Non-Coding RNAs in Oral Cancer: Emerging Roles and Clinical Applications

Simple Summary Oral cancer (OC) is one of the most prevalent cancers in the world. Despite improvements in therapies, OC still has a poor survival rate of about 50%, with metastasis being the worst-case scenario. Thus, there is an urgent need to understand the disease process and to develop diagnostic and therapeutic strategies for OC. Advancement of high throughput genome sequencing shows that more than 90% of the human genome encodes non-coding transcripts that do not code for any protein. In this review, we discuss the role of various types of these non-coding RNAs (ncRNAs) in OC and their promising clinical implications. Dysregulated expressions of ncRNAs are associated with OC initiation and progression, as well as therapy resistance. Differential expressions of these ncRNAs in blood or saliva have indicated their potential diagnostic and prognostic importance. In this review, we have summarized all the promising aspects of ncRNAs in the management of OC. Abstract Oral cancer (OC) is among the most prevalent cancers in the world. Certain geographical areas are disproportionately affected by OC cases due to the regional differences in dietary habits, tobacco and alcohol consumption. However, conventional therapeutic methods do not yield satisfying treatment outcomes. Thus, there is an urgent need to understand the disease process and to develop diagnostic and therapeutic strategies for OC. In this review, we discuss the role of various types of ncRNAs in OC, and their promising clinical implications as prognostic or diagnostic markers and therapeutic targets. MicroRNA (miRNA), long ncRNA (lncRNA), circular RNA (circRNA), PIWI-interacting RNA (piRNA), and small nucleolar RNA (snoRNA) are the major ncRNA types whose involvement in OC are emerging. Dysregulated expression of ncRNAs, particularly miRNAs, lncRNAs, and circRNAs, are linked with the initiation, progression, as well as therapy resistance of OC via modulation in a series of cellular pathways through epigenetic, transcriptional, post-transcriptional, and translational modifications. Differential expressions of miRNAs and lncRNAs in blood, saliva or extracellular vesicles have indicated potential diagnostic and prognostic importance. In this review, we have summarized all the promising aspects of ncRNAs in the management of OC.


Introduction
Oral squamous cell carcinoma (OSCC) or oral cancer (OC) is the most prevalent type of head and neck cancer that arises in the tongue, lips, and floor of the mouth. In the year 2020, GLOBOCAN estimated around 377,713 total cases and 177,757 deaths worldwide from lip and oral cavity cancer, in which India alone showed a high burden of the disease [1]. The American Cancer Society has recently estimated the incidence of cancers in the oral cavity and pharynx, with around 54,000 new cases and about 11,580 deaths in the year 2023 alone [2]. Some of the most common factors with which OC progression is often To date, RNAseq studies have revealed more than 200 ncRNAs (including miRNA, lncRNA, circRNA, snoRNA, and piRNA) to have an association with OC progression. In this review, we have described the types of ncRNAs, their functions, and their possible role as diagnostic and prognostic markers in OC. The correlations between the expressions of these ncRNAs and OC have been elaborated further in this review. Results from different studies obtained so far suggest a better understanding of the relationship between non-coding RNA transcriptomic alterations and disease development, since these modifications have the prospective to be used as diagnostic and therapeutic biomarkers. Thus, the review has practical implications for the treatment of OC and may offer fresh perspectives and ideas for future mechanistic research in various preclinical and clinical settings.

Oral Cancer: Current Diagnostic and Prognostic Markers
Tissue biopsy and histological evaluation are the gold standard for diagnosing oral cancer (OC). However, this technique is painful for patients and causes a delayed diagnosis [11]. As non-invasive methods, metachromasia using iodine staining, or toluidine blue, which stains cancerous lesions, chemiluminescence-based lumenoscopy, auto-fluorescence based techniques like laser-induced auto-fluorescence (LIAF) or visually enhanced lesion scope (Velscope), and optical coherence tomography (OCT) provide assistance in early diagnosis and identification of oral pathophysiologic lesions [11]. The advancement of complete human genome insight, and the numerous potentials of cellular and epigenetic research, can be utilized as prognostic and diagnostic techniques for conducting rapid evaluation and treatment of oral lesions. The molecular diagnostic measures are divided into two types: nucleic acid-associated markers and proteinassociated markers. In recent years, several biomarkers have been identified that could be utilized for the prognosis, diagnosis, differential diagnosis, prediction of recurrence, distant metastasis, and chemotherapy or radiotherapy resistance of OC. The biomarkers in OC include up-regulated expression of epidermal growth factor receptor (EGFR), vascular endothelial growth factor receptor (VEGFR), matrix metalloproteinases (MMPs), proliferating cell nuclear antigen (PCNA), Ki-67, Cyclin D1, Cathepsin-d, CD44, cytokeratins, p53 and p16 mutation or expression status, and the prevalence of human papilloma virus (HPV) and its oncogenes [12]. These biomarkers can aid in screening and

Oral Cancer: Current Diagnostic and Prognostic Markers
Tissue biopsy and histological evaluation are the gold standard for diagnosing oral cancer (OC). However, this technique is painful for patients and causes a delayed diagnosis [11]. As non-invasive methods, metachromasia using iodine staining, or toluidine blue, which stains cancerous lesions, chemiluminescence-based lumenoscopy, auto-fluorescence based techniques like laser-induced auto-fluorescence (LIAF) or visually enhanced lesion scope (Velscope), and optical coherence tomography (OCT) provide assistance in early diagnosis and identification of oral pathophysiologic lesions [11]. The advancement of complete human genome insight, and the numerous potentials of cellular and epigenetic research, can be utilized as prognostic and diagnostic techniques for conducting rapid evaluation and treatment of oral lesions. The molecular diagnostic measures are divided into two types: nucleic acid-associated markers and protein-associated markers. In recent years, several biomarkers have been identified that could be utilized for the prognosis, diagnosis, differential diagnosis, prediction of recurrence, distant metastasis, and chemotherapy or radiotherapy resistance of OC. The biomarkers in OC include up-regulated expression of epidermal growth factor receptor (EGFR), vascular endothelial growth factor receptor (VEGFR), matrix metalloproteinases (MMPs), proliferating cell nuclear antigen (PCNA), Ki-67, Cyclin D1, Cathepsin-d, CD44, cytokeratins, p53 and p16 mutation or expression status, and the prevalence of human papilloma virus (HPV) and its oncogenes [12]. These biomarkers can aid in screening and early detection and prognosis of OC, which can improve treatment outcomes by allowing early intervention. However, these biomarkers are not enough as a primary diagnostic tool, and can only be useful in combination with other diagnostic methods to confirm malignancy and its stage. In addition, these markers are often not organ-specific. Therapeutic intervention by targeting these biomarkers shows limited efficacy and side effects.

Diverse ncRNAs in OC
Multiple ncRNAs have been identified in association with OC growth, invasion, migration, and therapy resistance. The importance of the ncRNAs in OC is summarized below.

Role of miRNAs in Oral Cancer
The effect of microRNAs (miRNAs) has gained considerable attention in the past few years for regulating the biological processes of multiple malignancies. The miRNAs are 19-25 nucleotide small RNA molecules that mainly interact with the 3 UTR of the target mRNA, and thus regulate gene expression. However, interactions with 5 UTR, promoter, and coding region have also been reported [13]. The miRNA can function as an oncogene or tumor suppressor gene, and the deregulation of miRNAs is seen in many cancers, including OC. Defects in miRNA biogenesis machinery, alterations in miRNA genes or transcriptional regulation, or epigenetic regulations are often associated with the miRNA deregulation mechanism.

Role of Long Non-Coding RNAs in Oral Cancer
Long non coding RNAs (LncRNAs) are more than 200 nucleotides long and they have similar biogenesis to messenger RNAs (mRNAs). They are transcribed by RNA-Polymerase II (Pol II), and undergo polyadenylation, splicing, and 5 -capping [115]. LncRNAs are localized either in the nucleus, cytoplasm, or both. Depending on the localization, they perform a variety of cellular processes by interacting with DNAs, RNAs, and proteins such as epigenetic modification, transcriptional regulation, RNA splicing, mRNA stabilization, translational regulation, sequestration of proteins, protein stabilization, and miRNA sponges [115][116][117][118]. Due to these regulatory functions, lncRNAs are associated with numerous disorders such as diabetes, inflammatory bowel diseases, AIDS, neurodegenerative diseases, different blood-related disorders, as well as cancers [119,120].
The lncRNAs exhibit aberrant expression patterns in OC, and play a significant role in the advancement of the disease [121]. In OC, the lncRNAs may either act as oncogenes or as tumor suppressor genes. The functions of some reported lncRNAs are summarized in Table 2. Oncogenic lncRNAs, such as lncRNAs MALAT1, NEAT1, PVT1, ELDR, DLEU1, HOTAIR, HIFCAR, PCAT1, DANCR, etc., are associated with tumor proliferation, invasion, metastasis, angiogenesis, and drug resistance. For example, high MALAT1 expression is correlated with OC recurrence and metastasis [122]. It has also been demonstrated that MALAT1 promotes cellular proliferation and metastasis via controlling multiple signaling events, including Wnt/β-catenin signaling and PI3K/AKT/mTOR signaling pathways [122]. MALAT-1 is present in circulation, and shows importance as a serum biomarker. The MALAT1 is associated with resistance in multiple drugs, such as cisplatin, 5-fluorouracil, and paclitaxel [122]. Currently, the lncRNA MALAT1 is in OC clinical trial. The MALAT1 promoter contains transcription factors SP1 and SP3 binding sites [123]. Cooperative function of SP1 and SP3 activates MALAT1 expression [123]. In addition, STAT3/ TGF-β signaling axis, hypoxia-inducible factor (HIF)-2α, Yes-associated oncoprotein (YAP)-1, Jumonji C-domain-containing protein (JMJD)-1A, Octamer-binding transcription factor (OCT)-4, chemokine (C-C motif) ligand (CCL)-5, β-catenin, and Lysine-specific demethylase (KDM)-5B are involved in up-regulation of MALAT1 in OC and other cancers [123]. Another lncRNA ELDR is found to be up-regulated in OC cell lines and patient samples [124]. The ELDR inhibits miRNA-7, resulting in stabilization of EGFR. In addition, the ELDR interacts with RNA binding protein Interleukin Enhancer Binding Factor 3 (ILF3), and stabilizes cell cycle gene Cyclin E1. Interestingly, targeted inhibition of ELDR could inhibit in vivo tumor growth in a mouse model [124]. Overexpression of ELDR in normal oral keratinocytes (NOKs) induces cell proliferation and G2/M cell cycle progression through activation of CTCF/FOXM1/AURKA axis showing importance of the lncRNA as one of the OC driver genes [125]. It is not clear why the ELDR is up-regulated in OC; however, around 11% of samples contain the gene amplification in TCGA database (www.cbioportal.org, accessed on 28 June 2023).
On the other hand, the lncRNAs MEG3, GAS5, FENDRR, and PTCSC3 are downregulated in OC. For example, the down-regulation of the lncRNA GAS5 increases miR-21 expression and helps in proliferation, invasion, EMT, and migration [126]. MEG3, being another tumor-suppressive lncRNA, suppresses miR-421 and the Wnt/β-catenin pathway [127,128]. Down-regulation of the lncRNA helps in the induction of the cell cycle, cell proliferation, metastasis, and suppression of cell apoptosis. Promoter methylation of MEG3 and GAS5 is one of the mechanisms of down-regulation in cancers [129,130]. As stated in Table 2, many lncRNAs are associated with OC diagnosis and prognosis, and are thus suggested to be OC biomarkers and therapeutic targets.      Therapeutic target [192] lncRNA BBOX1 ↑ Encourages proliferation and migration, and suppresses apoptosis miR-3940-3p/laminin subunit gamma 2 axis Tissue, saliva and cell lines Therapeutic target [193] lncRNA IFITM4P ↑ Induces cell proliferation and enhanced immune escape

Role of Circular RNAs (circRNA) in Oral Cancer
CircRNAs are closed-loop, extremely stable ncRNA molecules with no 3 or 5 ends and a poly (A) tail. They belong to the lncRNA kingdom, and have a higher half-life compared to linear RNAs [222,223]. Back-splicing and exon skipping of pre-mRNAs are the two processes by which circRNAs are produced. The transcription process is carried out via RNA pol II [224]. These RNA structures have the ability to withstand exonucleolytic breakdown by RNase R. It has been seen that 80% of circRNAs are present in the cytoplasm. However, the presence of circRNAs in the nucleus makes the control of gene expression possible [225].
Due to their closed-loop orientation, tissue specificity, high stability, and conservation, they serve as significant biomarkers for several diseases. These RNA molecules perform several regulatory functions, such as miRNA sponging, direct protein binding, and certain circRNAs are even translated into proteins [226,227]. Numerous studies have revealed that the majority of aberrantly expressed circRNAs play a significant role in controlling the progression of cancer by influencing a number of cancer hallmarks. Proliferative signaling, encouraging tumor and antitumor immunity, triggering angiogenesis, promoting invasion, metastasis, and deregulating cellular energetics are some functions that are associated with an aberrant circRNA profile [228].
Research on circRNA in OC has increased in recent years, and it has been discovered that these RNA molecules have a significant influence on the development, management, and prognosis of OC [229]. Like miRNA and lncRNA, these RNA molecules also play a role as both oncogenes and tumor suppressors in OC. For example, circ_0002185, circ_PVT1, circ_100290, circ_0001742, circ_HIPK3, circ_0001971, circ_DOCK1, circ_FLNA, circ_GOLPH3, circ_CLK3, circ_CDR1, circ_0014359, circ_LPAR3, circ_SEPT9, etc., are up-regulated, whereas circ_0000140, circ-PKD2, circ_0005379, circ_0004491, circ_SPATA6, circ_0086414, circ_0008309, circGDI2, circ_0007059, etc., are down-regulated in OC (Table 3). The mechanism of differential expression of the circRNA in OC is not clear. The oncogenic circular RNA Circ_100290 acts as competing endogenous RNA (ceRNA), and inhibits miR-378a mediated suppression of glucose transporter GLUT1, resulting in the induction of glycolysis and cell growth [230]. The circ_PVT1 is derived from exon 3 of the oncogene of lncRNA PVT1 [231]. Recently, it was discovered that the mutant p53/ YAP/ TEAD transcription-competent complex is responsible for the up-regulation of circ_PVT1 in head and neck squamous cancer [231]. Through sponging miR-125b, Circ_PVT1 worked as a competitive endogenous RNA (ceRNA) to induce STAT3 signaling and cell proliferation [231]. Another circular RNA, called Circ_CDR1, has been stated to encourage autophagy under the hypoxic condition to enhance cell survival in OC, via control of the AKT/ERK-1/2/mTOR signaling pathway [232]. On the contrary, a study showed the association between suppressed expression of circ_0007059 and OC, via regulation of the AKT/mTOR pathway [233]. A high throughput sequencing study of OC samples identified significant down-regulation of circ_0005379 as compared to the adjacent normal tissues [234]. Up-regulation of circ_0005379 enhances cetuximab sensitivity, efficiently reduces OC proliferation, migration, invasion, and angiogenesis in vitro, and slows tumor growth in nude mice via inhibiting EGFR signaling [234]. All these studies suggest that these RNA molecules regulate OC via control over the major signaling pathways like MAPK, WNT/β-catenin, Notch, VEGF, and PI3K/AKT in OC [235]. The aggressive trait of OC was found to be linked to circ_PKD2 down-regulation. Overexpression of circ_PKD2 induces cell cycle arrest and apoptosis, and inhibited proliferation, migration, and invasion of OC through inhibiting miR-204-3p [236]. All these studies indicate the potential role of circRNA in OC. Table 3 enlists several other circRNAs whose aberrant expression level regulates OC.     Therapeutic target [268] Up arrows ↑ indicate upregulation and down arrows ↓ indicate downregulation.

Role of Small Nucleolar RNA (SnoRNA) in Oral Cancer
SnoRNAs are one of the many classes of non-coding RNA molecules present in the body. There are around 300 snoRNA sequences identified in the human genome. Although, snoRNAs are small in size, they are present in large quantities within the nucleus of cells [269]. Most snoRNAs are expressed in the intron of both coding and noncoding genes, while the remaining gets transcribed by RNA polymerase II (RNA pol II). Splicing, debranching, and co-transcription are the stages involved in their biogenesis. They play diverse roles in the development of ribosomal, small nuclear, and other pre-mRNA molecules via endonucleolytic disintegration and post-transcriptional regulation [269]. They also have the ability to control gene expressions by modifying and splicing mRNA. The snoRNAs form small nucleolar ribonucleoprotein complexes (snoRNP complexes) by binding to protein molecules, which then leads to the modification of rRNA bases [269].
SnoRNAs have roles in a variety of pathological and physiological processes. Studies have shown that snoRNAs control tumor growth, invasion, and metastasis, as well as cell death during the carcinogenesis process. More importantly, snoRNAs play a significant role in the development of OC tumors. Today, the differential expression of snoRNAs in OCs leads to the possibility of them being used as diagnostic and prognostic biomarkers [270]. However, the mechanism of differential expression of snoRNA in OC is not known clearly. Alteration in snoRNA biogenesis and post-transcriptional regulation may be involved in differential expression of different snoRNAs in OC. An in silico investigation using RNA seq data of 567 samples from the TCGA head and neck cancer cohort could identify 113 snoRNAs using p < 0 .05 as the cut-off [271]. The top significantly modulated snoRNAs were associated with DNA template regulation, RNA editing, regulation of cell proliferation, adhesion, invasion, metastasis, PI3K-AKT signaling, EMT, and angiogenesis pathways. Further analysis with the top five snoRNAs (SNORD114-17: ENSG00000201569, SNORA36B: ENSG00000222370, SNORD78: ENSG00000212378, U3: ENSG00000212182, and U3: ENSG00000212195) showed association with patient survival, indicating the importance of snoRNAs in disease progression and as biomarkers in OC [271]. A microarray analysis from eight OC samples identified 16 significantly modulated snoRNAs as compared to control samples; among them, 15 were significantly down-regulated and associated with patient survival [270]. The SNHG3, a snoRNA that gets up-regulated in OC patients, induces migration and cell proliferation of oral squamous cells. It targets the nuclear transcription factor-Y subunit gamma (NFYC) via the SNHG3/ miR-2682-5p axis and functions as a biomarker [272,273]. SnoRNA SNHG15 also gets overexpressed in OC cell lines, and facilitates the malignant behaviors of OC via miR-188-5p/ DAAM1 as a target [274]. Thus, snoRNAs play a role in assisting tumor growth in OC. Further studies can increase their significance in cancer therapy in the future. Table 4 enlists expressions and functions of some snoRNAs in OC.

Role of piRNAs in Oral Cancer
PIWI-interacting RNAs (piRNAs) are a subclass of ncRNAs that can be divided into three primary categories: transposon-derived piRNAs, mRNA-derived piRNAs, and lncRNA-derived piRNAs. They are 24-31 nucleotides long, have 5 -end uridine or 10th position adenosine bias, and lack proper secondary structural features [278]. The piRNAs, composed of an array of different nucleotide sequences, are single-stranded ncRNAs that interact with P-element-induced wimpy testis (PIWI) proteins [279]. They are the largest group of ncRNAs, and are multi-functional. PiRNAs are instrumental in genome rearrangement, spermiogenesis, protein regulation, transposon silencing, epigenetic regulation, and germ stem-cell maintenance by binding to PIWI proteins to make a piRNA/PIWI complex [280].
The piRNAs are mostly known to be expressed in germ cells; however, their existence is also observed in cancer cells. Therefore, the question of employing these RNA molecules as a prognostic marker or therapeutic target arises. Current research has given evidence of the piRNAs/PIWI complex being used for the occurrence, development, metastasis, and recurrence of breast cancer [281] and lung cancer [282]. Some piRNAs have been found to have a role in the development of OC, and can be a potential biomarker or therapeutic target for OC in the future [283]. In the OC mouse model, piR354, piR415, piR832, and piR1584, have been found to interact with mRNA molecules [284]. Longer survival of patients with head and neck cancer is associated with low levels of piR-58510 and piR-35373 [285]. It is observed that genes like GALNT6, SPEDF, and MYBL2 that are paired with piRNAs are responsible for the suppression or progression of several OC tumors [284]. An in silico analysis using the RNA sequencing data of 455 head and neck cancer samples, and 43 matched non-tumors from The Cancer Genome Atlas (TCGA), showed a total of 305 piRNAs in both tumor and non-tumor tissues [286]. Among a total of 247 significantly altered genes, 25 piRNAs were exclusively expressed in non-tumor samples, and 87 were only expressed in tumors. The significantly up-regulated piRNAs, including the topmost gene FR140858, were associated with poor patient survival. This indicates the importance of piRNAs in OC as diagnostic and prognostic biomarkers [286]. Another study identified a panel of 30 piRNAs in 77 HPV positive head and neck cancer samples from the TCGA RNA seq data [287]. Simultaneous validation in cell lines further reported key piRNAs NONHSAT077364, NONHSAT102574, and NONHSAT128479 in HPV associated head and neck cancer development. Based on analysis of the tongue cancer GEO database (GSE196674 and GSE196688), 406 differentially expressed piRNAs were identified [288]. Further investigation identified a down-regulated piRNA: piR-33422 and its association with mevalonate/ cholesterol-pathway-related gene FDFT1 in tongue cancer. Using TCGA RNA seq data of 256 smoking-related head and neck cancer samples, a panel of 13 piRNAs were identified [289]. Among them, NONHSAT123636 and NONHSAT113708 were found to be associated with tumor stage, NONHSAT067200 with patient survival, and 6 other piRNAs with TP53 mutation and 3q26, 8q24, and 11q13 amplification. Further studies are needed to know the regulation of their expression and functional mechanism of piRNAs in OC.

NcRNAs in Oral Cancer Progression
OSCC or OC is a multistep process originating from epithelial cells by progressive accumulation of genetic and epigenetic alterations. The histological changes that occur during the carcinogenesis begin with atypical squamous cell hyperplasia to carcinoma in situ (CIS) through stages of dysplasia [290]. The molecular events associated with alterations in different protein coding genes during the development of OC are extensively studied [290]. However, the role of ncRNAs is not well-studied in this regard. Few studies have reported the potential modulation of miRNAs, lncRNAs, and circRNAs in premalignant lesions, including oral leukoplakia (LK), oral lichen planus (OLP), oral submucous fibrosis (OSF), and oral dysplasia, with respect to OC (Figure 2).
Up-regulation of miR-7, miR-31, miR-1293, and down-regulation of miR-133a, miR-204 and miR-206 were reported in OC samples. Among these miRNAs, significantly high expressions of miR-31 and down-regulation of its target gene C-X-C motif chemokine ligand 12 (CXCL12) were seen in LK and OLP tissues, suggesting their importance in OC progression from pre-cancerous stages [291]. The miR-21, miR-181b, miR-345, miR-549 and miR-205, were found to be overexpressed both in progressive dysplasia and OC [292]. Another study reported up-regulation of miR-145, lncRNA RoR, and SNHG1 and downregulation of miR-34a from low-grade to high-grade dysplasia and, finally, to OC during carcinogenesis [293]. The lncRNA FGD5-AS1 inhibits NF-kB signaling, and gets downregulated in chronic periodontal samples as compared to the healthy tissues [294]. On the other hand, lncRNA MALAT1 was found to be up-regulated in periodontal samples, and induced inflammation through TLR4 by targeting miR-20a [294]. In OSF, up-regulation of lncRNA LINC00974, HIF1A-AS1, and down-regulation of GAS5-AS1 were seen during the development of OSF [294]. An in silico study examined microarray data of 167 OC, 17 dysplasia, and 45 normal oral tissues from the GEO database for expression analysis of lncRNAs [295]. Among these groups, 200 lncRNAs were found to be common in three groups, and 1206 genes are common in OC vs. dysplasia groups. The differentially expressed genes (DEGs) identified among the three groups were found to regulate OC development through PI3K-Akt signaling and NF-kB signaling. Among the DEGs, lncRNA DUXAP10 is relatively new, and associated with the progression of OC development [295]. A high throughput sequencing study identified 366 significantly modulated circRNAs, including 65 up-regulated and 301 down-regulated in LK tissues as compared to the normal mucosa indicating their importance in OC development [296]. Some of the top significantly up-regulated circRNAs were Circ_HLA-C, Circ_PLIN4, Circ_MTX2, Circ_RNF13, and the down-regulated ones were Circ_SENP2, Circ_PLEKHM2, Circ_ERICH1, Circ_EMB, Circ_ALDH3A2, and Circ_ZNF720. The Circ_HLA-C showed stage-wise up-regulation from mild to severe dysplasia [296]. All these studies indicate the importance of ncRNAs in OC development from precancerous lesions to the most aggressive form; however, more mechanistic investigation is needed in this regard.

NcRNAs in Body Fluid and Exosomes of Oral Cancer as Diagnostic Markers
Even though there has been a lot of improvement in the treatment of OC in the last few years, the prognosis for OC remains poor. Involvement of extracellular vesicles or exosomes in biofluids, like blood and saliva, is seen in disease progression, cellular communication, and metastasis of many cancer types, including OC [297][298][299]. Moreover, studies have revealed that biomarkers in the blood are more intriguing, due to their low invasiveness and increased stability. In a recent study, aberrant expressions of 18 different circulating miRNAs have been discovered that possess a direct association with a poor prognosis for head and neck cancer [300]. Salivary exosomal miRNA-1307-5p was seen as a potent prognostic indicator for oral malignancies, as it has shown the ability to indicate poor prognosis as well as poor patient outcomes [301]. Increased levels of lncRNA TIRY derived from exosomes have been found to reduce miR-14 expression levels which, in turn, enhances OC progression and metastasis [302]. In another study, Li et al., have mentioned two lncRNAs, namely MAGI2-AS3 and CCDC144NL-AS1, derived from serum exosomes that encourage cellular proliferation, migration, and invasion in OC, via regulating the PI3K-AKT-mTOR pathway [303]. Several circRNAs have also been found to have their role as potent biomarkers in OC. A high level of circ_0000199 was seen in circulating exosomes of OC patients, which was found to be associated with poor survival outcomes, proving itself as a potent biomarker for OC [247]. The circ_0001874 and circ_0001971 derived from the saliva of patients were up-regulated in OC and identified as a potential diagnostic biomarker [250]. Table 5 enlists the expression of some of the

NcRNAs in Body Fluid and Exosomes of Oral Cancer as Diagnostic Markers
Even though there has been a lot of improvement in the treatment of OC in the last few years, the prognosis for OC remains poor. Involvement of extracellular vesicles or exosomes in biofluids, like blood and saliva, is seen in disease progression, cellular communication, and metastasis of many cancer types, including OC [297][298][299]. Moreover, studies have revealed that biomarkers in the blood are more intriguing, due to their low invasiveness and increased stability. In a recent study, aberrant expressions of 18 different circulating miRNAs have been discovered that possess a direct association with a poor prognosis for head and neck cancer [300]. Salivary exosomal miRNA-1307-5p was seen as a potent prognostic indicator for oral malignancies, as it has shown the ability to indicate poor prognosis as well as poor patient outcomes [301]. Increased levels of lncRNA TIRY derived from exosomes have been found to reduce miR-14 expression levels which, in turn, enhances OC progression and metastasis [302]. In another study, Li et al., have mentioned two lncRNAs, namely MAGI2-AS3 and CCDC144NL-AS1, derived from serum exosomes that encourage cellular proliferation, migration, and invasion in OC, via regulating the PI3K-AKT-mTOR pathway [303]. Several circRNAs have also been found to have their role as potent biomarkers in OC. A high level of circ_0000199 was seen in circulating exosomes of OC patients, which was found to be associated with poor survival outcomes, proving itself as a potent biomarker for OC [247]. The circ_0001874 and circ_0001971 derived from the saliva of patients were up-regulated in OC and identified as a potential diagnostic biomarker [250]. Table 5 enlists the expression of some of the ncRNAs, which are derived from exosomes, blood, serum/ plasma, and saliva samples, and have significance as biomarkers.

Non-Coding RNA in Oral Cancer Clinical Trials
The utilization of cancer-specific ncRNAs as biomarkers and therapeutic targets could aid in tailoring treatment options to specific patients or patient subgroups. In contrast to the typical tumor markers being used in latest medical applications, such as protein biomarkers and metabolic products, growing research indicates that ncRNAs could be ideal agents in cancer diagnosis and therapy. This is because the ncRNAs target multiple druggable and non-druggable targets and signaling events at a time. Further, they have tissue specificity, distinct RNA attributes of rapid detection, extra tissue-associated activity, and a far more stable structure [325]. Several clinical trials have been conducted using ncRNAs, especially using miRNAs as diagnostic and therapeutic biomarkers [326]. Through head-on targeting of gene sequences using antisense oligonucleotides (ASOs) and siRNA-associated therapeutic applications, the most sophisticated and concise therapeutic efforts at RNA screening have been achieved to date [326]. The lncRNA H19 promoter sequence has been introduced with the coding sequence of diphtheria toxin in BC-819 plasmid in clinical trials of the bladder, pancreatic and ovarian cancer [115,327]. The lncRNA HOTAIR and CCAT1 (ClinicalTrials.gov) are in clinical trials for thyroid and colorectal cancer diagnostic biomarker studies, whereas inhibitors of LINC01212 and lncMyoD are used as therapeutic markers for melanoma (US2016271163) and sarcoma therapy (WO2015020960) [115,327]. All these studies indicate possible clinical implications of ncRNAs in cancer diagnosis and therapy. A clinical study has been initiated to evaluate the sensitivity and specificity of miRNA-412 and miR-512 in extracellular vesicles from saliva in the malignant progression of OC (ClinicalTrials.gov Identifier: NCT04913545). Another trial has recently been initiated to identify diagnostic and prognostic miRNA biomarkers from blood, saliva, and tissue samples of head and neck cancer (ClinicalTrials.gov Identifier: NCT04305366). The diagnostic and therapeutic importance of lncRNA MALAT1 and its target miR-124 has been studied in saliva samples from OC patients (ClinicalTrials.gov Identifier: NCT05708209). A randomized phase II study has been initiated to identify salivary and plasma miRNAs of head and neck cancer patients, and monitor their change during the dietary intervention (ClinicalTrials.gov Identifier: NCT02869399). Two studies have been performed to assess the correlation of head-neck cancer immunotherapy with blood and plasma miRNAs profiles (ClinicalTrials.gov Identifier: NCT03843515 and NCT04453046). Low EGFR-AS1 lncRNA expression is determined as companion diagnostic biomarker of OC, and a phase II clinical study is recently ongoing to evaluate therapeutic efficacy of a certain drug of EGFR-associated advanced OC with low-EGFR-AS1 OC patients (ClinicalTrials.gov Identifier: NCT04946968). Outcomes of the study may indicate the diagnostic and therapeutic importance of EGFR-AS1 in OC treatment.

Future Perspectives
Specific ncRNA expression changes have been linked to disease progression and poor outcomes in OC patients. Moreover, the ncRNAs show a promising role in regulating multiple signaling pathways in the modulation of OC progression, invasion, and metastasis. Nevertheless, such conclusions still have to be transformed into medical settings for OC patient populations. This could be attributable to a lack of scientific and technological evaluation research in pre-clinical systems, a dearth of prospective cohort clinical studies, inadequate standardization of the work process to extract and enhance ncRNA, and an absence of standardization targets. Since OC is linked to both behavioral risk factors and genetic predisposition, investigations must account for population-level variability; thus, substantial research regarding patient cohort study from various global locations are required to confirm these benchmarks. In short, the field of ncRNA research in OC is rapidly advancing, and holds great promise for improving our understanding of the disease and developing new strategies for diagnosis, treatment, and prevention. However, there are still many unknowns, and much more research is needed to fully understand the complex roles that ncRNAs play in OC. There are many ncRNAs that are yet to be explored. Many modified ncRNAs are designed to improve their stability and target specific efficacy. Thus, studies in pre-clinical and clinical systems, modifications to increase in-vivo stability and organ-specific targeting, extensive validation, and follow-up studies are needed to fully realize the potential of ncRNAs in OC. A rapid detection kit to identify ncRNAs in blood or saliva samples would greatly improve the diagnostic accuracy of OC. Thus, the acquired information in these areas will pave the way for more important healthcare, prognostic, and treatment options for the management of OC in the near future.

Conclusions
In summary, the review describes the substantial role of both quantitative and qualitative alterations of different ncRNAs (such as miRNA, lncRNA, circRNA, snoRNA, and piRNA) in the development of OC. The ncRNAs target multiple signaling molecules at a time, and regulate cell proliferation, survival, angiogenesis, metastasis and drug resistance. Studies have revealed that the high death rate and morbidity in OC are correlated with the complexity of conducting a quick diagnosis and appropriate management. Therefore, a timely diagnosis can prove to be crucial for controlling potential invasion and metastasis of oral premalignant conditions, and can also increase the overall life expectancies of patients. Keeping this in mind, a lot of emphasis is being given to the regulative function of different ncRNA profiles, as they demonstrate great promise in identifying OC lesions. Large numbers of ncRNAs like miRNAs, lncRNAs, and circRNAs obtained from exosomes or blood, serum, and saliva unveil their implicit role as non-invasive diagnostic and prognostic biomarkers for OC. Some clinical studies have been initiated to identify blood or saliva miRNA biomarkers in OC patients. miR-412, miR-512, miR-124, lncRNAs MALAT-1, and EGFR-AS1 are currently in OC clinical trials as diagnostic and therapeutic biomarkers, and the outcomes are yet to be received. Recent research on the role of other ncRNA molecules, like snoRNAs and piRNAs, behind the cause of OC development also makes them potential contenders for early diagnosis tools. However, in order to properly integrate liquid biopsy tests into the clinical practice in OC diagnosis, more in-depth pre-clinical research, as well as experimental studies with large cohorts, would indeed be required to confirm these findings and the effectiveness of these biomarkers in OC.

Acknowledgments:
The authors would like to thank Ratna B. Ray, Saint Louis University, USA for the critical review of our manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.