New Delhi, Dec 27 (NationPress) A group of researchers from the United States has successfully created and validated an artificial intelligence (AI)-driven, non-invasive tool designed to forecast the risk of head and neck cancer metastasis.

This innovative AI tool, developed by scientists from Mass General Brigham, is capable of predicting the likelihood that a patient's oropharyngeal cancer—a specific form of head and neck cancer originating in the throat—will progress. This could assist healthcare professionals in determining which patients require more aggressive treatment.

According to senior author Benjamin Kann of the Artificial Intelligence in Medicine (AIM) Programme at Mass General Brigham, "Our tool may assist in identifying which patients should undergo multiple interventions or are the best candidates for clinical trials involving intensive strategies such as immunotherapy or additional chemotherapy."

Kann further noted, "Our tool can also help recognize patients who might benefit from a reduction in treatment intensity, such as opting for surgery alone."

The findings of this research have been published in the Journal of Clinical Oncology.

Therapeutic options for oropharyngeal cancer, which include various combinations of surgery, radiation therapy, and chemotherapy, can be challenging for patients to tolerate and may have long-term adverse effects. Thus, it is crucial to identify patient subgroups that may benefit from either more or less intensive treatment modalities.

One method to achieve this is by evaluating whether the patient exhibits pathologic extranodal extension (ENE), which occurs when cancer cells invade beyond the lymph node into surrounding tissues. Currently, ENE can only be conclusively diagnosed through surgical removal and examination of lymph nodes.

The new AI-based tool utilizes imaging data from computed tomography (CT) scans to predict the number of lymph nodes exhibiting ENE—an important indicator of a patient’s prognosis and their likelihood of benefitting from intensified therapies.

When applied to imaging scans from 1,733 patients diagnosed with oropharyngeal carcinoma, the tool successfully predicted uncontrolled cancer spread and poorer patient survival outcomes. Integrating the AI's assessments with established clinical risk predictors enhanced risk stratification, resulting in more accurate survival predictions and assessments of cancer spread for individual patients.

"The AI tool facilitates the prediction of the number of lymph nodes affected by ENE, something that was previously unattainable, demonstrating its potential as a powerful and novel prognostic biomarker for oropharyngeal cancer that could enhance current staging protocols and treatment planning," said Kann.