Multiparameter Imaging Method Enhances Head and Neck Cancer Diagnosis

Researchers from the University of Helsinki, University of Turku, and the Max Planck Institute for Molecular Biomedicine have developed an innovative machine learning-based method for analyzing head and neck squamous cell carcinoma, one of the ten most common cancer types. This new imaging analysis technique combines biomarkers of cell behavior with morphological analyses of individual cells and tumor tissue structure, creating a unique "fingerprint" for each patient. 

The study's most significant outcome was the identification of two previously undetected patient groups with markedly different prognoses. One group showed an exceptionally good prognosis, while the other had a particularly poor outlook. The difference was attributed to a specific combination of cancer cell state and the composition of surrounding tissue. In the group with poor prognosis, the disease's aggressiveness was linked to signaling between cancer tissue and surrounding healthy connective tissue, mediated by the epidermal growth factor (EGF). 

Sara Wickström, senior author of the paper published in Cell, emphasized the breakthrough nature of these findings, stating, "For the first time, we have shown that specific combinations of malignant cells and tissue cell types in what is considered healthy tissue have a strong prognostic effect on cancer progression."

The method's ability to identify patients with poor prognosis who could benefit from aggressive treatment, as well as those with good prognosis who might only require less aggressive interventions, could significantly impact treatment strategies and patient quality of life. 

The researchers are now developing a diagnostic test based on this method, which could lead to more precise diagnoses for head and neck cancers. They are also exploring its potential application in other cancer types, such as colorectal cancer. 

Wickström highlighted the affordability of the method, noting that it primarily requires their developed algorithm and a special combination of antibodies, making it a cost-effective option in the context of overall cancer treatment expenses. “Imaging of cancer biomarkers using antibody stainings is already in clinical use. Therefore, the method will not be particularly expensive, since it only requires the algorithm developed by us and a special combination of antibodies. Considering the cost of cancer treatment, this is actually quite affordable.”