The U.S. Food and Drug Administration (FDA) has granted Breakthrough Device designation to ORAL ONC-InCYT, an investigational cytology test designed to help clinicians assess suspicious oral lesions.  

With the ORAL ONC-InCYT test, a brush is used to collect cells from a suspicious white or red oral lesion. Advanced imaging and diagnostic deep learning models are then used to evaluate cellular morphology and biomarker-related features and generate an Oral Cancer Numerical Index, a quantitative risk score.  

The Breakthrough Device designation applies to use of the test in adults with suspected oral potentially malignant disorders or lichenoid conditions.  

The Oral Cancer Numerical Index was developed and validated in a retrospective study of individuals with oral potentially malignant disorders who were divided into a training cohort (n = 384) and a holdout test cohort (n = 164). At the prespecified rule-out threshold of 37.6 or lower, the index demonstrated a sensitivity of 92% and a negative predictive value of 97%. The index score increased with worsening histopathologic severity.  

In a separate study, the Oral Cancer Numerical Index achieved an area under the receiver operating characteristic curve of up to 0.99 in distinguishing malignant oral lesions from healthy controls.  

“Cancer is, at its core, a disease of cellular dysregulation, so the cells collected from a suspicious lesion carry some of the most direct evidence of what is happening in that tissue,” stated John T. McDevitt, PhD, Chief Scientist at OraLiva. “ORAL ONC-InCYT is designed to capture those cellular signatures, combining cell morphology and biomarker-related features with deep learning to produce a quantitative index. We see cytosignatures as a distinct class of biomarker, complementary to the nucleic acid, protein, and microbiome approaches the field has pursued for years, where reaching robust clinical validation has proven difficult. FDA brings rigor to that challenge, and Breakthrough Device designation gives us a framework to work closely with the agency as we generate the clinical evidence needed to support marketing authorization.” 

OraLiva, the company developing the test, noted that the Oral Cancer Numerical Index is intended to be used alongside clinical evaluation to support clinical decision-making.  

OraLiva is also exploring the use of its AI cytology platform for the early detection of other epithelial cancers.  

ASCO AI in Oncology is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO^®). The ideas and opinions expressed in ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.