Researchers at Florida International University and First Ascent Biomedical have used an artificial intelligence (AI)-powered, functional precision medicine (FPM) platform to identify unique therapeutic treatment options for children with relapsed cancer diagnoses. They published their study, “Feasibility of Functional Precision Medicine for Guiding Treatment of Relapsed or Refractory Pediatric Cancers,” in Nature Medicine.
“Children with rare, relapsed, or refractory cancers are often faced with limited treatment options and few predictive biomarkers that can enable personalized treatment recommendations,” wrote Diana Azzam, PhD, an assistant professor at Florida International University and co-founder of First Ascent Biomedical, and her colleagues. “Implementation of FPM, which combines genomic profiling with drug sensitivity testing (DST) of patient-derived tumor cells, has potential to identify treatment options when standard-of-care is exhausted.”
Although cancer diagnoses are often grouped together and are publicly thought of as a singular condition, any single cancer can be as unique as the individual diagnosed. Treatment options are varied and work differently for each patient. Personalized treatments that work toward remission are challenging to identify and use effectively.
The new study began with 25 patients with relapsed solid or blood-based cancers who were selected to undergo DST and genomic profiling. Nineteen patients were able to complete the process and receive treatment recommendations. Of these patients, 14 remained in the study with six patients receiving FPM-guided treatments, and eight patients receiving treatments based on physician’s choice (PC).
Of the six patients to get FPM-guided treatments, the authors found that “83% (five patients) experienced a greater than 1.3-fold improved progression-free survival compared to their previous therapy.” They also “demonstrated a significant increase in progression-free survival and objective response rate compared to non-guided patients,” wrote the authors.
FPM treatment in action
The process of using FPM includes sampling tumors or blood and growing them to mimic the environment in the body. These samples are then exposed to more than 120 FDA-approved drugs individually and in combination. The use of AI in FPM “gets rid of the guesswork and delivers a list of the most effective drugs that the oncologist can work with,” said Tomás R. Guilarte, PhD, dean of Stempel College and a co-author of the study.
Actionable results can be produced in about a week for clinicians to begin treatment. “We have shown the ability to return data rapidly, and that [these] data improved outcomes in patients with guided treatments. Similarly, we foresee the potential of an FPM approach to reduce toxicity and cost,” Azzam told GEN.
One patient, who’s cancer relapsed 14 months after initial treatment, was one of the six patients to receive FPM-guided treatment. After 33 days of treatment, he reached remission, while his prior treatment took 150 days for results. He has been in remission for more than two years following the FPM-guided treatment. “I believe that AI is a critical component of the future of FPM and driving better outcomes for cancer patients,” said Azzam.
The future of AI in cancer treatment
“FPM and AI represent the future of personalized cancer treatments,” Azzam said. “This approach should move earlier in a patient’s treatment journey. However, we recognize the need for additional clinical validation. Ultimately, compelling clinical data will drive acceptance and bring FPM and AI to patients earlier, similar to how genomics is becoming available to patients in earlier stages of treatment.”
This proof-of-principle study is the first step in establishing a new approach to cancer treatments using AI-informed precision medicine. Azzam is hopeful that future studies will establish the impact on a larger scale to “demonstrate the benefit to patients, doctors, hospitals, and insurance companies.”
Azzam also acknowledged a large hurdle to implementing FPM on a large scale is “education and availability.” She pointed out to GEN that “clinicians worry about efficacy, safety, and toxicity. The benefit of this FPM approach, testing 100s of FDA approved drugs, is these are drugs [that] the clinicians are familiar with… When they see what drugs work and the underlying mechanism why they work they have confidence in treating based on the data.”
Data and its appropriate analysis, here using an AI platform, are the driving forces for improved cancer treatments. “Our published study and the current open studies provide the blueprint to drive education and acceptance of this new approach to personalized cancer care,” concluded Azzam.
