Chemotherapy, radiation, and surgical removal of tumors have long been standard cancer treatments. However immunotherapies which leverage the body’s immune system to target cancer cells, have become available in recent decades. A particular kind of immunotherapeutic — checkpoint inhibitors — have demonstrated improved outcomes for many cancer patients, though they are not effective for everyone. Identifying patients likely to benefit from these treatments remains a challenge.

A team led by Dr. Eytan Ruppin and Dr. Luc Morris developed a predictive tool named LORIS (logistic regression-based immunotherapy-response score). By analyzing data from over 2,880 cancer patients, the team used machine learning to identify six key variables that predict a patient’s response to immune checkpoint inhibitors. These variables include tumor mutational burden and five clinical features: patient age, cancer type, therapy history, blood albumin, and blood NLR.

LORIS outperformed existing models, offering a simple, accessible method for clinicians to predict immunotherapy outcomes and guide treatment decisions. Further clinical validation is needed to confirm its effectiveness.

To read more, click here. 

[Source: NIH Research Matters, June 25th 2024.]