The Breakthrough

Breast cancer is hard to predict: patients with the same type of cancer often have different responses to the same medication. So a team of researchers tried to identify which genetic factors lead to which outcomes, which could lead to individualized treatment regimens. They looked at 40 genes, and then used AI combined with other data to narrow down which genetic signatures were most important for predicting response to medication and treatment. The team was able to identify 84 percent of women who would go into remission in response to paclitaxel. For the drug gemcitabine, the team was able to predict remission with 62 to 71 percent accuracy.

The Implications

Artificial intelligence could be a key to predicting drug outcomes because it doesn't only look at individual genes but at the sum of all the interacting genes. The tech could improve patient outcomes and help guide doctors as to which medications are best for which patients -- which hopefully means fewer people dying from common cancers.


Share This Article