Review new research on a one-hour brush test that could help dentists identify which suspicious oral lesions warrant biopsy and which may be safely monitored.
By Genni Burkhart, Editor
Not every suspicious oral lesion needs a scalpel biopsy, but deciding which ones do and which don't isn’t always exact.
Fortunately, a new non-invasive brush test could spare more than 90% of patients with low-risk oral potentially malignant disorders from unnecessary and sometimes painful biopsies. In a recent study published in Biomarker Research, researchers found that the brush test distinguished oral squamous cell carcinoma from oral leukoplakia and oral lichen planus with 95.5% accuracy.
While this test isn't designed to replace a biopsy, it could help better determine who needs one. That distinction is especially useful when a lesion remains suspicious, but the examination alone doesn’t provide a clear answer. In most cases, dentists must decide whether to continue monitoring the area or refer the patient for a tissue biopsy. But now, the brush test may offer more information before they make that decision.
This new test collects cells with a small brush, analyzes the activity of four genes, and produces a malignancy score in less than an hour. As such, a higher-risk result could support faster referral and biopsy, while a lower-risk result could help clinicians continue surveillance without immediately removing tissue.
Easier, Less Invasive
The test starts with a brush sample taken directly from the lesion. The clinician rotates the brush 10 times, and the collection takes less than five minutes. No incision or tissue removal is required.
This test could also make follow-up easier for patients with lesions that need to be watched over time. Clinicians could then repeat the brush test without putting the patient through another scalpel biopsy.
A higher malignancy score would still lead to a conventional biopsy and histopathological examination. A lower-risk result, however, could help some patients avoid an invasive procedure that would only confirm a benign or low-risk lesion. Researchers estimate that the test could prevent more than 90% of unnecessary scalpel biopsies among patients with low-risk oral potentially malignant disorders. It may also help patients who hesitate to undergo a biopsy because of pain or anxiety.
How the Brush Test Works
After collecting the cells, the laboratory measures the activity of four genes. Two of them, INHBA and S100A16, are associated with oral squamous cell carcinoma. The other two, YAP1 and POLR2A, give the laboratory a reference point for interpreting the results.
Researchers had originally used a 16-gene test on small tissue biopsies. However, brush samples contain fewer cells, and several genes didn’t produce reliable results. Therefore, the current four-gene version performed better with the material collected by the brush.
The laboratory measures gene activity using RT-qPCR, a widely used technology for detecting and quantifying genetic material. An algorithm then turns the results into a malignancy score, with the full process taking less than an hour.
It's important to note that the samples in this study held up remarkably well during storage. Researchers also found that the malignancy score remained stable for 10 months at room temperature. That fact could allow a dental practice to collect the sample and send it to an outside laboratory without refrigeration or rush processing.
Methods and Results
Researchers collected 1,090 brush samples from 545 adults, collecting one brush sample from the lesion and a second from normal oral mucosa on the opposite side of the mouth. After excluding samples that didn’t meet the study requirements, researchers tested 893 samples with the final four-gene assay.
The study included 443 patients with oral squamous cell carcinoma, 63 with oral leukoplakia, and 39 with oral lichen planus. Histopathology had already confirmed each diagnosis, so researchers could compare the brush test results with the established standard.
When they compared oral squamous cell carcinoma with the oral leukoplakia and oral lichen planus samples, the test reached 95.7% sensitivity, 95.1% specificity, and 95.5% overall accuracy. False-positive and false-negative rates both remained below 5%, and age and sex didn’t affect its performance. However, the results changed when researchers compared the cancer samples with normal-looking tissue from the same patients. In that analysis, sensitivity, specificity, and accuracy each dropped to 84%.
Researchers believe field cancerization may explain the difference. Molecular changes can extend beyond the visible tumor, so nearby tissue may appear normal during an examination even as it already shows cancer-related changes.
The findings also confirm that the brush test would serve as a triage tool rather than a stand-alone diagnosis. Clinicians would still need to consider the lesion’s appearance, history, location, and other clinical findings.
More Research Before Clinical Use
The results are promising, but researchers still need to test the brush assay in broader patient groups. All participants in this study were from a single region of India, and most had oral squamous cell carcinoma. Furthermore, the study included far fewer patients with oral leukoplakia or oral lichen planus and didn’t include a separate healthy control group.
Earlier studies validated the tissue-based 16-gene qMIDSV2 assay in patient groups from the United Kingdom, China, and India. However, for this study, the researchers adapted it for use on brush samples and narrowed it down to only four genes. Before the test can move into clinical use, researchers would need to confirm the results in broader patient populations and secure support for further development.
Queen Mary University of London is now seeking a commercial partner to help advance that work. If that works out, the university estimates that the test could reach the market within two years.
Reference:
- Teh, MT., Patil, R., Tekade, S.A. et al. INHBA–S100A16 dysregulation enables a non-invasive molecular stratification platform for rapid detection of oral squamous cell carcinoma: results from a large diagnostic case-control study. Biomark Res 14, 76 (2026). https://doi.org/10.1186/s40364-026-00963-7
