Choosing the right therapy for non-small cell lung cancer patient is not always an easy decision. Biomarkers can change during the therapy, making that treatment ineffective. The researchers at Moffitt Cancer Center have made new advancements in predictive lung cancer treatment. They have developed a non-invasive method to analyze tumor mutations and biomarkers with the purpose to determine the best course of lung cancer treatment.
In the article published in Nature Communications, the research team has demonstrated how a deep learning model can identify which non-small cell lung cancer patients are sensitive to tyrosine kinase inhibitor treatment and who would benefit from immune checkpoint inhibitor therapy. The researchers have utilized PET/CT imaging with radiotracker 18F-Fluorodeoxyglucose. The process works to help in accurately characterizing tumors.
In this study, the team has developed a deep learning model using retrospective data from non-small cell lung cancer patients. The model classifies EGFR mutation status by generating an EGFR deep learning score of each patient. Matthew Schabath, Ph.D., associate member of the Cancer Epidemiology Department said, “This type of imaging, 18F-FDG PET/CT, is widely used in determining the staging of patients with non-small cell lung cancer. The glucose radiotracer used is also known to be affected by EGFR activation and inflammation.”
The researchers have found that EGFR deep learning score was positively associated with longer survival of patients treated with tyrosine kinase inhibitors and negatively associated with durable clinical benefit. Robert Gillies, Ph.D., chair of the Cancer Physiology Department said that we would like to perform further studies but believe this model could serve as a clinical decision support tool for different treatments.
The information shared in this blog is for educational purposes only. If you face any symptoms or medical health problems, please consult your healthcare practitioner immediately.