Look Beyond Molecular Testing to Predict Response to Immunotherapy Agents, Study Says

Source: Dana-Farber, January 2020

For all their potential to curb or even cure some cancers, drugs known as immune checkpoint inhibitors come with a caveat: They’re effective in only a subset of patients. Predicting who those patients are, and understanding why others don’t respond as well, remains a major challenge.

In a new study, researchers at Dana-Farber in collaboration with investigators in Germany, the Broad Institute of MIT and Harvard, and Massachusetts General Hospital found that, for patients with metastatic melanoma, a combination of molecular data and clinical information — including whether the patient has received previous treatment with a different checkpoint inhibitor — provides the best forecast of whether a checkpoint inhibitor is likely to be effective.

“Our findings show that while molecular data about the tumor is very powerful by itself, it isn’t sufficient, at this point, for determining which patients will respond to this type of checkpoint inhibitor,” says David Liu, MD, MPH, of Dana-Farber and the Broad Institute of MIT and Harvard, the first author of the study, which is being published in Nature Medicine. “Tools for predicting responsiveness will need to take multiple factors — clinical information as well as genomic analysis — into account.”