How are scientists looking for a way to treat cancer?
How are scientists looking for a way to treat cancer?This research team was led by Dr. Itan Rupin and Dr. Luke Morris. They analyzed a large dataset containing information on more than 2880 cancer patients. These patients had 18 different types of solid tumors and all were treated with immunotherapy. The team assessed more than 20 different clinical, pathologic and genomic features. Using machine learning, they tried to find out which combination of features could best predict the patient's response to immunotherapy. Their results were published in Nature cancer on june 3, 2024.After creating and testing different machine learning models, scientists have created a new type of AI scoring system called LORIS (Logistic Regression-Based Immunotherapy Response Score). It is based on tumor mutational burden as well as five clinical features that are routinely collected from patients. These include the patient's age, type of cancer, history of cancer therapy, blood albumin and blood NLR.How can the new AI scoring system help in cancer treatment?This program works better than other earlier programs. This program can not only tell whether the patient will benefit from this medicine or not, but can also tell how long the patient will live. Scientists say that this program can help doctors choose the right treatment for patients. Howevermore large studies are needed so that this program can be used in hospitals. In the NIH report, Dr. Morris says, "We have created a new program for patients of many different cancers that works with just six simple pieces of information. Some earlier programs were very complex, but this program is easy for doctors to understand."