Khaberni - A British research team succeeded in developing a non-invasive diagnostic tool capable of predicting the likelihood of throat cancer spread.
Throat cancer, a type of head and neck cancers, and the AI-based tool assists doctors in identifying patients who need intensive treatment and others who may have their treatment intensity reduced.
The results of the study, prepared by scientists from the Brigham and Women's General Hospital and the Dana-Farber Cancer Institute, were published in the Journal of Clinical Oncology, showing that the new tool can contribute to improving the accuracy of predicting disease progression and survival.
Dr. Benjamin Kahn, the lead researcher in the study and a member of the Artificial Intelligence in Medicine program at Brigham and Women's General Hospital and a radiation oncologist at the Dana-Farber Institute and Brigham and Women's Hospital, explained that the tool could help in determining the most suitable patients for receiving multiple treatments or participating in clinical trials based on intensive therapeutic strategies such as immunotherapy or adding chemotherapy. It could also be used to identify patients who may benefit from reducing the intensity of treatment, such as by undergoing only surgery.
The treatment of oral pharyngeal cancer usually depends on a combination of surgery, radiation therapy, and chemotherapy, which are treatment options that can be strenuous and leave long-term side effects. Therefore, the need arises for accurate tools that help in classifying patients according to the degree of risk and directing treatment accordingly.
One of the main factors affecting the treatment plan is what is known as extranodal extension, a condition in which cancer cells spread outside the lymph node to the surrounding tissues. Until now, this condition can only be confirmed through surgical intervention and examination of the lymph nodes.
To overcome this challenge, researchers developed an artificial intelligence tool that analyzes CT images to predict the number of lymph nodes affected by extranodal extension, an important indicator of the severity of the disease and the likelihood of benefiting from intensive treatment.
When testing the tool on medical images belonging to 1,733 patients with oral pharyngeal cancer, the tool demonstrated its ability to predict uncontrolled cancer spread and decreased survival chances.
Additionally, integrating the tool's results with established clinical indicators improved the accuracy of risk classification and predicting the disease course for each patient individually.
Dr. Kahn affirmed that the tool represents a "new and powerful predictive indicator" that was previously unmeasurable, and it may contribute to the development of current disease staging systems and improve treatment planning, supporting a move towards more precise and personalized medicine for head and neck cancer patients.
