One of the key challenges in developing new cancer therapeutic drugs is effectively matching the best combination of drugs with the right patients.

For a study recently published in Cancer Cell, however, researchers developed a deep learning algorithm that was able to analyze tumor data and recommend the best possible treatments for patients. 

According to Trey Ideker, PhD, professor at University of California San Diego School of Medicine and Moores Cancer Center, “Most AI systems are 'black boxes'—they can be very predictive, but we don't actually know all that much about how they work.”

For AI to be useful in health care, researchers have to be able to see inside the black box to understand how the system comes to its conclusions, explained Ideker, who is also co-director of the Cancer Cell Map Initiative and the National Resource for Network Biology. “We need to know why that decision is made, what pathways those recommended drugs are targeting and the reasons for a positive drug response or for its rejection.” 

Currently, just four percent of all cancer therapeutic drugs under development earn final approval from the FDA, often because developers have an incomplete understanding of the mechanisms governing drug response. 

To overcome this issue, researchers developed a deep learning model called DrugCell, which was specifically designed so that its recommendations can be explained to human clinicians.

The team trained DrugCell on more than 1,200 tumor cell lines and their responses to nearly 700 FDA-approved and experimental therapeutic drugs, totaling more than 500,000 cell line/drug pairings. Researchers also validated some of the model’s conclusions in laboratory experiments.

Using DrugCell, researchers can input data about a tumor and the system returns the best-known drug, the biological pathways that control response to that drug, and combinations of drugs to best treat malignancy.

“We were surprised by how well DrugCell was able to translate from laboratory cell lines, which is what we trained the model on, to tumors in mice and patients, as well as clinical trial data,” said Brent Kuenzi, PhD, a postdoctoral researcher in Ideker’s lab.

The team's ultimate goal is to get DrugCell into clinics for the benefit of patients, but the study authors caution there's still a lot of work to do.

"While 1,200 cell lines is a good start, it's of course not representative of the full heterogeneity of cancer," said Jisoo Park, PhD, another of the team’s postdoctoral researchers. "Our team is now adding more single-cell data and trying different drug structures. We also hope to partner with existing clinical studies to embed DrugCell as a diagnostic tool, testing it prospectively in the real world.”