The research, led by Ravina Vats and Rashmi Bhardwaj at the Centre for Medical Biotechnology, Maharshi Dayanand University in Rohtak, in collaboration with colleagues at Swami Shraddhanand College in Delhi and the Postgraduate Institute of Dental Sciences in Rohtak, focused on a class of tiny regulatory molecules called microRNAs, or miRNAs. These short RNA strands, roughly 18 to 24 nucleotides long, do not code for proteins themselves. Instead, they act as molecular dimmer switches, binding to messenger RNA molecules and fine-tuning how much protein gets produced from each gene. When this regulatory system goes awry, the consequences can be profound: genes that drive cell proliferation can be unleashed, while genes that restrain growth can be silenced. In cancer, particular miRNAs often become abnormally abundant or scarce, and because these changes can be detected in body fluids, they have become one of the most actively pursued categories of liquid biopsy markers.

The study zeroed in on five specific miRNAs with well-documented roles in cancer biology. Three of them, miR-21-5p, miR-31-5p, and miR-155-5p, are known onco-miRNAs, meaning their overexpression tends to promote tumor development. MiR-21-5p, for example, has been shown in previous work to drive invasion in oral cancer cells through the Wnt/beta-catenin pathway, while miR-31-5p is considered a pivotal oncogenic factor in oral squamous cell carcinoma and miR-155-5p has been associated with metastasis and poor prognosis. The other two, miR-125-3p and miR-133a, are tumor suppressor miRNAs whose reduced expression is thought to remove brakes on malignancy. Crucially, the researchers did not stop at the miRNAs themselves. They also measured two of the key protein targets downstream of these regulatory molecules: p53, the famous guardian-of-the-genome tumor suppressor whose dysfunction is a hallmark of many cancers, and EGFR, the epidermal growth factor receptor, a cell surface signaling molecule frequently overproduced in head and neck cancers and a well-established therapeutic target.

The logic of combining miRNAs with their protein targets reflects a growing recognition in the biomarker field that single molecules are often deceptive. A lone miRNA fluctuating in saliva might reflect inflammation, tissue injury, or any number of non-cancerous conditions, producing false positives that would undermine a screening program. A lone protein marker might similarly lack the specificity needed to distinguish malignant from benign changes. By tracking a coordinated pattern, an oncogenic miRNA rising in parallel with its target protein shifting in the oral epithelial cells, the researchers hoped to build a signal robust enough to survive the noise of everyday biology. This systems-level approach, reading the conversation between regulator and regulated rather than either voice alone, is increasingly seen as the most credible path toward clinically useful liquid biopsies.

The experimental design was straightforward but technically rigorous. The team recruited 40 individuals divided among oral cancer patients, tobacco consumers without cancer, and healthy controls. From each participant, they collected saliva and exfoliated oral cells, the loose epithelial cells that naturally shed from the mouth’s lining and can be harvested by simple scraping or rinsing. Using quantitative real-time PCR, or qRT-PCR, a technique that amplifies and measures specific RNA sequences with exquisite sensitivity, they quantified the expression levels of the five miRNAs in saliva and the messenger RNAs for p53 and EGFR in the exfoliated cells. To confirm the protein-level findings, they employed immunocytochemistry, a method that uses antibodies to visualize where specific proteins accumulate within cells, providing an independent line of evidence that the molecular changes were real and not artifacts of RNA measurement.

The results were striking, particularly for two molecules. MiR-31-5p, the oncogenic microRNA, was elevated 23-fold in the saliva of tobacco consumers compared with healthy controls, a difference that reached statistical significance at p equals 0.002. In oral cancer patients, the elevation was even more dramatic: a 38-fold increase, also significant at p equals 0.002. Meanwhile, EGFR expression in the exfoliated oral cells rose 12-fold in tobacco consumers, significant at p equals 0.001, and 20-fold in oral cancer patients, significant at p equals 0.0003. The parallel trajectory is what makes these findings compelling. Tobacco users, who are at elevated risk of developing oral cancer but have not yet done so, already show a partial version of the molecular signature seen in full-blown disease, suggesting that these markers may capture the intermediate state of tissue transformation before a tumor becomes visible or symptomatic.

That intermediate signal is precisely what an effective screening tool needs. The other studied miRNAs and their target messenger RNA were also found to be deregulated across the participant groups, though the standout performance of the miR-31-5p and EGFR pair led the team to propose this combination as the most promising candidate panel. To assess diagnostic potential formally, the researchers used receiver operating characteristic curve analysis, a statistical framework that plots true positive rates against false positive rates across all possible diagnostic thresholds. The area under the resulting curve, a standard measure of diagnostic accuracy, allows researchers to judge how well a biomarker distinguishes diseased from healthy individuals, and the ROC framework will be essential in determining whether this panel can move from pilot observation to validated clinical test.

The implications for tobacco users are significant. Tobacco consumption, whether smoked or chewed, remains the dominant risk factor for oral cancer in many parts of the world, including India, where smokeless tobacco products are widely used and where the study was conducted. Current screening relies on visual examination by dentists or physicians and, when suspicious lesions appear, on incisional biopsy, a procedure that removes a piece of tissue for histopathological analysis. Biopsy is accurate but invasive, uncomfortable, and impractical to deploy repeatedly across large populations of asymptomatic at-risk individuals. A saliva-and-scrape test, by contrast, could in principle be administered in a dental chair in minutes, repeated at regular intervals, and used to flag individuals who warrant closer follow-up, potentially catching malignancies at stages when survival odds are dramatically better.

The researchers themselves are careful to frame the work appropriately. This was a pilot study with a small, single-cohort design involving just 40 participants, and the authors explicitly describe their findings as preliminary and hypothesis-generating. Before any clinical application, the biomarker panel must be validated in larger, independent cohorts drawn from different populations, with careful attention to confounding factors such as age, sex, oral hygiene, alcohol use, and the specific type and duration of tobacco exposure. Saliva composition can also vary with circadian rhythm, diet, and collection method, all of which will need standardization. These are the ordinary hurdles of translational diagnostics, and they do not diminish the value of the initial observation; they simply mark the road ahead.

Still, the study adds to a rapidly accumulating body of evidence that the mouth offers a uniquely accessible window into cancer biology. Because oral squamous cell carcinoma begins in the epithelium, the very layer of cells that lines the oral cavity and sheds into saliva, the disease leaves molecular traces in the easiest body fluid to collect. Previous studies have implicated salivary miRNAs in oral cancer detection, and work on miR-21, miR-30c-5p, miR-146a, and other candidates has steadily refined the field’s understanding of which molecules carry the most diagnostic weight. What this new pilot study contributes is the demonstration that pairing salivary miRNAs with protein markers measured in exfoliated cells can produce a coordinated, statistically robust signal, and that this signal appears in tobacco users before cancer develops. If larger studies confirm the pattern, the humble spit test could become a routine weapon in the fight against one of the world’s most preventable and most neglected cancers, turning the saliva of millions of tobacco users into an early warning system that currently does not exist.

Subject of Research: Salivary microRNA and p53/EGFR biomarkers for noninvasive early detection of oral squamous cell carcinoma in tobacco users

Article Title: Salivary miRNAs along with p53 and EGFR in oral exfoliated cells: a pilot study on candidate noninvasive biomarkers for the early detection of oral cancer in tobacco users

Article References: Vats, R., Yadav, P., Bano, A., Vashishtha, A., Narwal, A., & Bhardwaj, R. (2026). Salivary miRNAs along with p53 and EGFR in oral exfoliated cells: a pilot study on candidate noninvasive biomarkers for the early detection of oral cancer in tobacco users. Molecular Biology Reports, 53(1), Article 1661. https://doi.org/10.1007/s11033-026-12829-0

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12829-0

Keywords: oral cancer, salivary miRNAs, miR-31-5p, EGFR, p53, tobacco users, liquid biopsy, noninvasive diagnostics, oral squamous cell carcinoma, biomarkers, qRT-PCR, early detection