reanalysis indicated that the Toll-like receptor (TLR) signaling pathway plays a significant role in the development of oral SCC and CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2 genes were up-regulated and enriched significantly in the signaling pathways’ interactions in oral SCC. Genetic mutation analysis of hub genes in OSCC revealed that STAT1 have 2.5% mutation rate and 0% for other genes. It was revealed that the development and prediction of OSCC may be affected by hsa-mir-146a-5 and hsa-mir-155-5p.

Conclusion:

Novel potential biomarkers and signaling pathways associated with OSCC have been identified, which may be important in the transformation of OSCC adenocarcinoma and may serve as therapeutic targets for OSCC.

Keywords

bioinformatics analysis differentially expressed genes oral squamous cell carcinoma microrna mutation rate

Highlights

Identification of 6 hub genes associated with oral cavity squamous cell carcinoma.

Toll-like receptor signaling pathway was significantly enriched.

STAT1, CXCL8, CXCL10, IL1B, TLR2, and CCL5 were upregulated.

Potential involvement of hsa-miR-146a-5p and hsa-miR-155-5p in OSCC pathogenesis.

Introduction

Oral squamous cell carcinoma (OSCC) can arise from various regions of the oral cavity, including the lips, tongue, palate, buccal mucosa, gums, floor of the mouth, vestibule, and retromolar region.¹ OSCC represents a major global health challenge due to its aggressive biological behavior and poor prognosis.² Despite advancements in diagnostic and therapeutic techniques, the survival rate of patients with OSCC remains low.³ Early detection of oral cancer, including its precancerous stages, significantly improves patient outcomes. Benign oral cavity lesions commonly occur in the anterior tongue, buccal mucosa, floor of the mouth, hard palate, retromolar trigone, and gingiva.⁴ Globally, oral cancer accounts for approximately 200 000 new cases and 100 000 deaths annually. The global age-standardized prevalence of lip cancer in 2012 was 0.3 per 100 000 people (0.4 in men and 0.2 in women).⁵ Several etiological factors contribute to the development of OSCC, including human papillomavirus (HPV) infection, tobacco smoking, alcohol consumption, and betel nut chewing.⁶ Surgery remains the primary treatment modality for OSCC, with the choice of surgical technique depending on the tumor’s anatomical location, extent of invasion, margins, and histopathological characteristics.⁷ However, the modest improvement in the 5-year overall survival rate from 63% to 65% over the past 8 years highlights the limited impact of recent clinical advances on OSCC prognosis.⁸ Therefore, deciphering the molecular mechanisms underlying OSCC development and progression is crucial for identifying effective therapeutic targets.³ Comprehensive molecular studies are needed to elucidate the key pathways and biomarkers involved in OSCC pathogenesis.

The present study aims to address these gaps by integrating multiple open-access gene expression datasets to identify differentially expressed genes (DEGs) and the major molecular pathways associated with OSCC. By combining data from several high-quality sources, this study provides a more comprehensive and accurate molecular profile of OSCC. Rigorous selection criteria were applied to ensure data reliability and relevance. Through integrated bioinformatics analysis, this approach minimizes dataset-specific bias and enables a robust identification of DEGs and key signaling pathways. Furthermore, survival and mutation analyses of the identified hub genes offer deeper insights into their prognostic and functional roles in OSCC progression.

Materials and Methods

Data Collection

The Gene Expression Omnibus (GEO) database⁹ contains numerous gene expression datasets; however, only a subset is suitable for identifying hub genes associated with OSCC tumor progression. To ensure data relevance and reliability, the following inclusion criteria were applied:

Samples must originate from human subjects.

Datasets must be analyzable using the GEO2R online tool.

Each dataset must include more than 6 samples.

Both normal and OSCC (disease) groups must be available for comparison.

All samples must represent oral cavity tumor tissues.

Based on these criteria, 3 gene expression datasets (GSE37991, GSE30784, and GSE107591) were retrieved from the NCBI-GEO database (https://www.ncbi.nlm.nih.gov/geo/) the platform identifiers for GSE30784, GSE37991, and GSE107591 were GPL570, GPL6883, and GPL6244, respectively. These datasets were selected to identify key molecular targets and pathways potentially involved in OSCC. Among the available OSCC-related GEO datasets, GSE37991, GSE30784, and GSE107591 were selected because they fulfilled all predefined inclusion criteria and provided high-quality expression profiles with sufficient sample sizes. The datasets were generated using well-established microarray platforms (GPL570, GPL6883, and GPL6244) and contained both OSCC and normal oral tissue samples. To avoid potential batch effects caused by platform or sample heterogeneity, each dataset was normalized individually, and only overlapping DEGs consistently identified across all 3 datasets were included for downstream analysis. This intersection-based approach ensured analytical robustness and minimized cross-platform bias.

Data Processing of DEGs

Differentially expressed genes (DEGs) between normal and OSCC samples were identified using the GEO2R online analysis tool (https://www.ncbi.nlm.nih.gov/geo/geo2r/). Expression data were normalized by quantile normalization prior to analysis. Genes with an adjusted P-value <.01 and |log FC| > 1 were considered significantly differentially expressed. Identified DEGs were categorized as upregulated or downregulated based on fold-change values. Overlapping DEGs among the 3 datasets were determined using a web-based Venn diagram tool (http://bioinformatics.psb.ugent.be/cgi-bin/liste/Venn/calculate_venn.htpl).

Enrichment Analysis

To explore the biological significance of the DEGs, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed. These analyses identified key biological processes (BP), molecular functions (MF), and cellular components (CC) potentially associated with OSCC. Functional enrichment was conducted using the Enrichr web server^10
-13 and the Database for Annotation, Visualization, and Integrated Discovery (DAVID; https://david-d.ncifcrf.gov). Significance was defined as an adjusted P-value <.05^14,15 Enrichr was also applied to annotate genes within specific modules identified from network analyses.

Protein‒protein Interaction (PPI) Network and Module Analysis

The Search Tool for the Retrieval of Interacting Genes (STRING; https://string-db.org/) was used to construct a PPI network of the DEGs, with a minimum required interaction score > 0.4.^16,17 The network was visualized and analyzed using Cytoscape version 3.7.2.¹⁸ Highly interconnected gene clusters (modules) were identified using the Molecular Complex Detection (MCODE) plugin with the following parameters: degree cutoff = 2, node score cutoff = 0.2, k-core = 2, and maximum depth = 100. Hub genes were determined using the CytoHubba plugin based on node degree and network connectivity.¹⁹ Four topological analysis algorithms (Degree, MCC, MNC, and EPC) were applied in CytoHubba to comprehensively evaluate gene centrality from different perspectives. The intersection of the top-ranked genes across these methods was defined as the final set of hub genes, ensuring robust and consistent selection.

Survival Analysis of Hub Genes

The prognostic significance of hub genes was evaluated using the Kaplan–Meier Plotter tool. In addition, the Gene Expression Profiling Interactive Analysis (GEPIA) web server.²⁰ Which integrates data from TCGA and GTEx, was used to assess expression differences between OSCC and normal tissues. The log-rank test P-value, hazard ratio (HR), and 95% confidence interval (CI) were calculated for each hub gene.

Mutation Analysis of Hub Genes

To identify genetic alterations in hub genes, mutation data were analyzed using the cBioPortal database^21,22 this platform provides comprehensive OSCC genomic data, including whole-genome sequencing results from 40 OSCC tumors and matched normal tissues.²³

Potential miRNA‒mRNA Interactions

Potential interactions between differentially expressed miRNAs (DEmiRs) and hub genes were predicted using the miRWalk database. The overlap between miRNA target genes and module genes was visualized using a Venn diagram. A bipartite miRNA–mRNA regulatory network was constructed and analyzed using Cytoscape software.²⁴

Results

Identification of DEGs

These 3 profiles included 101 samples of normal oral mucosa and 224 samples of OSCC (Table 1). Three gene expression datasets, GSE30784, GSE37991, and GSE107591, were selected from the GEO-NCBI for OSCC after the DEGs between OSCC and normal oral mucosa were screened. For all analyses, adjusted P-values <.01 and |log2FC| > 1 were taken into account. There was an overlap across the 3 datasets, depicted by a Venn diagram showing 1157 downregulated DEGs and 941 upregulated DEGs (Table 2 and Figure 1A and B).

Table 1.

The Details of the GEO Datasets.

ID	Data type	Tissue	Platform	Tumor	Normal
GSE30784	mRNA	SCC	GPL4133	167	45
GSE37991	mRNA	SCC	GPL2895	40	40
GSE107591	mRNA	SCC	GPL570	17	16

Table 2.

Venn Diagram Results from 3 GEO Datasets for SCC.

Cancer type	DEGs	Gene names
SCC	Upregulated 941 gene	CLMP MMP2 FAM219A APBB2 EDN1 MMP7 CDK14 CXCL13 SUGCT CMPK2 BRCA1 SLC12A4 CDC25B COL1A1 NUDT1 SEL1L3 SMYD3 STK17A CNPY4 CD38 DUSP4 FOXM1 CHST15 PLXNA1 IL1F10 IL24 RGS4 RAG1 TMEM104 SNAPC1 LILRA6 C1orf112 TFEC ENTPD7 GTSF1 PDGFRB CENPL FST SCRN1 KIF4A UCN2 POLR3G CASP1 XAF1 NOL4L DSC1 PSMB8 APOL1 MME SOCS3 TMEM132A DUSP10 AMPD3 GNLY NDE1 CDK18 CACNA1C PLBD2 IFNE PCDH12 LPAR3 CD276 CLIC4 SH3PXD2B ATP13A2 SCG2 GALNT6 CPXM1 NFE2L3 OAS2 GLIS3 DPH7 CDC45 HTRA1 WISP1 SLC15A3 ITGAX HEATR6 FCRL5 GPR84 ZNF474 COL4A6 CABYR ADAM8 MAP4K2 SULF1 CDCA2 GJC1 EXO1 KIRREL WFDC12 IGSF6 SNRPB2 PDCD1LG2 COL24A1 IGFBP3 SEC61A2 COL11A1 TREM1 GBP5 ODF3B HOXB7 LBH SFXN5 WNT7B JRKL RELL2 TYMP GGT5 APOC1 NEK2 AP1S1 TMEM206 CCDC51 RPTOR SLC6A2 NID1 MMP13 HOXC13 HCP5 MSN TBC1D31 GTF2E1 SLC39A14 ABCG1 EML1 NEDD4 FBLIM1 PELO BORA CHEK1 PAG1 GRK3 CD40 MED10 STRIP2 PLEKHG2 SERPINA1 FADS2 ENAH ATP2C1 RAPGEF1 HSF2BP STC1 TCP11L1 RAB23 FAM167A NLRP7 ARSB HARBI1 EIF5A2 ITM2C MCAM RDH16 RFC4 ARRDC4 POLQ SPAG4 SOAT1 OASL PRR11 DFNA5 GLMP TCEAL9 ISG15 PDPN P3H4 COL3A1 P4HA2 SMC2 CCR8 IFITM3 OCIAD2 CDYL2 CCR1 NCS1 LYN DUSP6 TGFB3 SERPINE2 NTMT1 LILRB1 GINS4 MRPS23 AEBP1 TMEM177 CXCL1 PKMYT1 NRIP3 MYO7A EGLN3 UNC119 AIM2 MORC2 PPP1R14C DSN1 SPATC1L TTC9C TMEM2 PRAME HIP1 GSDMD CDKN3 FJX1 NUDCD1 PCGF1 PPP4R4 SUSD1 FCHSD1 ZNF200 SGPP1 ITPR3 BIRC3 NRSN2 APOL2 ITPK1 SLC38A5 BIRC5 LAIR1 LRRC8D CD300A GALNT18 FN1 GPR132 EML5 SORCS2 PRIM2 PGM2L1 NUAK1 DSE SPP1 ENO2 PTGS2 NEFL IL4I1 BMP2 ACOT9 RASGEF1A CDR2L PIK3CD SLC7A11 ACAN CHST2 TXNDC12 GTSE1 CD86 STAT1 COL8A1 MMP12 CDCA5 CARD16 BNC1 AJAP1 CYTH4 MMP11 XCL1 CCDC102B CHGB DIAPH3 COL5A3 COL4A2 UBASH3B DYRK3 MTFR2 IGF2BP2 CMSS1 ACOT13 HIST1H2AM LILRB3 HOXD10 FSCN1 SIRPA CLEC7A C1QTNF6 LIPG FEZ1 RBBP8 STARD4 ACVR1 TTI1 SLC43A2 EXTL3 NID2 ADTRP SMTN NACC1 ESCO2 BFSP1 FLRT3 LCP2 CCL5 CLSPN DZIP1L LOX LAMC2 SEMA4F IMPDH1 ICOS FCER1G SHOX2 TPST1 DHX58 IFITM1 PGS1 FSD1L EPHB2 COL27A1 HAVCR2 PHLDB2 FAM101B LPCAT1 TMC7 HSD17B6 TCOF1 SLC16A1 C1QTNF1 KRT75 MOB3B MSC HMGB3 HOMER3 FPR3 SLCO1B3 TMEM86A ADAMDEC1 FBN2 CARD18 P3H1 HENMT1 NUP62CL REEP3 MRGBP DTX3L TTYH3 ZFP69B KLHDC7B HAUS6 CDC25C NRP2 ACP2 TENM2 DSG2 TMEM44 CXCL8 PARP14 MSX2 IFIT2 GUCY1B3 SNAPC4 SLC9A7 CDK6 FURIN AKIP1 BLM SMAD7 WARS PTK7 DDX60 KYNU PRPS1 DRAM1 NOX4 SLC4A2 OSTM1 CLCN7 SPECC1 VAV2 GLI3 ZBP1 MB21D1 SPATS2 SLC7A8 GRAMD1A CCL24 MX2 NXPH4 HLA-F SULT1E1 BRD9 HSPA13 RARA NPHP4 PML DPP3 SERPINH1 IL2RA FAP TNFSF4 BCL2L11 INHBA SLC36A1 PPEF1 LOXL2 CDK5RAP1 TRIM21 NLRC5 FADD VNN1 DOK3 CALU C9orf84 DENND3 CKAP2 AUNIP MICALCL CA2 ETS1 WDHD1 TCIRG1 SIGLEC10 FAM26E FTSJ1 CYP26B1 GNB5 SERPINE1 BEND6 RSAD2 RAD51 NEDD9 CFB CXCL6 GMNN CSF2RA KIF20A ACD COL10A1 CERCAM ZNFX1 BCAT1 SLC2A9 CEBPB NRG1 TTK TGFBI PSMB10 PGM3 SLC38A7 CASK CD274 SLC52A2 ANO6 SDCCAG3 MAD2L2 MISP FARP1 ITGA5 CNGB1 EPB41L4B COL4A5 AQP9 MET TNS3 LHFPL2 ITGA3 RUNX3 LUZP1 EFNB1 C5AR1 DNAH17 SLFN12 TNFRSF9 CCNE1 APOE ST8SIA4 TOR3A SYT12 PXDN CHST11 IRS1 CDC42EP3 GLIS2 PILRA COL7A1 CMTM3 UBE2L6 ALDH1L2 PTPN2 KIF14 ACTR5 SH2D2A IFI27 TNFAIP6 KIAA0226L GPR158 APBA2 RGS20 TRAM2 PI15 ELAVL2 CA9 OSM OLR1 CLIP2 FOXN2 CEP164 SMS LAMA1 CDH11 NDC80 CEP72 OIP5 NELL2 CHPF2 E2F7 LAMA3 SLC11A1 TNFAIP3 MMP1 SFXN3 ADAMTS6 SMURF2 ADAMTS12 KIF23 VNN3 QSOX2 STK10 EIF2AK2 CCL11 IDO1 GOLGA7B ZNF114 MMP3 CCL18 SPAG5 BRCA2 SDSL ERCC4 SP110 AGRN CEP290 IRX4 TREM2 RPS6KA4 ARMC9 MINPP1 MPHOSPH6 CLPTM1L DDIAS PLAC1 APLN MCM2 RAB32 LRP12 L3HYPDH CXCL10 GDPD2 CKAP2L CYP27C1 HERC5 NHS MTERF3 EVA1A FAT1 NREP ZNF668 ADAT1 CENPI FPR2 CCNF KLHL5 LOC90768 POLD1 CALD1 SKIL AMPD2 ACTN1 FBXO6 RFC5 COL1A2 PROCR ARSJ KCNE4 COL12A1 TTC26 ACKR3 ESM1 RCN3 BCL2L12 KIF26B CCNA1 BID MACF1 SEC24D TGFB2 TRIO RAB42 NRM CALB1 DOLK ABL2 PTPRZ1 PDLIM7 ALKBH1 RGS3 RAC2 UAP1L1 ITGB4 DCBLD2 BTN3A2 OSMR PTHLH NCF2 KNSTRN CCM2 TAP1 SGIP1 CCL7 DCLRE1B RBMS2 MYO10 IL1RL1 SNX10 RIPK2 ABCC4 SLC25A22 TRPV2 TIGIT FKBP10 VEGFC SUPT3H HELZ2 ADAMTS14 GPR176 Gene.symbol BCAR1 LRRC15 CASP7 EDEM2 AATF SCARB1 GREM1 PLOD3 KDELR3 BUB1 TRIP13 FBXL2 MVP PSTPIP2 XPR1 C16orf74 PODNL1 C19orf66 TSPAN15 NOCT NT5C3A BST2 STX1A KLF7 IFI6 MGAT2 IL12RB2 IQCA1 SASS6 STIL UBE2C FHOD1 SLAMF9 SDK1 IFI44L LEMD1 CAPRIN2 MAGEA6 CAV1 NAV1 DBF4B RAD54L ABCD1 ZNF319 PRSS23 USB1 KIF18A RTP4 SLC47A2 MTCL1 CLEC12A ZFP64 IL7R IFI35 AZIN2 DOCK5 GLIPR1 MMD DDX58 COL6A3 ERCC6L PIP5KL1 ENC1 SLFN5 VOPP1 STC2 COL6A1 IL11 TDO2 S100A7A COL5A1 CTHRC1 UHRF1 MFHAS1 RBP1 KIF3C RUNDC3A SCG5 FGD6 B3GAT3 TTC7B HAPLN1 TNIP3 ADAM12 CSPG4 IGF2BP3 HOXD11 SP140L BAK1 PCSK1 LAX1 TANGO2 RHOC HTR7 GSG2 BTN3A1 LAMP3 DNMT3B LTBP1 C3orf52 PSMC3IP CDC6 KDELC1 CHI3L1 ADAMTS2 APOBEC3B AURKA NEK6 CDSN SLC17A9 PIEZO2 FANCA LARGE2 DUSP14 COL5A2 XRCC4 THY1 C17orf80 STK3 MSR1 EPSTI1 POFUT2 TGFB1 SLC31A2 ABCA12 ANAPC10 BCL2A1 CDH3 PLPP4 VAC14 NXN CENPE PARP12 CTNS ACP5 PGF DKK3 POSTN CYP27B1 TACC3 DMXL2 PPP2R5B TMED8 CTLA4 WNT2 CSMD2 DCBLD1 SLC2A6 PDGFA P3H2 MUCL1 CEP55 TENM3 F2R GZMB BMP1 DPF1 PDE7A RASL11B ZCCHC17 C2 TRPC6 ARPC1B GLDC PLA2G7 CEP19 MFAP2 SOCS1 PJA1 COL16A1 GPX8 TIMELESS RAD51AP1 PHTF1 COL4A1 FFAR2 TGIF1 NAV3 EME1 RPS6KA1 HAPLN3 ASPHD2 SPRY4 IFIH1 MANBA IL1B HIST1H2BD CPT1C GPR68 TLR2 GNA12 LIMK1 PARP9 MMP10 AMIGO2 PIM2 C4orf46 BMP8B ADGRE1 IFIT3 PLOD1 NAT14 MYO1B PANX1 PLAU SLC28A3 PRKCDBP GPR180 SLC22A4 BIK SMIM3 TGM2 STX2 WDR54 PLK1 RFTN1 BRMS1 YEATS2 ABCA1 PCGF6 RAI14 FKBP14 REC8 CTSV CLP1 PTRH1 PAPPA CXCL11 PFN2 TNFRSF4 OAS3 LIMA1 INF2 SMOX NCAPG2 ETV4 STARD3NL MYH10 KIF2C MYO5A FLNB MFSD10 LAT2 LIG4 TNFAIP8L1 WRAP73 MTBP NEDD1 GRIA3 HAUS8 ATG101 RAB31 SLC6A14 ARNTL LAMC1 FAM26F COL13A1 TLDC1 ANGPT2 LAMP5 NCAPH PTPRK TK1 LETM2 ST3GAL2 ODF2 IKBIP SHCBP1 COTL1 TNC LAYN PRIM1 PLEK2 AFAP1L2 SP100 CD80 FMNL3 KCNAB2 HPCAL1 PLA2G15 IFIT1 HCG4 TSHZ3 ZBTB9 RAB29 SPIRE1 MICB CXCL9 EGFL6 CTSL TMEM138 MFSD12 HJURP NMI STAT2 HMMR JAK3 CXCL5 FSTL3 MYO1G ORC6 BMPR1B PMEPA1 HOXC6 CHN1 C18orf54 SNAI2 GPR39 IL2RB SEMA3C WDR66 IFI44 PSMB9 CSF2 MMP9 DHRS2 SLC39A6 CTSC
	Downregulated 1157	ELMO2 ABRA CLK4 DECR1 REEP6 SEMA4D SFTA2 C15orf41 USP6NL TEK ANKRD6 LTBP4 ABR SIAH3 ZNF257 RMND5B MAOA GDF10 PMM1 CPEB3 METTL25 TF UNC13B SETMAR LAMB4 EML3 MRM3 HDHD2 SLC9A3R1 ASAH1 ZNF350 KRT15 FAM175A FHL1 CHMP2B CAMSAP1 LMO4 KRCC1 PLSCR4 PPARGC1A ETFA EXOC1 CYP3A5 PITX1 FETUB SEMA3G GRHL1 ARHGEF26 PREX2 PGM1 DCT NTN4 TSHZ1 DCAF11 KALRN THOC7 TLR5 FAM129B KIAA0232 FBXO3 ZNF750 SNCA SLC48A1 RBP7 EPB41L3 PROS1 AIF1L WDR26 MALL MAGI1 BLVRB STK39 KLHDC1 C11orf57 GRAMD1C SERPINB11 SYNPO2L OGFRL1 FBXO34 ALOX12 ABHD5 TRAF6 SHE CTNNA1 COMMD1 FOXN3 TACC1 KRT4 ABCA6 ZDHHC3 SLC25A4 SESN1 GBE1 FARP2 MAPKAPK3 CYSLTR1 RNF146 LGALS3 RBSN AGL ZBTB21 MIR99AHG TMEM43 OBFC1 TMEM50B SCARA3 TACSTD2 PPP2R2A F10 HCFC2 TCAP PPP1CB SYNE1 SPTBN2 SLC26A2 ADIRF C21orf33 FGF10 MBNL3 SOX6 CRIP1 CHP1 PRDM5 LRRN4CL CA5B FBXL3 PLPP1 SIM2 PIK3R1 ADCY4 CDH19 TPPP GSTM5 ZNF570 DOPEY1 TIMP4 SLIT2 LONRF1 SLC25A23 DYNLT3 FAM3D SLC4A4 KAZN RPL14 ZNF682 AOX1 MAFF PHKB NDUFA5 ZADH2 ST6GALNAC1 CHPT1 GPHN EIF4EBP2 HSPB6 COX5A CHRDL1 SMARCC2 CCDC12 STRADB RPS23 ASB8 KRT13 PRKAR2A SKP1 FYCO1 RASAL1 PIK3C3 FAM135A HACE1 MAL SLC47A1 IDNK CYP4F3 CROT STOX2 DNAH1 ABCB1 PLAGL1 USP54 LYRM5 TM7SF2 OXCT1 ABCA8 NDUFS2 EFCAB14 MAML3 EDNRB ATP5S INPP1 OXR1 TGM3 HK1 GALNT12 ABI3BP TRMT13 GRHL3 TOM1 CENPC PRPSAP1 GCHFR PELI1 UBE2B OMA1 ABLIM2 DIMT1 PYM1 TCAIM TXNIP HEBP2 TLR3 CSRNP3 KANSL1L TMEM100 SUCLG1 TUBE1 NDUFV3 LANCL3 ZDHHC11 MYH14 CC2D2A TPRG1 NUCB2 FOXO4 ID4 AKAP1 NFIX SLC16A6 HIBCH RREB1 GDPD3 SECISBP2L FGF2 CEP76 LDB2 PGAP3 PCM1 PIGN SHROOM3 JAM2 COQ8A PLEKHG6 CASZ1 CDKN2AIP TMEM116 LARP1B GSTM3 EIF3L SERAC1 FBXW11 ZNF626 C18orf25 RARG RBMS3 TSPAN6 FGFBP2 COBL GAS7 ARHGAP6 RAB11A SNX1 RBM8A UGP2 ECHDC2 VPS52 SCRIB GPX3 RRAGD KYAT3 ANK3 SNX24 DIS3L GPRC5B FASTKD1 DOCK3 ALDH3B1 DLAT THSD4 SRSF5 TRPS1 ACOX2 COA3 PDHB C1orf116 PHYHD1 MTMR3 TRNP1 GATM PGD TMEM168 APPL1 SFRP1 DAPL1 SUOX ALDH9A1 SLC18B1 TMEM106B C11orf54 ZNF420 SMAD5 FBLN5 ZNHIT6 CIPC DLG1 FIBIN TACC2 CYCS GNAL TAF9B PADI1 NDUFS1 UPK1A AES SLC16A14 CEACAM6 GOT1 FRY DKK2 ARL4A HNMT DCP1A C15orf52 CES2 GHR HS3ST1 ALDH2 SCP2 GUCY2C BARX2 PPP2R5A BDH2 ZNF395 LGI4 STX19 NEBL STK40 C15orf59 NDNF SMPD2 MLIP ZNF672 ACVR2A ZC3H6 CSTA USP4 PPL RALGPS1 CCDC6 PLA2G6 SYTL4 ALDH6A1 SVEP1 FAM3B ZNF254 PAX1 SELP LRBA PANK1 PLAC9 SH3BGRL2 FDFT1 BMP7 ADH1C CYP2C9 RUNX1T1 DMXL1 RUFY1 RNPC3 PCSK2 ARID2 ZBTB4 BNIPL ZC2HC1A SSBP2 CXCL17 MIPEP ARHGAP20 GREM2 HDDC2 GAREM1 MKNK2 ZNF44 ZNF555 SASH1 PHC1 HECTD1 ZNF493 HDAC4 FAM102A SCGB1A1 USP46 GYS2 BLVRA RASL12 IKBKB ZNF706 CBR3 USF3 LYVE1 MUT VPS37A BAZ2B AKAP12 APOD PAOX ZNF423 MACC1 SUSD4 MZF1 TST CHCHD10 IL1RN ABCA9 PCBD1 TGFBR3 CASP9 PPA2 FAM219B GCNT3 TMEM120A FXYD1 USO1 RUFY3 SLC24A3 TOP2B FAM117B CEACAM1 RNF11 HSDL2 AKR1A1 PSCA HIGD2A LDOC1 MID2 RANBP9 LNX1 CFD SOX2 KAT2B PDLIM5 DUSP13 CD1C EYA2 WLS OAT COBLL1 SLC30A9 TMPRSS11B ZSCAN18 TC2N CILP SLC6A4 SMDT1 SYNPO2 NR2F1 PRDX6 LYRM1 RNF141 ACAT1 MPP7 PLA2G2A SOCS6 LMBRD1 SCIN MYO5B ACOX1 EFHD1 ELF3 RCOR3 FUT6 AIM1L SIPA1L2 RCAN2 PPARG RGAG4 MAPT ECM1 EPB41L1 C1orf21 DYM PIM1 MAP2K6 CCDC120 RALGAPA2 P2RY1 AKTIP UQCRFS1 ADH7 S100A14 ASPA TNNI2 DISP1 SPATA6 CYP7B1 INTS10 USP25 PCYOX1 PLEKHA7 AP5M1 REV1 CYP2E1 CRYBG3 TJP1 PER2 HPGD PLEKHH2 CXXC5 ADSSL1 SMIM19 RHCG BCOR NIPAL2 DHRS3 CPPED1 GAB2 VPS13D IL12A PDE1A MKRN2 EPHX2 ICK DUOX1 ADGRG6 PHLPP1 SH2D4A CD34 PHACTR4 MYLIP RPIA KCNT2 ANKRD16 MGLL CD24 DENND2C SCARA5 CLCA4 ACADSB SEPP1 CADM3 PCF11 SORBS1 FGF7 FOXA1 ALDH3A1 LIMCH1 RWDD1 ENSA MMAA HOOK2 CCNG1 HADH ELOVL6 EPHX1 2-Mar MAPK3 MPZL3 BCAS1 STAP2 CRNN SERHL2 ADCY6 ZNF273 CTDSPL CLYBL PCCA FMO2 RAVER2 LPAR6 ZMYND11 LRRFIP2 PLLP WDR41 FAM221A CDK19 SCNN1A HSD17B8 NNT TAPT1 ZNF654 TCEA3 DUSP16 PI16 AGAP1 FREM1 NDUFA10 GGTA1P KLK11 CAPN7 PARD3 RETSAT CES1P1 VSIG10 TIFA OLFML1 TNS2 KAT6B LANCL1 SNX21 BCKDHB N4BP3 GNL3 CYP11A1 TMX4 CAPNS2 ING3 C4orf3 NFIA ATXN10 ECH1 ACTR8 HES1 ETNK2 FBXL14 KIAA1211L N4BP2L2 OPHN1 OR7E37P TMEM255A COPS4 DUSP5 TNXB SLC9A9 GSTM4 SLC13A4 TSPAN7 BRK1 TACR1 PAQR8 MYO5C ZBTB20 SPRY2 LONRF2 AHNAK HEATR5B SORBS2 NFE2L2 MUC15 SPINK5 SVIL NDRG2 TYRP1 ACADVL HBP1 PHGDH DOPEY2 SUMF1 SC5D NOTCH2 MGP CYP4F12 SCNN1B GPRIN2 CUL3 HMBOX1 MUC20 NR1D2 UQCRQ SLC37A1 GGT6 ARHGAP10 TRAPPC6A GPD1L KLHDC2 WDR82 TRIP10 HRASLS TRIM24 FIG4 LMO7 FNDC4 FOXP2 ATP5A1 RMND5A F13A1 ACADM CHMP2A SLC44A3 SLC6A1 OCLN CDON SH3YL1 TDRD3 ZC3H8 FBXO9 CEACAM5 FUT3 DBI D2HGDH CPEB4 MYOC BCKDHA RPL3 SPATA18 MKL2 DIO2 TRIOBP RBM5 NCOA1 PKHD1L1 PDZD2 CLCN3 ESYT2 SYBU SCN7A ZDHHC21 DCXR FAXDC2 GPC3 CSTB NOVA1 PLA2G16 PGM5 SLURP1 HMGCS1 RAB10 ZNF823 SUCLG2 CYB5A KCTD3 PLP1 ABCA5 NDUFS3 PCOLCE2 ROR1 ZFAND2B PEBP1 HSBP1 GULP1 CYP4B1 LYPLAL1 ABLIM1 PSD3 HOPX ZNF519 KANK1 PTN RPL10A HSPB8 RNF170 MFSD6 ZFP3 MYOM1 DIAPH2 CLN8 SPAG16 SCOC SLC16A7 PGRMC2 ANKRD35 MYO6 CNKSR3 SOD3 CRYAB C6orf203 LMF1 PAX9 ABO GALNT5 LIPE DEGS2 ZNF721 BAG1 VPS36 NUPR1 ATP13A4 TP53INP2 RASEF POGK CLDN7 SNX31 DLG3 FAM171A1 NAA16 ARIH2 CYP4V2 PITX2 KIAA1217 FZD7 GATSL3 CHL1 AIM1 GSN PLS1 ANK2 VLDLR CGGBP1 P4HTM ARHGEF3 SLIT3 MAN1C1 COX20 GID4 IL34 PKP4 DDAH2 Gene.symbol SHPRH P2RY2 UACA CD9 USP53 C4orf33 SMAD4 ACPP NFIC ITPR2 SUN2 PDIK1L SHMT1 CRISP3 CPT2 SLC7A2 PRKAA2 DPT RGS5 OPLAH ABCG2 BBS2 SELK EZH1 CACNB4 MAOB PTGDS ATP5L IGSF10 NOL12 UBL3 TRAPPC8 ERBB3 CYP2C18 MYCBP2 TMCO4 CLU PAMR1 CXCL12 IDI1 KCTD6 GLTSCR1L LGI1 EHD3 FAR1 ALDH1A1 STRBP ZNF256 SPARCL1 IL6ST FAM172A PPM1L TEC MON2 FAM98C COX5B MSRA CD1E TMOD1 RASAL2 WSB2 GPD2 HIPK2 ZMIZ1 CRTAC1 FAM107A SLC25A12 EIF3K TCP11L2 SREK1 MIOS MMRN1 GMDS ZNF662 PYGM ZBTB5 LPAR1 RELN MAMDC2 YTHDC1 EVA1C AR LEXM GOLGA4 SELENBP1 ARHGAP27 GPAT3 SAR1B PTGIS PPIL6 PDE12 FAM214A FRAS1 ATP9B MITF C15orf48 ENDOU HOOK3 UTRN IMPDH2 ZNF529 ALDH3A2 SORT1 RPLP1 RAB5B LRRK2 RASSF5 PLD1 HLF PBX1 SYTL1 VIT NPRL2 BARD1 C1QTNF7 ANO10 RNF150 WNK4 PIK3C2G PDCD6IP RASSF9 IK PRSS27 PCBD2 CAND2 MDH2 TPD52L1 RAB25 L3MBTL4 KRT78 PDK4 ARGLU1 SAMD5 IGFBP5 NR2F2 CTTNBP2 ANGPTL1 SMAP1 P2RY14 KLHDC8A NR3C2 FAM63A EMCN BLNK MCCC1 TTC19 ARHGAP32 SLC27A6 EIF2B4 IL17RD CCDC125 MPHOSPH10 OARD1 AASS RCHY1 MED7 NAAA MGST2 OSBPL1A PID1 CRIPAK ARSG SYT8 ST3GAL4 MINA VASN ZNF507 RAB11FIP1 CYSRT1 PKIA KRT19 ZNF106 ISOC1 MYZAP ZNF438 SUCO ZNF512B ZNF615 UBE2D4 ACACB TRAK1 ETFB RORC NMU MXD1 AZI2 SEMA4A BOC MANSC1 GFRA1 SIAE TESK2 TXN2 PRELP ZNF91 LPIN1 PGAM2 CYBRD1 PLEKHA6 EVPL SYNGR1 REPIN1 PPM1D XYLT1 GOLPH3L TMTC2 L3MBTL1 C3orf14 RNF135 PDE6A FGL2 MPZL2 EHF AFF1 ANXA1 ZFP28 ARL1 AK3 KIAA0141 MRPS14 BTNL9 LIN54 SLC44A2 KIAA1683 GAN HDAC5 GSTA4 PM20D2 MUC1 RNF180 DCLK1 CFH TANC1 CNN3 MMAB RAPGEFL1 SMAGP MPC1 TGM5 EMP1 FBXW4 GSTO2 RAF1 SYT15 PHKA1 CD207 STXBP5 TMEM9B SMPDL3A WDR49 TXNL1 ICOSLG GID8 SCEL NTRK3 BEX4 TADA1 DIS3L2 CLASP2 TMPRSS2 MED23 UFC1 KDM3B ATP6V0A4 ACSS2 CLOCK TMBIM4 SORCS1 FAM13C PDGFRL HMGN5 MFAP4 DOCK9 DLG2 ZYG11B PPARA SCFD2 RORA PIAS2 CYP4X1 ATP1A2 ATP5O BTC SDPR SLC2A12 KLF8 PLXDC2 ACKR1 ETFDH LIFR MXI1 TSEN2 EXPH5 KMT5B TSPO CLINT1 C2orf54 STX12 OSR1 PTGR1 SERPINB6 XPA FAIM2 RPAIN IL36A PLEKHM1 ANAPC16 CPNE3 RECK ABAT RBM47 GBP6 CRBN ATPAF1 ELF2 CEACAM7 METTL7A ITM2A KIAA1468 EIF4E3 CAPN5 COX7A1 FAM189A2 KCNRG KLHDC10 LGI3 C6orf136 DEPTOR TMEM220 BRWD3 PPFIBP2 PRPF38A ROBO2 GNA14 NT5C2 ADH1B MLANA SPRR3 CAB39L SGSM2 EPHB6 RGMA SLC9A8 RALGAPA1 PPP1R3C ABLIM3 SMARCA2 TMEM254 TXN GPT2 ZNF436 PHF10 PARD3B GAB1 CYP2J2 FCHO2 NDUFS7 GRAMD3 ZNF383 ANXA11 CNST CYP2R1

Figure 1.

Identification of common differentially expressed genes (DEGs) across 3 OSCC datasets. (A) Venn diagram of upregulated DEGs identified from GSE30784, GSE37991, and GSE107591 datasets. (B) Venn diagram of downregulated DEGs from the same datasets. The overlapping regions represent genes consistently up- or downregulated across all 3 datasets. A total of 943 upregulated and 1159 downregulated DEGs were found to be shared, which were used for subsequent functional and network analyses.

Functional Enrichment Analysis

Functional enrichment was examined using Enrichr, a feature-rich web service for the analysis of gene set enrichment. The KEGG pathways were assessed to specify the significant pathways associated with the DEGs. Based on KEGG analysis, the DEGs for SCC were both upregulated and downregulated (Table 3).

Table 3.

KEGG in SCC.

Expression	Term	P-value	Gene
Upregulated	Cytokine-cytokine receptor interaction	4.03E−11	CXCL6;CD40;CXCL9;CCL11;CSF2;CXCL8;IL24;CXCL1;CXCL13;CSF2RA;CXCL5;IL1RL1;CCL7;CCL5;CCR8;CCL18;TNFRSF4;IL12RB2;ACVR1;CCR1;IL11;CCL24;TGFB2;TGFB1;TGFB3;TNFRSF9;OSM;BMP8B;INHBA;OSMR;IL1F10;CXCL10;BMP2;CXCL11;TNFSF4;IL1B;IFNE;IL2RA;IL2RB;XCL1;ACKR3;BMPR1B;IL7R
	Protein digestion and absorption	5.32E−11	SLC36A1;COL16A1;COL27A1;COL13A1;COL24A1;MME;COL11A1;COL12A1;COL1A1;COL3A1;COL1A2;COL4A2;COL5A1;SLC7A8;COL4A1;COL5A3;COL7A1;COL5A2;COL6A1;COL4A6;COL4A5;COL8A1;COL10A1;COL6A3
	Pathways in cancer	8.11E−10	CXCL8;PIK3CD;LAMC2;LAMC1;BRCA2;ETS1;GLI3;CASP7;RAC2;PIM2;JAK3;PDGFRB;EDN1;ITGA3;MMP1;MMP2;F2R;MMP9;PGF;CCNA1;COL4A2;CCNE1;COL4A1;RARA;COL4A6;BIRC5;COL4A5;MET;BIRC3;LAMA1;LAMA3;PDGFA;LPAR3;PTGS2;CSF2RA;BCL2L11;GNA12;BAK1;FADD;BID;WNT2;IL12RB2;EGLN3;TGFB2;TGFB1;STAT1;TGFB3;STAT2;WNT7B;FN1;VEGFC;PML;BMP2;CDK6;RAD51;IL2RA;IL2RB;GNB5;IL7R
	Focal adhesion	1.37E−09	ITGB4;LAMA1;LAMA3;TNC;PDGFA;PIK3CD;LAMC2;LAMC1;RAC2;SPP1;FLNB;PDGFRB;ITGA3;ACTN1;CAV1;FN1;VEGFC;PGF;VAV2;COL1A1;COL1A2;COL4A2;COL4A1;COL6A1;RAPGEF1;COL4A6;COL4A5;COL6A3;ITGA5;MET;BCAR1;BIRC3
	Amebiasis	1.63E−09	TGFB2;CSF2;TGFB1;CXCL8;TGFB3;LAMA1;ACTN1;LAMA3;FN1;PIK3CD;CXCL1;LAMC2;LAMC1;COL1A1;COL3A1;COL1A2;COL4A2;COL4A1;IL1B;COL4A6;COL4A5;TLR2
	ECM-receptor interaction	3.40E−09	ITGB4;ITGA3;LAMA1;LAMA3;TNC;FN1;LAMC2;LAMC1;HMMR;COL1A1;COL1A2;COL4A2;COL4A1;COL6A1;COL4A6;SPP1;COL4A5;COL6A3;ITGA5;AGRN
	Rheumatoid arthritis	9.42E−09	CD86;CXCL6;IL11;TGFB2;CSF2;CXCL8;TGFB1;MMP1;TGFB3;CD80;MMP3;CXCL1;TCIRG1;CXCL5;CTSL;IL1B;CCL5;ACP5;CTLA4;TLR2
	AGE-RAGE signaling pathway in diabetic complications	3.47E−08	TGFB2;EDN1;TGFB1;CXCL8;TGFB3;STAT1;MMP2;SERPINE1;FN1;PIK3CD;VEGFC;COL1A1;COL3A1;COL1A2;COL4A2;COL4A1;IL1B;COL4A6;COL4A5;NOX4
	Viral protein interaction with cytokine and cytokine receptor	3.47E−08	CCR1;CXCL6;CCL24;CXCL9;CXCL8;CCL11;IL24;CXCL1;CXCL13;CXCL5;CXCL10;CXCL11;CCL7;CCL5;IL2RA;IL2RB;CCR8;XCL1;ACKR3;CCL18
	PI3K-Akt signaling pathway	3.63E−08	IRS1;ITGB4;LAMA1;LAMA3;TNC;PDGFA;PIK3CD;LPAR3;LAMC2;BRCA1;LAMC1;RPTOR;BCL2L11;SPP1;JAK3;PDGFRB;ANGPT2;ITGA3;F2R;FN1;OSM;PPP2R5B;VEGFC;OSMR;PGF;COL1A1;COL1A2;CDK6;COL4A2;CCNE1;COL4A1;IL2RA;COL6A1;IL2RB;COL4A6;COL4A5;COL6A3;GNB5;ITGA5;IL7R;MET;TLR2
Downregulated	Valine, leucine and isoleucine degradation	6.23E−10	BCKDHA;HMGCS1;BCKDHB;MCCC1;ABAT;ACADSB;ACAT1;ALDH3A2;ALDH6A1;ALDH2;PCCA;OXCT1;AOX1;ACADM;HADH;HIBCH;ALDH9A1
	Propanoate metabolism	2.22E−07	BCKDHA;ALDH6A1;ACSS2;PCCA;ACOX1;BCKDHB;SUCLG2;ABAT;SUCLG1;ACACB;HIBCH;ACAT1
	Fatty acid degradation	5.45E−07	ACADVL;ADH1C;ADH1B;ADH7;ACADSB;ACAT1;ALDH3A2;ALDH2;CPT2;ACOX1;ACADM;HADH;ALDH9A1
	Pyruvate metabolism	1.69E−06	ACSS2;ADH1C;ADH1B;MDH2;AKR1A1;PDHB;ADH7;ACACB;ACAT1;ALDH3A2;ALDH2;DLAT;ALDH9A1
	Histidine metabolism	1.76E−06	ALDH3A2;ALDH3A1;MAOB;ALDH2;MAOA;ALDH3B1;HNMT;ASPA;ALDH9A1
	Drug metabolism	3.36E−06	GSTM4;GSTM3;MAOB;GSTO2;ADH1C;MAOA;ADH1B;MGST2;FMO2;ADH7;CYP3A5;ALDH3A1;CYP2C9;GSTA4;ALDH3B1;IMPDH2;AOX1;CYP2E1;CES2;GSTM5
	Tyrosine metabolism	3.64E−06	ALDH3A1;ADH1C;MAOB;GOT1;DCT;ADH1B;MAOA;ALDH3B1;TYRP1;AOX1;ADH7
	Glycolysis/Gluconeogenesis	5.00E−06	ACSS2;ADH1C;ADH1B;PGAM2;AKR1A1;PDHB;ADH7;HK1;ALDH3A2;ALDH3A1;ALDH2;ALDH3B1;DLAT;PGM1;ALDH9A1
	Metabolism of xenobiotics by cytochrome P450	5.72E−06	GSTM4;GSTM3;GSTO2;ADH1C;ADH1B;EPHX1;MGST2;ADH7;CYP3A5;ALDH3A1;CYP2C9;GSTA4;ALDH3B1;CYP2E1;CBR3;GSTM5
	Beta-alanine metabolism	3.33E−05	ALDH3A2;ALDH3A1;ALDH6A1;ALDH2;ACOX1;ALDH3B1;ABAT;HIBCH;ALDH9A1

PPI and Modular Analysis and the Identification of Hub Genes

A total of 1878 genes 13 602 edges in OSCC were detected as differentially expressed genes in OSCC, utilizing Cytoscape 3.8.0 and the STRING 11.0 database. Afterward, the resulting PPI network was evaluated. The uppermost 20 genes were determined by the techniques of Degree, MNC, EPC, and EPC using the CytoHuuba plugin in Cytoscape software. Twenty-four hub genes associated with OSCC were identified via the Venn diagram (Table 4). Seventeen genes were upregulated; 7 were downregulated.

Table 4.

The Top Core Genes Screened in 0SCC Based on the Degree/MCC/MNC/EPC.

Expression status	Gene symbol	Gene description
SCC up	AURKA
	MMP2
	PLK1
	BRCA1
	CXCL8
	TGFB1
	CXCL10
	CD274
	IL1B
	FN1
	CCL5
	COL1A1
	CSF2
	STAT1
	MMP9
	PXDN
	TLR2
SCC down	NDUFS3
	ACADM
	ACOX1
	CYCS
	SUCLG1
	ATP5F1A
	MDH2

Hub Gene Enrichment

KEGG pathway enrichment analysis was performed to identify the key genes connected to SCC. The top-ranked genes identified in our analysis are considered hub genes central to the development of SCC. The reanalysis indicated that the Toll-like receptor (TLR) signaling pathway plays a significant role in the development of oral SCC and CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2 genes were up-regulated and enriched significantly in the signaling pathways’ interactions in oral SCC (P < .05, Figure 2).

Figure 2.

Analysis of the top genes SCC by KEGG pathway enrichment.

Expression Analysis of Core Genes in SCC

The 6 hub genes’ expression identified in the core gene analysis was validated using the GEPIA database (P < .05, Figure 3).

Figure 3.

Differentially expressed 6 hub genes in OSCC patients, compared to healthy controls according to the GEPIA online database (P < .05).

Prognosis and Survival Rates of the Core Genes in OSCC

The Kaplan-Meier plotter was utilized to evaluate the prognostic significance of the top genes associated with OSCC, determining their relation to low survival outcomes in OSCC patients (Figure 4).

Figure 4.

The Kaplan-Meier plotter used to determine the prognostic value of 10 core genes in SCC related to the survival rate.

Mutational Analysis of Hub Genes Involved in OSCC

Signal transducer and activator of transcription 1 (STAT1) was the most significant gene linked to amplification mutations according to a study’s findings performed in 2013²⁵ on hub genes involved in OSCC conducted by MD Anderson, Cancer Discov, which involved the mutational analysis of 40 samples (Figure 5).

Figure 5.

Genetic mutation analysis of hub genes in OSCC. The mutation rate is 2.5% for STAT1 and 0% for other genes.

Bipartite miRNA-mRNA Network Analysis

The mirWalk database predicted potential microRNAs, possibly interacting with specific hub genes (CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2) to examine their function in the development of OSCC. Additionally, the miRTarBase filter was taken into consideration. Four miRNAs that are significant in OSCC were found by evaluating the shared miRNAs among these genes (Table 5 and Figure 6).

Table 5.

The miRNAs Involved in the Hub Genes.

miRNA	Gene	Total
hsa-miR-146a-5p	CCL5, CXCL8, TLR2, STAT1	4
hsa-miR-155-5p	CXCL10, CXCL8, IL1B, STAT1	4

Figure 6.

Bipartite miRNA‒mRNA regulatory network of important miRNAs involved in SCC.

Validation of the Key Candidate DEGs

To analyze cancer transcriptome data, UALCAN (http://ualcan.path.uab.edu/index.html) is utilized. The expression of several genes in various tumor types and the correlation between genes and prognosis may be obtained by UALCAN through comprehensive studies of TCGA gene expression data¹² The UCLAN database received the potential important genes, and the association between the expression of these genes and the prognosis of OSCC was confirmed using the TCGA data (accession date: November 21, 2018).

Discussion

Oral squamous cell carcinoma (OSCC) is among the most prevalent types of head and neck squamous cell carcinoma (HNSCC)¹ Although OSCC is characterized by high mortality, aggressive invasion, and rapid metastasis, reliable biomarkers and therapeutic targets for early diagnosis and effective treatment remain limited. Identifying genes differentially expressed between OSCC and normal tissues can provide valuable insights into the underlying molecular mechanisms and lead to the discovery of potential diagnostic and therapeutic targets.

Advancements in microarray and high-throughput sequencing technologies have enabled large-scale genomic analyses, offering new opportunities to understand the molecular basis of OSCC. This comprehensive bioinformatics study aimed to integrate multiple datasets to overcome the limitations of individual studies and provide a more robust and accurate molecular landscape of OSCC. By combining data from multiple populations and platforms, this analysis contributes to improving our understanding of OSCC pathogenesis and supports the development of new diagnostic and therapeutic strategies.

In this study, 3 OSCC gene expression datasets comprising 224 tumor samples and 101 normal controls were analyzed. Venn diagram analysis identified 941 upregulated and 1157 downregulated differentially expressed genes (DEGs) common to all datasets. Subsequent bioinformatics analyses—including PPI network construction, enrichment analysis, module identification, and hub gene selection—highlighted several key genes and pathways potentially involved in OSCC progression. The Toll-like receptor (TLR) signaling pathway, in particular, appeared to play a crucial role, and 6 hub genes (CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2) were identified as key molecular markers.

Because gene expression can vary across populations, our inclusion of datasets from the USA, Taiwan, and Italy helped minimize sample heterogeneity and identified DEGs consistently associated with OSCC across different ethnic backgrounds. This cross-population validation strengthens the robustness and potential generalizability of our findings.

Several previous bioinformatics studies have explored molecular alterations in OSCC.^5,6 For instance, Xu et al²⁶ identified 651 DEGs enriched in cytokine–cytokine receptor interaction and xenobiotic metabolism pathways suggesting potential biomarker roles for CSF2 and EGF. Similarly, Zou et al²⁷ identified 34 DEGs across 4 datasets, emphasizing SPP1 and PLAU in OSCC cell migration and proliferation. Mathavan et al⁷ reported 9 hub genes, including APP and EHMT1, associated with OSCC and demonstrated that fisetin may regulate these genes. Compared with these studies, our analysis integrated 3 independent datasets to reveal 6 hub genes highly consistent across populations and biologically relevant to inflammation and tumor immune responses, thus providing broader and more reproducible results.

The hub genes identified in our study play important roles in inflammation, immune modulation, and tumor progression. For example, CXCL8 and CXCL10, both members of the chemokine family, have been shown to promote cancer development and predict patient outcomes by mediating pro-inflammatory and pro-oncogenic signaling^28,29 Elevated CXCL8 levels in endothelial–tumor cell co-cultures suggest its involvement in promoting tumorigenesis³⁰ TLR2, another identified hub gene, is known to regulate immune responses in OSCC and other epithelial malignancies. While TLR2 activation on inflammatory cells may promote antitumor immunity, its overexpression in tumor keratinocytes can enhance resistance to apoptosis and favor tumor survival^31
-34 Furthermore, STAT1 is widely recognized for its role in inducing apoptosis and cell cycle arrest through immune-mediated signaling, and its activation status has been linked to OSCC prognosis^35,36 Similarly, IL1B, a pro inflammatory cytokine, contributes to extracellular matrix degradation and OSCC invasiveness by promoting MMP production.^37,38 Lastly, CCL5 promotes tumor invasion and migration by interacting with its receptor CCR5 and upregulating MMP-9 expression in oral cancer cells.^39
-42

Together, these findings underscore the importance of immune and inflammatory pathways—particularly chemokine signaling, STAT1-mediated apoptosis, and TLR activation—in the pathogenesis and progression of OSCC. Targeting these pathways may provide new opportunities for therapeutic intervention.

Despite these promising findings, several limitations should be acknowledged. First, our study primarily relied on publicly available transcriptomic datasets and bioinformatics analyses. Experimental validation, such as in vitro and in vivo functional assays, is still required to confirm the biological roles of the identified hub genes. Second, although integrating datasets from different populations reduced heterogeneity, potential batch effects and sample biases cannot be completely ruled out. Third, only 6 hub genes were investigated in depth; other DEGs may also play important roles and warrant further exploration. Additionally, the mutation analysis was based on 40 OSCC samples from the MD Anderson 2013 dataset. The relatively small sample size may limit the accuracy of estimated mutation frequencies, and these findings should therefore be interpreted with caution. Finally, a larger clinical cohort is needed to validate the diagnostic and prognostic utility of these hub genes in independent patient populations. The hub genes and pathways identified in this study, including CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2, may serve as potential biomarkers for OSCC diagnosis or prognosis. Their involvement in immune and inflammatory signaling suggests they could also represent therapeutic targets, particularly for interventions aimed at modulating the tumor microenvironment. While these findings are currently based on bioinformatics analyses, they provide a foundation for future experimental studies and clinical validation, which could ultimately support the development of targeted therapies or predictive biomarker panels in OSCC patients. Future research should therefore focus on experimental and clinical validation to translate these findings into practical clinical applications.

Conclusion

Using integrated bioinformatics analysis, this study identified 941 upregulated and 1157 downregulated differentially expressed genes (DEGs) across 4 OSCC datasets. These DEGs were mainly associated with Toll-like receptor (TLR) and chemokine signaling pathways. Six hub genes—CXCL8, CCL5, CXCL10, STAT1, IL1B, and TLR2—were highlighted as key molecules potentially involved in OSCC pathogenesis. While several of these genes, such as STAT1 and CXCL8, have been previously reported to contribute to OSCC progression, our integrated analysis reinforces their importance across multiple independent datasets and provides additional network- and pathway-level insights. These results enhance the current understanding of the molecular mechanisms underlying OSCC and may assist in identifying biomarkers or therapeutic candidates worthy of further validation. Nevertheless, additional experimental and clinical studies are required to confirm the functional roles of these genes and to explore their potential translational applications in OSCC diagnosis and therapy.

Footnotes

Acknowledgements

The authors thank to the Tehran University of Medical Science for their financial support.

Author’s Note

Mohammad Reza Eskandarion is now affiliated with Legal Medicine Research Center, Legal Medicine Organization, Tehran, Iran.

ORCID iD

Mohammad Reza Eskandarion

Ethical Considerations

This study was conducted using publicly available datasets and in accordance with relevant ethical guidelines and regulations. Ethical approval was obtained from the research Ethics Committees of Amiralam Hospital Tehran University of Medical Sciences (IR.TUMS.AMIRALAM.REC.1402.011).

Consent to Participate

The study used publicy available datasets; therfore, informed consent was not required.

Consent for Publication

Written informed consent was obtained from all participants included in the study.

Author Contributions

MR. E is a PhD candidate in Cancer Research Institute, Tehran University of Medical Sciences, Tehran, Iran, and he was involved in writing-original draft, Visualization, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization with support from M.A, E.K that were supervised the total of this project and monitoring all processes of function as the head of the team in collaboration with M.VR. SH.EM and SF.MH were two researchers that investigated the summary of papers for extraction of the data and quality evaluation. H.M and A.MK was involved in review & editing. All authors contributed to the article and approved the submitted version.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported by the Amiralam Hospital Tehran University of Medical Sciences.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets used during the current study, were deposited into the Gene Expression Omnibus database under accession number (GSE37991, GSE30784, and GSE107591 datasets) are available at the following URL .

References

Warnakulasuriya

Global epidemiology of oral and oropharyngeal cancer. Oral Oncol. 2009;45(4-5):309-316.

Masaoka

Shinozuka

Ohara

Tsuda

Imai

Tonogi

Bioinformatics analysis of dysregulated exosomal microRNAs derived from oral squamous cell carcinoma cells. J Oral Sci. 2021;63(2):174-178.

Scully

Bagan

Oral squamous cell carcinoma: overview of current understanding of aetiopathogenesis and clinical implications. Oral Dis. 2009;15(6):388-399.

Montero

Patel

SG.

Cancer of the oral cavity. Surg Oncol Clin. 2015;24(3):491-508.

Shield

Ferlay

Jemal

, et al. The global incidence of lip, oral cavity, and pharyngeal cancers by subsite in 2012. CA Cancer J Clin. 2017;67(1):51-64.

Kerawala

Roques

Jeannon

Bisase

Oral cavity and lip cancer: United Kingdom national multidisciplinary guidelines. J Laryngol Otol. 2016;130(s2):S83-S89.

Mathavan

Kue

Kumar

Identification of potential candidate genes for lip and oral cavity cancer using network analysis. Genom Inform. 2021;19(1):e4.

Siegel

Miller

Jemal

Cancer statistics, 2018. CA Cancer J Clin. 2018;68(1):7-30.

Clough

Barrett

. The gene expression omnibus database. In: Statistical Genomics: Methods and Protocols. 2016:93-110.

10.

Xie

Bailey

Kuleshov

, et al. Gene set knowledge discovery with Enrichr. 2021;1(3):e90.

11.

Kuleshov

Jones

Rouillard

, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44(W1):W90-W7.

12.

Wang

Liu

Gong

Cheng

Establishment and validation of an aging-related risk signature associated with prognosis and tumor immune microenvironment in breast cancer. Eur J Med Res. 2022;27(1):317.

13.

Eskandarion

Eskandarieh

Shakoori Farahani

, et al. Prediction of novel biomarkers for gastric intestinal metaplasia and gastric adenocarcinoma using bioinformatics analysis. Heliyon. 2024;10(9):e30253.

14.

Huang

Sherman

Lempicki

RA.

Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44-57.

15.

Lin

Huber

Kibbe

WA.

Model-based variance-stabilizing transformation for illumina microarray data. Nucleic Acids Res. 2008;36(2):e11.

16.

Szklarczyk

Kirsch

Koutrouli

, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023;51(D1):D638-D46.

17.

Zhang

, et al. Identification of a potential bioinformatics-based biomarker in keloids and its correlation with immune infiltration. Eur J Med Res. 2023;28(1):476.

18.

Shannon

Markiel

Ozier

, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498-2504.

19.

Chin

C-H

Chen

S-H

H-H

C-W

M-T

Lin

C-Y.

CytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 2014;8 Suppl 4:1-7.

20.

Tang

Kang

Gao

Zhang

GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017;45(W1):W98-W102.

21.

Cerami

Gao

Dogrusoz

, et al. The cbio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2(5):401-404.

22.

Gao

Aksoy

Dogrusoz

, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cbioportal. Sci Signal. 2013;6(269):l1.

23.

Guo

Chang

Huang

, et al. Mutation hotspots at CTCF binding sites coupled to chromosomal instability in gastrointestinal cancers. Nat Commun. 2018;9(1):1520.

24.

Kakiuchi

Nishizawa

Ueda

, et al. Recurrent gain-of-function mutations of RHOA in diffuse-type gastric carcinoma. Nat Genet. 2014;46(6):583-587.

25.

Pickering

Zhang

Yoo

, et al. Integrative genomic characterization of oral squamous cell carcinoma identifies frequent somatic drivers. Cancer Discov. 2013;3(7):770-781.

26.

Jiang

Identification for exploring underlying pathogenesis and therapy strategy of oral squamous cell carcinoma by bioinformatics analysis. Med Sci Monit. 2019;25:9216-9226.

27.

Zou

, et al. Identification of key candidate genes and pathways in oral squamous cell carcinoma by integrated bioinformatics analysis. Exp Ther Med. 2019;17(5):4089-4099.

28.

Yang

Sun

Wei

Lin

The role of CXC chemokines in cancer progression. Cancers. 2022;15(1):167.

29.

Tokunaga

Zhang

Naseem

, et al. CXCL9, CXCL10, CXCL11/CXCR3 axis for immune activation – a target for novel cancer therapy. Cancer Treat Rev. 2018;63:40-47.

30.

Christofakis

Miyazaki

Rubink

Yeudall

WA.

Roles of CXCL8 in squamous cell carcinoma proliferation and migration. Oral Oncol. 2008;44(10):920-926.

31.

Apetoh

Ghiringhelli

Tesniere

, et al. Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med. 2007;13(9):1050-1059.

32.

Sheu

JJ-C

Hua

C-H

Wan

, et al. Functional genomic analysis identified epidermal growth factor receptor activation as the most common genetic event in oral squamous cell carcinoma. Cancer Res. 2009;69(6):2568-2576.

33.

Rich

Hussaini

, et al. Toll-like receptor 2 is present in the microenvironment of oral squamous cell carcinoma. Br J Cancer. 2011;104(3):460-463.

34.

Ikehata

Takanashi

Satomi

, et al. Toll-like receptor 2 activation implicated in oral squamous cell carcinoma development. Biochem Biophys Res Commun. 2018;495(3):2227-2234.

35.

Zhang

Molavi

Lai

The clinical and biological significance of STAT1 in esophageal squamous cell carcinoma. BMC Cancer. 2014;14:1-14.

36.

Laimer

Spizzo

Obrist

, et al. STAT1 activation in squamous cell cancer of the oral cavity: a potential predictive marker of response to adjuvant chemotherapy. Cancer. 2007;110(2):326-333.

37.

Sherwood

Toliver-Kinsky

Mechanisms of the inflammatory response. Best Pract Res Clin Anaesthesiol. 2004;18(3):385-405.

38.

Kamatani

Shiogama

Yoshihama

Kondo

Shirota

Shintani

Interleukin-1 beta in unstimulated whole saliva is a potential biomarker for oral squamous cell carcinoma. Cytokine. 2013;64(2):497-502.

39.

Korbecki

Grochans

Gutowska

Barczak

Baranowska-Bosiacka

CC chemokines in a tumor: a review of pro-cancer and anti-cancer properties of receptors CCR5, CCR6, CCR7, CCR8, CCR9, and CCR10 ligands. Int J Mol Sci. 2020;21(20):7619.

40.

McCormack

Moriarty

O’Donoghue

McCormick

Sheahan

Baird

AW.

Tissue cytokine and chemokine expression in inflammatory bowel disease. Inflamm Res. 2001;50:491-495.

41.

Wang

S-W

Liu

S-C

, et al. CCL5 and CCR5 interaction promotes cell motility in human osteosarcoma. PLoS One. 2012;7(4):e35101.

42.

Chuang

Yang

Chen

, et al. CCL5/CCR5 axis promotes the motility of human oral cancer cells. J Cell Physiol. 2009;220(2):418-426.