Precision medicine approach combining AI, DNA and drug testing shows gains against relapsed childhood cancers

17 April 2024

Conor Hale / FiercePharma

Researchers have demonstrated that a functional precision medicine approach—combining artificial intelligence, DNA sequencing, mutation mapping and drug sensitivity testing—could feasibly help tackle hard-to-treat cancers regardless of their type, including among children with relapsed solid tumors and blood-based malignancies.

The process, being developed by First Ascent Biomedical, showed promise in a small study published this week in Nature Medicine. It begins by collecting, culturing and analyzing tumor cells from each individual patient and then exposing those samples to more than 10 dozen FDA-approved drugs, including some that are aimed at cancer and some that are not, such as statins for high cholesterol.

Picking out the best performers and potential combinations takes about a week. Out of six patients who received these guided therapies, five showed strong improvements in their progression-free survival—this time outpacing the periods they had spent cancer-free, before the disease returned after at least two initial treatments.

“The results are exciting because cancer that comes back is much harder to treat. Seeing improvement in 83% of patients is incredibly promising,” said Florida International University cancer researcher Diana Azzam, Ph.D., who also serves as First Ascent’s co-founder and scientific advisor. “This could be the way we turn cancer into a manageable disease.”

While performing ex vivo experiments with drug libraries isn’t necessarily new, turning around actionable information such as this to oncologists and clinicians fast enough that it can be used has proven to be a challenge.

In the study, delivering drug sensitivity results took a median of nine days for hematological cancers and 10 days for solid tumors, but not longer than four weeks. According to researchers, that was significantly faster than genomic profiling, which may take as long as two to nine weeks—and in pediatric patients with aggressive disease, where standard therapies have been exhausted, that speed is essential.

“The Azzam lab approach gets rid of the guesswork and delivers a list of the most effective drugs that the oncologist can work with,” said Tomás Guilarte, Ph.D., dean of the FIU Robert Stempel College of Public Health & Social Work and one of the authors of the Nature Medicine paper. “It’s accelerating our understanding of which cancer treatments work best for patients and their specific needs.”

In the study, the patient’s regimen was altered with an additional targeted drug if one was identified in subsequent genetic testing. 

The researchers said the study is additionally important because the majority of its participants were members of ethnic minority populations, who can have different responses to treatments. And, while survival rates have improved in childhood cancers such as leukemia, progress has been slower in relapsed and refractory cases, which may lack genetic targets.

“The First Ascent FPM platform is a sophisticated integration of drug sensitivity testing and genomics, which are analyzed by our novel AI-driven technology. These results mark a true breakthrough, as we were able to provide the vast majority of study participants with a highly personalized and actionable treatment plan,” said First Ascent co-founder and CEO Jim Foote. “As a father who tragically lost his son to cancer, I wanted something like this for my son and I understand the urgency and importance of our work.”

Miami-based First Ascent is working to establish a larger, CLIA-certified commercial laboratory to help scale up the reach of its personalized approach, while the researchers said the methods warrant further validation in larger prospective studies. 



Our news

All news

Media Center

Read more