Introduction

Over the entire Head and Neck cancer incidence, more than 95% comprises oral squamous cell carcinoma (OSCC) and is placed among the top ten cancer-related mortality worldwide [1]. In India, OSCC and oral cancer remain the most frequently seen, representing nearly one-third of the total incidences worldwide [2]. The common risk factors of OSCC include betel quid chewing with areca nut, excess tobacco usage or smoking, alcohol consumption, poor oral hygiene, and dietary deficiencies. Other factors include environmental factors, sustained viral infections, i.e., human papillomavirus (HPV), and microbial infections. Besides, genetics and epigenetic alteration contribute to the formation of oral cancer. Despite advancements in the knowledge gap and scientific findings over the previous 20 years, an oral cancer patient's 5-year survival rate remains below 50%. Diagnosis of OSCC early on with pharmacological and surgical interventions results in a better prognosis. However, due to unavailable appropriate biomarkers, recognizing at later stages where metastasis would have occurred results in high morbidity and low survival rate [3].

Most biomarkers, including the coding genes, constitute approx. 2% and the rest are non-coding genes that are often ignored. Nearly 98% of the human genome contains non-coding RNA (ncRNA) that does not possess a protein-encoding ability but functions as a master regulators of gene expression by serving as protein scaffolds, transcription factor regulators, post-transcriptional and epigenetic regulators. Alteration in gene expression is identified as a new hallmark in the development of cancer. Among ncRNAs, the long noncoding RNA (lncRNA) started to gain attention due to its underlying mechanism either as a tumor suppressor or an oncogene. Based on its over or under expression lncRNAs are regarded as a biomarkers in oral pre-malignant lesions, associated with tumor initiation, progression, and metastasis, as well as in drug resistance. LncRNAs are analyzed within the saliva and site-specific expression profile helping in the early diagnosis of OSCC and monitoring recurrence post-surgery [4].

In the same way, circular RNA (CircRNA) has also been investigated, perhaps providing a biomarker because of its similar property of miRNA sponging and regulating gene expression. Numerous MicroRNAs (miRNAs) are investigated for their diverse role in hallmarks of cancer and are putative markers for cancer diagnosis and prognosis. It is to be noted that comparatively fewer studies have been conducted on lncRNA and circRNA than miRNA exploiting its clinical utility as a non-invasive diagnostic and prognostic marker in saliva and blood samples.

Initially, the development of pre-malignant lesions is manifested as erythroplakia, leukoplakia, lichen planus, or the mixed forms that are triggered down to oral malignant forms. The incessant progression of the cell cycle is pivotal for tumor growth and thus frequent mutation in the cell cycle checkpoints is commonly seen in OSCC. The mutations in the p53 gene are found in more than 50% of the oral cancer tissues followed by CDKN2A mutations [5]. Surgical biopsy is established to be a gold standard in the diagnosis of OSCC, although there are certain setbacks as it is an invasive technique that targets a single tumor at a specific point in time and does not reveal spatiotemporal heterogeneity of solid tumors [6]. Hence discovering a better strategy to precisely characterize the disease to assist in diagnosis, prognosis, and treatment is of prime necessity.

Methodology

The current review provides comprehensive knowledge on the non-coding RNAs (lncRNA, circRNA, and miRNA) as prospective markers for oral cancer diagnosis and prognosis, considering their viability in saliva, serum, and plasma. The literature search was performed using major search engines like PubMed and Google Scholar between the timeline of 2012 to 2023. Around 2300 articles were obtained from the first keyword combination (“non-coding RNA” AND “oral cancer”), 312 from (“non-coding RNA” AND “oral cancer” AND “cell cycle”), 78 from (“serum non-coding RNA” AND “oral cancer”), 81 from (“plasma non-coding RNA” AND “oral cancer”), 143 from (“salivary non-coding RNA” AND “oral cancer”). The compiled data focuses mainly on the diagnostic and prognostic significance of miRNAs, circRNAs, and lncRNAs in oral cancer and all other cancers. The subject-relevant articles published in a language other than English are excluded from the review. Moreover, DNA, metabolites, protein, and peptide markers are not included in the review.

Overview of cell cycle: CKIs, dysregulated cyclin and Cdks

The cell cycle consists of gap phases G1 and G2, S, and M phases regulated by serine-threonine kinases called Cdks with their regulatory subunit cyclins. The human genome encodes around 20 Cdks and 29 cyclins. Cdk 1,2,4,6 is directly involved in the cell cycle, whereas Cdk 7,8,9 has a major role in the regulation of gene expression. Besides participating in DNA damage and response, stemness, metabolism, and angiogenesis. Upon external stimulus, the Cdk4, 6/ Cyclin D complex initiates phosphorylation of the Rb pocket in the early G1 phase releasing transcription factor E2F which further transcribes genes involved in the cell cycle progression. In the late G1 phase, the Cyclin E/Cdk2 complex completes the phosphorylation of E2F and initiates the S phase. Cyclin A/Cdk2 and cyclin B/Cdk1 fulfil DNA synthesis and mitotic phases. The G0 phase also called the quiescence stage or dormancy period, is mediated by the cyclin C/Cdk3 complex.

Endogenous CKIs (Cyclin-dependent kinase inhibitors) can restrain the activity of specific cyclin-Cdk complexes thus regulating the cell cycle. The different families of CKI include INK4, Cip/Kip, and ribosomal protein-inhibiting Cdks (RPICs). Compounds resembling the action of CKIs are clinically developed to destroy the hyperproliferative cells. These include first-generation pan CKIs Flavopiridol and roscovitine though lacking specificity further led to the invention of selective second-generation CKIs mainly pan-Cdk inhibitor AT7519 and dinaciclib which even failed in the clinical trials. Consequently, it paved the way for third-generation Cdk4/6 selective inhibitors such as Palbociclib, Ribociclib, and Abemaciclib to get clinically approved for breast cancer [7]. A study on Histone deacetylase inhibitors (HDACis) in combination with Cdk4/6 inhibitor Abemaciclib showed better sensitivity and enhanced antitumor activity against OSCC [8].

The dysregulation of cyclins, Cdks, and CKIs are commonly implicated in both pre-cancerous (dysplastic) lesions and oral cancer. Mutations in retinoblastoma (Rb) protein alongside the loss of p16 activity on cyclin D-Cdk4/6 are commonly manifested in oral carcinogenesis. The downregulation of p16, a tumor suppressor and cell cycle controller, by targeting cyclin D1 is associated with the transformation of the dysplastic lesion and poor survival in OSCC patients. Inactivation of tumor suppressor genes via hypermethylation of promoter regions of p14 and p16 is commonly manifested in pre-cancerous lesions and OSCC [9]. A study conducted had [10] identified that overexpression of p53 and EGFR of cytoplasmic and nuclear sublocation decreased the overall survival rate in OSCC patients. The cyclin-dependent kinase inhibitor p27KIP1 internally controls the cell cycle at G0 to S phase by regulating cyclin D/Cdk4 and cyclin D/Cdk6 complexes. p27 is also identified as an anti-oncogene and functions mainly in regulating cell proliferation, and apoptosis and is found to be a therapeutic target and prognostic marker of OSCC. The level of p27 is reported to be reduced in correspondence to the increased expression of cyclin D1 and corresponding Cdks in the early stage of OSCC development [11]. Silencing of Cdk5 slowed down G1 to S phase and vice versa is directly correlated with poor survival and tumor development in tongue squamous cell carcinoma (TSCC) [12].

At the same time, overexpression of Cdk1 and cyclin B1 are commonly observed in patients with recurrent OSCC or lymph node metastasized OSCC. The cumulative 5-year survival period of Cdk1-positive patients remarkably declined than the negative ones validating its role to be manipulated as a prognostic marker for OSCC survival [13]. The overexpression of cyclin E and B1 is associated with advanced tumor stage and severity level of OSCC and larynx cancer. Cyclin B1 is a reliable indicator of lymph node metastasized tongue carcinoma. Among the D cyclins, Cyclin D1(a proto-oncogene) is commonly overexpressed in OSCC by inappropriate mitogen expression or receptor activation and the commencement of the Rb pathway evading the inhibitory signals. The risk factors of OSCC leave the chromosomal region 11q13 vulnerable to aberrant amplification of genes associated with it predominantly CCND1 coding for cyclin D1 [14]. Owing to the mutation in cyclin D1, it cannot be cleared via the 26S proteasomal ubiquitin mechanism in the cytoplasm leading to its nuclear accumulation. Further altering the Cdks, inhibitory proteins, cell cycle checkpoints, obstinate towards growth factors, and eventually progressing into a pile of abnormal cells [15]. The upregulation of cyclin D1 Cdk4/6 results in the increased severity of dysplasia to cancerous tissue is attributed to its early and late-stage characterization in carcinogenesis and in concordance with the downregulation of tumor suppressor genes that code for the INK family of inhibitors [16].

Non-coding RNAs

All the cyclins, Cdks, and CKIs are regulated by ncRNAs either at the transcriptional or post-transcriptional level. ncRNAs are 98% of human genes that do not encode any proteins. The ncRNAs are mainly categorized according to their size into small and long non-coding RNAs consisting of miRNAs, and circRNAs proven to have a constituent role in oral cancer. Short interfering RNAs (siRNAs), PIWI-interacting RNAs (piRNAs), small nucleolar RNAs (snoRNA), small nuclear RNA (snRNA), repeat-associated RNAs (rasiRNAs), and additional ncRNAs yet to be decoded and characterized [17].

Cell cycle regulatory miRNAs in OSCC

miRNAs and siRNA are short single-stranded RNAs of around 20–22 nucleotides that play a huge role in gene expression and regulation. miRNAs are initially transcribed into primary miRNA (pri miRNA), a stem-loop structure by RNA pol II. If there is perfect complementarity between the sequences of miRNA and mRNA it leads to endonucleolytic cleavage of the target mRNA whereas its partial complementarity leads to translational silencing of the mRNA. miRNAs are involved in both tumor development and suppression. They are involved in regulating a large number of coding and non-coding genes encompassing some of the crucial oncogenes RAS, MYC, and EGFR along with the tumor suppressors TP53, PTEN, and BRCA1. miR21 is identified as an oncomiR since it is overexpressed in progressive leukoplakia and OSCC. miR-184 is overexpressed in TSCC and functions as an anti-apoptotic factor altering the pro-apoptotic factor c-myc and promoting cell proliferation whereas its downregulation has also been reported. MiR24 is studied for its role in regulating CDKN1B gene encoding p27kip1, inevitably cell proliferation and evasion of apoptosis. In addition, miR-196a and miR-10b are also found to be overexpressed in OSCC. Downregulation of miR-125b and miR-100 [18], miR-596, miR-494 [19], miR-145 (targets c-Myc and Cdk6) [20], and miR-195 are involved in the regulation of cell cycle proteins cyclin D1 and Bcl2 [21] in oral cancer indicates its involvement as a tumor suppressor. MiR-137 and miR-193a were demonstrated to be downregulated in OSCC possessing a crucial role in inhibiting the expression of Cdk6 and E2F6 resulting in the arrest of cell cycle progression and apoptosis. miR-503 and miR-15a are involved in the downregulation of cyclin D1 and E and as a consequence, the level of these miRNAs is found to be declining in OC [22]. miR-222-3p a tumor promoter miRNA showed a higher expression in OSCC tissues and is associated with clinical staging and lymph node metastasis while the 5-year survival rate declined in comparison to the low expression group. miR-222-3p modulates its action through downregulation of CDKN1B/p27 [11].

Mir5100 promotes proliferation in OSCC, attenuating the G1-S phase and decreasing the action of SCAI, a suppressor of cancer cell invasion. Consequently, the mir5100 inhibitor inhibited the G1-S transition by lowering the expression of proliferative genes Cdk2 and Cyclin D and upregulating p27 expression. Furthermore silencing of miR5100 suppressed the migration and invasion of OSCC via tuning of SCAI [23]. Upregulation of miR-504 decreased the expression of Cdk6, cyclin D1, and E2F1 implying its function as a tumor suppressor in OSCC [24]. Downregulation of miR-191 silenced the expression of PCNA, CDK4, and other proteins in the downstream Wnt/β-catenin signaling pathway exerting an anti-tumor effect on OSCC cells via. Phospholipase C delta1 (PLCD1) [25]. Oral pre-malignant disorders (OPMDs) are early epithelial lesions seen in the oral cavity possessing a greater chance of transforming into a malignancy. They manifest features such as oral leukoplakia (OLK), oral lichen planus (OLP), and oral submucous fibrosis (OSF). Among a variety of miRNAs altered in OLK, miR-10b-3p is one found in the saliva of mild dysplastic OLK. Eight miRNAs were found to be intensified in progressing OLK notably miR-345/21/181b subset, and therefore miRNAs demonstrate a strong attribute in differentiating between progressing and nonprogressive mild OLK. Chronic inflammation and degeneration of keratinocytes are the underlying pathological causes recognized in OLP. miR-27b and miR-125b are found to be downregulated in OLP whose function is to inhibit keratinocyte proliferation and promote apoptosis. Thus aiming miRNA can be an approach to hinder the transformation of OLP to OSCC [26]. Mir-9 was found to inhibit cell proliferation of OSCC through the downregulation of Cdk6 and Cyclin D1 and stopping the cell from G1/S transition thus functioning as a tumor suppressor [27].

Cell cycle regulatory circular RNAs in OSCC

The other major class of ncRNA is the ones bearing more than 200 nucleotides called lncRNAs processed by RNA pol II which includes long intergenic ncRNAs (lincRNAs), antisense RNAs (asRNAs), pseudogenes, and circRNAs. CircRNAs are ncRNAs with covalently linked 5′3’ends without a 5’polyadenylated tail forming a circular loop mainly because of back splicing of pre mRNA and realignment of exons. circRNA contains more than 200 nucleotides and is resistant to endonucleolytic cleavage. They are also called miRNA sponges since they bind to miRNA and stop its action of silencing mRNA besides the regulatory function. Because of these reasons, circRNAs are used as a therapeutic strategy to alleviate the overexpressed miRNA in cancer. In addition, circRNAs are involved in epithelial-mesenchymal transition (EMT) and, therefore involved in tumor metastasis and invasion [17, 28]. Knockdown of circYap, a circular RNA obtained from the YAP (Yes-associated protein) gene, upregulated PCNA and C-MYC as well as Rb and Akt phosphorylation. Further overexpression of Cyclin D1/Cdk4 and transformation of the G1-S phase specifying circYAP could impede OSCC progression [29].

circ_SEPT9 enhanced OSCC progression via miR‐1225 and PKN2 [30]. Upregulation of circATRNL1 enhanced the sensitivity of OSCC towards radiotherapy through endogenous sponging of miR-23a-3p, alleviating the inhibitory action on PTEN and promoting apoptosis and cell cycle arrest [31]. Upregulation of circ_0000745 in OSCC increased the expression of CCND1(cyclin D1) via the mediatory interaction with miR-488 and RNA-binding protein (HuR) [32]. Expression of circCLK3 is linked with TSCC progression by negatively regulating its target miR-455-5p while positively regulating PARVA a cancer progressive gene associated with CAM, integrins, and cytoskeleton involved in metastasis and angiogenesis. In addition, the knockdown of circCLK3 preceded G0/G1 phase arrest by decreasing the expression of cyclin D1 and cyclin E1 and modulating [33]. Downregulation of hsa_circRNA_102459 and knockdown of upregulated hsa_circRNA_043621 induced apoptosis and arrest of the G0/G1 phase in OSCC cells [34]. The recently identified circRNAs as biomarkers in biological fluids, saliva, and circulation are mentioned in Tables 1 and 2.

Table 1 Salivary miRNA as biomarker for oral cancer
Full size table
Table 2 Circulatory miRNA as biomarker for oral cancer
Full size table

Cell cycle regulatory long non-coding RNAs in OSCC

Lack of consistent results in identifying miRNAs as a promising tumor biomarker shifted the paradigm of research towards lncRNAs which in the past was considered junk gene. LncRNAs are involved in the regulation of gene expression at the post-transcriptional level via binding to the chromatin-modifying complexes and transcription factors. Besides participating in epigenetic regulation, lncRNAs play a direct role in cell cycle regulation and differentiation. Also involved in inhibiting transcription by targeting RNA pol II and preventing translational silencing by masking the miRNA binding site i.e., miRNA sponging. Based on an initial study conducted on lncRNA expression profiling of oral mucosa, nearly 325 lncRNA was expressed, and more than half of it is altered in dysplasia of the oral cavity. Prominent lncRNAs such as HotairR (Hox Transcript Antisense RNA), Neat1(Nuclear Enriched Abundant Transcript (1), Uca1 (urothelial cancer associated (1), LncHIFCAR are found to be elevated. While Meg-3(Maternally expressed gene (3) possesses a suppressive action on Wnt/β-catenin signaling in tumors and functions as a regulatory element of enzyme DNA methyltransferase 3B is reported to be downregulated in saliva and tissue samples from OSCC and TSCC patients orchestrating poor patient survival. Hotair and Taurine upregulated gene 1 (Tug1) is overexpressed in lymph node metastasized OSCC. High expression of Uca1 is associated with migratory ability and regional lymph node metastasis in tongue squamous cell carcinoma. ZEB1-AS1, HOXA11-AS, LINC01116, and LEF1-AS1 were also found to have a role in the development of OSSC and showed a poor prognosis [35, 36]. HOTAIR performs its role by employing polycomb repressive complex 2 (PRC2) subunit, a chromatin remodeling complex whose function is in gene silencing or transcriptional repression and thus binding to the promoter region of tumor suppressor gene PTEN inactivates its expression and condensation of chromatin resulting in tumor progression [37]. HOTTIP (HOXA distal transcript antisense RNA)is involved in the carcinogenesis and migration of OSCC cells and acts as a competing endogenous (ceRNA) of miR-206 [38].

One among the frequently overexpressed lncRNAs in OSCC includes DLEU1 (deleted in lymphocytic leukemia 1) is associated with a low survival rate in HNSCC patients and is linked with the expression of multiple oncogenes demonstrating it as a possible target for therapeutics and prognosis [39]. HOXA-AS3 negatively regulates miR-218-5p a tumor suppressor that inhibits COL1A1 and LPCAT1 in the proliferation and migration of OSCC [40]. A study on integrative profiling of lncRNA identified the role of DUXAP8 in oral carcinoma (OC). DUXAP8 localized in the nucleus of OC cells binds to histone methyltransferase enhancer of zeste homolog 2 (EZH2) a commonly dysregulated methyltransferase enzyme. It represses the transcription of tumor suppressor gene KLF2 resulting in tumor formation although its complete underlying mechanism is unclear [41]. SOX21AS1 suppresses the growth and invasion of cancer cells either by acting as a decoy preventing the binding of the transcription factor to the DNA and transcription of oncogenes or by enhancing other tumor suppressors. Thus downregulation of lncRNA SOX21AS1 is associated with a poor prognosis of OSCC [42]. Overexpressed GASL1 induces the arrest of the G0/G1 phase by inhibiting cyclin D/CDK4 and apoptosis of OSCC cells [43]. LUCAT-1 (lung cancer-related transcript 1) is upregulated in OSCC and positively correlated with the expression of PCNA (proliferating cell nuclear antigen) [44]. lncRNA GACAT1(Gastric cancer-associated transcript 1) is involved in OSCC proliferation and migration, negatively targeting miR-149, as a result, the knockdown of GACAT1 promoted autophagy and apoptosis of the cell [45] Fig. 1.

Fig. 1
figure 1

The illustration represents the detection of ncRNA biomarkers in various biological samples such as blood (plasma or serum) (B) tissue (c) saliva from OSCCC patients

Full size image

In normal cells, MALAT1 (metastasis-associated lung adenocarcinoma transcript (1) participates in alternative splicing of pre mRNA by functioning along with various pre-mRNA splicing complexes such as the Ser/Arg (SR) splicing factor. It is a transcript of Neat 2 and is accepted to be involved in the metastasis of OSCC. The knockdown of MALAT1 downregulated N-cadherin and Vimentin while upregulation of E-cadherin. Moreover, the nuclear and cytoplasmic levels of β-catenin and NF-κB are suppressed elucidating the role of MALAT1 lncRNA in EMT and metastasis of OSCC through modulation of β-catenin and NF-κB pathway [46]. Enhanced expression of lncRNA TIRY from Cancer-associated fibroblasts (CAF)-derived exosomes notably reduced miR-14 expression and downstream activation of Wnt/β-catenin and EMT expression displayed in OSCC tissues [40]. ZEB1-AS1 and RBM5-AS1 are highly expressed in tumor tissues and sponging of its corresponding miRNA, miR-23a, and miR-1285-3p respectively contributed to the invasion, migration, and proliferation of OSCC [47, 48]. BRAF-activated long non-coding RNA (BANCR) overexpression is seen in OSCC correlated with its carcinogenic property [48]. LOLA1 (lncRNA oral leukoplakia progressed associated (1) is found to help predict the transformation stage from OL to high-risk dysplasia to early-stage OSCC via AKT/GSK‐3β pathway. It is involved in EMT transition [49]. Expression of RPSAP52 was identified to be negatively involved in the expression of miR‐423‐5p and promoted tumor growth,TNM staging of TSCC, and release of cytokines such as IL-1β,6, 8, 10 and TGF-β via MYBL2 [50]. LncRNA ZFAS1 promoted OSCC development through the sponging of miRNA-6499-3p and activation of CCL5 [51]. LncRNA PHACTR2-AS1 through ceRNA of miR-137 activates EMT transcription factor Snail as a result of TSCC migration [52]. The schematic illustration of various cell cycle regulatory ncRNAs is put together in Fig. 2.

Fig. 2
figure 2

Schematic representation of cell cycle regulatory genes regulated by ncRNAs

Full size image

LncRNAs, miRNAs, and CircRNAs displaying resistance to chemo regimen

Enhanced expression of CASC2 increased the chemosensitivity of OSCC cells resistant to cisplatin through the expression KANK1 and sponging of miR-31-5p [53]. Transcription factor FOXD1 transcribes lncRNA CYTOR sponges miR-1252–5p and miR-3148 and upregulation of lipoma-preferred partner (LPP) expression is positively linked with cisplatin chemoresistance [54]. CEBPA-DT overexpression augmented the cisplatin-induced chemoresistance in OSCC cells via CEBPA/BCL2 axis [55] Overexpression of chemotherapy-induced lncRNA 1(CILA1) via Wnt/β-catenin signaling involved in chemoresistance of TSCC [56]. One more factor implicated in resistance against cisplatin-induced death of OSCC is Medkine, a heparin-binding growth factor released by CAFs that elevates the expression of lncRNA ANRIL. Restriction of ANRIL in tumor cells had an impact on increasing the sensitivity towards cisplatin by blocking the action of drug transporters MRP1 and ABCC2 and therefore enhancing the cisplatin cytotoxicity, apoptosis, and inhibiting proliferation of tumor cells [57] Uca1 demonstrated enhanced chemoresistance in cisplatin-induced OSCC through the sponging of miR-184 and inhibiting its role in tumor suppression [58]. Overexpression of GAS5 in conjunction with downregulated GAS5 boosted cisplatin chemosensitivity in oral cancer cell lines and OSCC tissues [59]. KCNQ1OT1 sponges miR-211-5p and upregulation of Ezrin/Fak/Src signaling leading to cisplatin resistance in TSCC [60].

Inhibitory miRNA-485-5p attenuated keratin 17 gene and its signaling pathway involving integrin/FAK/Src/ERK/β-catenin through which OSCC cells attain the property of stemness, EMT, cisplatin resistance, and poor outcome of survival [61]. Under expression of miR-15b, upregulation of TRIM14, and further activation of ERCC1 and YAP led to chemoresistance on TSCC [62]. Intratumoral transfer of miR-155 inhibitor in cisplatin-resistant OSCC reduced the stemness, and EMT expression, declined the expression of efflux pump protein, and upregulated the expression of FOXO3a, thereby reducing the tumor [63]. In vitro and in vivo transfection of exosomal mir200c significantly sensitized the Docetaxel-resistant TSCC cells by targeting TUBB3 and PPP2R1B [64]. ZEB1 attaches to the promoter region of carbonic anhydrase IX (CA9) positively regulating its activity of chemoresistance, preventing the decrease in the pH and caspase 3-mediated apoptosis induced by chemotherapy [65]. miR-1224-5p bound with NSD2 partially neutralized the action of lncRNA APCDD1L-AS1 in 5-fluorouracil induced OSCC resistance [66]. Loss of expression of miR-132 upregulated TGF-β1/Smad2/3 signaling as a result of proliferation, invasiveness, and cisplatin resistance in OSCC [67]. The expression of circANKS1B and TGF-β are closely associated, and thus, targeting TGF-β through the modulation of circANKS1B is presumed to prevent cisplatin resistance in OSCC [68].

Stem cell biomarkers of cell cycle

Cancer stem cells (CSCs) are initiated from a cancer cell with the property of self-renewal and differentiation showing metastatic potential and resistance towards chemo and radiotherapy as well as tumor recurrence. Hence targeting specific biomarkers expressed on the CSCs can aid towards therapeutic innovation. The most frequently expressed CSC biomarkers in HNSCC are CD44, CD24, CD98, c-MET, CD133, CD166, Notch 1, ALDH1, SOX2, Nanog, OCT4. CD147 inhibits p53 signaling pathway and its higher expression is associated with a poor prognosis of OSCC. The expression of CD147 was in concurrent with the severity and progression of dysplasia from OLK to OSSC [69]. The same was observed on the expression of aldehyde dehydrogenase (ALDH1/2) differentiating dysplastic and non-dysplastic OSCC. Anterior gradient protein 2 (AGR2), is involved in the suppression of p53, regulating cyclin D1, and EMT Transition that is directly correlated with the stemness of OSCC [70].

The impact of lncRNAs on cell cycle

Around 42% of the genes dysregulated in the cell cycle belonged to ncRNAs, specifically lncRNAs. Most of it was overexpressed in the G1 phase demonstrating its role during G1 to S transition. LncRNAs are expressed upon DNA damage and regulate the transcription of the CCND1 gene. The tumor-suppressive lncRNA growth arrest specific 5 (GAS5) inhibits Cdk6 and upregulates p27kip1 restraining the proliferation of tumor cells. Besides the lncRNAs, ANRIL, p15AS, or CDKN2B-AS1 are involved in the development of tumors by regulating CKI via epigenetic repression of the INK4/ARF locus. MALAT 1, LAST (LncRNA-Assisted Stabilization of Transcripts), and MIR100HG are overexpressed in the G1/S transition. The cancer stem cells attain resistance to antiproliferative drugs by undergoing quiescence or entry into the G0 phase and the property of stemness is contributed by various lncRNAs such as GAS5 and DANCR however, the complete mechanism needs to be investigated further. lncRNA viz SUNO1 plays a role in S phase progression by its interaction between the helicases of DNA replication in tumor cells. The lack of lncRNA-RI, APAL, and OIP5-AS1 produced mitotic defects linking lncRNA in regulating mitotic kinases [22]. Downregulation of MALAT1 appeared to inhibit cell cycle proliferation arresting the G2/M phase and ceasing tumor growth of different cancer types in both in vitro cell lines and in vivo tumor xenograft nude mice models [71]. Knockdown of SLC16A1-AS resulted in decreased expression of Cyclin D1 and ceased OSCC cell proliferation at the G0/G1 phase[72]. CCAT1 aberrant expression downregulated Cdk4 and cyclin D1 while suppressing miR-181, a tumor suppressor, and activating Wnt signaling [73]. Overexpression of oncogenic lncRNA SLC16A1-AS prevented G0/G1 arrest, decreased cyclin D1, and proliferation of OSCC cells [72]. Identifying the function of the fifth genetic element within the INK4 locus ANROC (associated negative regulation of cell proliferation) is presumed to raise the expression of p16INK4A, p15INK4B, and ARF whereas inhibited cyclin B1 and arrest of G2 to M phase. Thus, denoting the action of lncRNA ANROC in ceasing the cell cycle by modulating cell cycle-related proteins [74].

Salivary miRNA, a biomarker for oral cancer

The saliva forms the environment in which the oral tissues are constantly immersed in and hence identifying a biomarker within the salivary secretions could be used as a non-invasive, easily accessible, storable, and safe to handle diagnostic and prognostic tool in oral cancer patients. Although saliva consists of various enzymes such as ribonucleases that degrade RNA, some protective measures are utilized by miRNA to remain stable within the harsh conditions of biological fluids with the aid of exosomes. The ncRNA miR-31 was found to be more highly upregulated in salivary specimens than that of plasma, implying its significant local contribution via salivary contents, and its secretion even from minute tumors illustrates its sensitivity in detection. This sensitivity remains the same when it comes to the case of advanced tumors as well. Following surgical excision of the tumor, the level of miR31 declined significantly, inferring its contribution to the salivary RNA pool specifically by the tumor tissue within the oral cavity [75]. Salivary levels of miRNA-21 were significantly increased in OSCC compared to the normal and diseased controls while there was no significant difference between the dysplastic PMD suggesting miRNA-21 could be an early molecular marker during the pathogenesis of oral cancer. Moreover, this study also revealed that miRNA-184 can be used to distinguish between OSCC and dysplastic PMD. Besides salivary miRNA -145 results demonstrated its decrease whose restoration in OSCC influenced cell apoptosis by arresting the G1 phase and hindering the activity of Cdk6 and c-Myc [76].

In a study on miRNA expression in saliva conducted within OSCC, OLP, OSCC-remission patients, and the healthy ones, miRNA-27b showed an increased expression in OSCC patients compared to the latter groups presuming it to be a promising marker in the detection of OSCC which has also been reported to be an oncogenic miRNA [77]. Upregulation of salivary miR-21 and miR-31 is observed in OPMD patients when compared to the healthy ones, whereas miR-31 showed a significantly high expression in OPMD transformed to malignant forms and on relapse of post-initial treatment as well. The oncogenic characteristic of miR-31 is attributed to targeting FIH-HIF-VEGF regulatory cascade and blockage of DNA repair mechanism, concomitant cell proliferation, and genomic instability. Moreover, previous studies reported an upregulated miR-31 in tumors of the tongue and OSCC of buccal mucosa [78]. miR-139-5p was found to be downregulated in the salivary samples of TSCC patients in contrast to healthy individuals. Subsequently, the resection of the tumor or post-operative analysis of TSCC saliva samples after 4–6 weeks demonstrated the restoration of salivary miR-139–5-p expression back to normal. Thus, specifying the downregulation of miR-139 was by the tumor and has been suggested to be a diagnostic and prognostic biomarker [79]. The recent OSCC salivary miRNA biomarkers are compiled in Table 1.

Circulatory miRNA, a biomarker for oral cancer

The stability of miRNA within the whole blood, serum, or plasma and their altered expression is utilized in the diagnosis and management of cancer and thus miRNA is demonstrated to be a non-invasive biomarker. A study comparing the miRNA expression in whole blood of OSCC, and healthy patients showed aberrant expression of miR-186, miR-3651, and miR-494. Upregulation of miR-3651, miR-494, and downregulation of miR-186 conclude its prominence in carcinogenesis and progression in OSCC [80]. The serum miR-9 expression is compared in patients with OLK and OSCC with healthy individuals. The results stipulated the significant downregulation of miR-9 expression, consequently reducing Cdk6 and Cyclin D1 compared to the healthy lots specifying it to be a tumor suppressor. It is correlated with advanced stages of OSCC and is a potential prognostic marker [81]. One of the works done on circulatory miRNA involves comparing miRNA expression from pooled plasma of around 5 patients with GSCC (gingival squamous cell carcinoma) versus healthy individuals. The highly expressed miRNA includes miR-26a, miR-126, miR-223, and miR-21, and due to the abundant expression of miR-223 was further investigated that revealed its upregulated expression not only within the plasma but by the noncancerous tissue as well. This implies that the plasma miR-223 is released by the healthy tissue as a part of a biological defense mechanism against tumor growth which correlates with its tumor suppressive action designating its potential as a diagnostic as well as therapeutic marker. The receiver operating characteristic (ROC) analysis curve produced an AUC value of 0.73 and a moderate level of specificity and sensitivity [82]. Plasma miR-196a/b was found to be highly expressed during the initial stages of tumorigenesis and its progression while miR-196a showed better specificity whereas mir-196b showed better sensitivity thus usage of multiple biomarkers complemented the efficacy of these two molecules in detecting oral cancer [83]. It was noted that plasma miR-187 level was 6.4-fold high in pre-operative OSCC patients(n = 63) whose level dropped by threefold following the tumor resection compared to the healthy controls (n = 26). Thus suggesting miR-187 as a marker for survival chances post-surgery [84]. Another study proved the metastatic potential of OSCC contributed by miR-296 in plasma in contrast to miR-130b, which is over-expressed in non-metastatic plasma samples [85]. A study conducted by Xu Huanxi and his colleagues on miRNA expression in the sera of OSCC patients (n = 101) before the surgery without any history of radio or chemotherapy identified significant upregulation of miR-483-5p especially of TNM staging III and IV category exhibiting AUC = 0.85 and sensitivity, specificity (0.853,0.746) correlating with lymph node metastasis [86]. The recent OSCC miRNA biomarkers in circulation are listed in Tables 2 and 3.

Table 3 Circulatory lnRNA as biomarker for oral cancer
Full size table

Salivary lncRNA, a biomarker for oral cancer

Haikuo tang et al. investigated the presence of lncRNA in saliva, a non-invasive sample from lymph node metastasized OSCC patients, and identified HOTAIR to be significantly expressed than other lncRNAs (HULC, MALAT-1, MEG-3, NEAT-1, UCA1) investigated in the study. MALAT-1 was expressed, but no significant difference was seen between metastatic and non-metastatic OSCC. Although NEAT-1 is abundantly present in metastatic tissues, there was no expression of it identified due to its least stability [87]. The recent salivary lncRNA biomarkers of OSCC are mentioned in Table 4.

Table 4 Salivary lnRNA as biomarker for oral cancer
Full size table

Circulatory lncRNA, a biomarker for oral cancer

Merdan Fayda and his colleagues looked for three lncRNAs GAS5, lincRNA-p21, and HOTAIR in plasma collected from Head and neck cancer patients (n = 41) pre and post-chemoradiotherapy (CRT) to determine its diagnostic and prognostic relevance in circulation. The level of GAS5 significantly declined post-treatment compared to pre-group, justifying that the decline in the malignant tissues decreased the release of GAS5 from these tissues while patients with the poorer response towards CRT exhibited raised levels in circulation. Thus, GAS5 outlines the progress of CRT in HNSCC patients [88]. There are only limited investigations done on circulatory and salivary lncRNA and hence further research needs to be done in these fields. The recently investigated lncRNA biomarkers of oral cancer are listed in Table 3 and Fig. 1 represents the ncRNA biomarkers in various biological samples.

Result and conclusion

The identification and quantification of non-coding RNAs in biological samples are widely performed owing to the sophisticated molecular detection methods and thus in the near future, there will be a complete shift towards these non-invasive biomarkers. Firstly, the rationale behind adopting a liquid biopsy over tissue marker is its non-invasive property, repeatable and quick sampling, less complexity, and feasibility. Secondly, they provide a broader picture of the tumor than just the snapshot, early detection, real-time monitoring for targeted therapy and personalized medicine, and detection of drug resistance and prediction of recurrence post-treatment [89]. In the present article, we have clubbed all the miRNA, circRNA, and lncRNAs detected within saliva and blood (Serum and plasma) of oral cancer patients and their influence on the cell cycle from 2012 to 2023. While generous amounts of reads were available on miRNA, a few studies are available about circRNA and lncRNA on oral cancer, specifically on the mentioned body fluids. Considering the diverse patient cohort, molecular techniques performed, sensitivity and specificity in detection, and based on the AUC curve we have curated a few of the ncRNA diagnostic panels. salivary miR-1307-5p, miR-3928, hsa_circ_0001874 and plasma ENST00000412740, NR_131012, ENST00000588803, NR_038323, miR-21, miR200b-3p in circulation which needs to be further deciphered on their mechanism of action, regulatory mechanism, and complex pathway through which their oncogenic property is obtained. Hence future investigations are required to identify the therapeutic targets and clinically approve these biomarkers for oral cancer diagnosis and prognosis.