A technology company uses artificial intelligence to assist in cancer drug development has launched a study that will collect data on up to 1,000 blood cancer patients over the course of a year.
San Francisco-based Notable said Wednesday it had launched the study, titled ANSWer, which will collect de-identified specimens with matched clinical data from participants in U.S. and Canadian clinical networks, at the time of their entry into the study and during subsequent visits. Patients with acute myeloid leukemia, acute lymphoblastic leukemia, chronic myelogenous leukemia, multiple myeloma, lymphomas, myeloproliferative disorders and others will be included. The goal is to establish a tumor registry with annotated clinical outcomes.
“The observational clinical trial that we’re kicking off will give us the opportunity to test more patients than ever before, allowing us to continue increasing the platform’s predictive value,” Notable CEO Matt De Silva said in a statement. “Notable’s platform improves with every patient, and the ANSWer study brings our technology one step closer to being used as a tool to help physicians and patients make complex treatment decisions in the clinic, which is our ultimate goal.”
Notable’s technology is designed to help predict patient responses to drugs in order to guide therapeutic decisions and also find clinical trials. Last year, it launched a feasibility study with Stanford University and Tempus to determine if it was possible to complete genetic sequencing and drug sensitivity recommendations in a clinically reasonable amount of time.
The study enrolled 21 patients with myelodysplastic syndrome – a type of blood cancer that can progress to acute myeloid leukemia – whose disease was refractory to drugs known as hypomethylating agents. The study collected bone marrow samples and screened them against up to 76 marketed and investigational drugs, while genomic data were entered into a computational biology model. Overall, the study found that Notable’s system had an 84% accuracy rate in predicting patient responses to drugs.
Photo: MF3d, Getty Images
