The Seattle-based Swedish Cancer Institute (SCI) and precision medicine company GNS Healthcare have announced a machine learning collaboration with the aim to strengthen, accelerate, and potentially expand the delivery of precision medicine in breast cancer care.
GNS Healthcare’s machine learning and simulation platform, REFS (Reverse Engineering and Forward Simulation), will be employed to build computer models and ultimately a software interface tool that will link clinical and molecular data to treatment and outcomes across SCI's patient population, officials attest.
What’s more, the collaboration will use aggregated patient data from SCI's breast cancer research registry, and personalized medicine research program (PMRP), to build these computer models. REFS-generated models can support clinical decision-making through a better understanding of treatment success based on individualized genetic backgrounds. As such, these models can be used to simulate potential treatment interventions and their effect on disease outcomes. The results of these analyses can then be interpreted through GNS's software interface to support the optimal delivery of treatments to individual patients, according to officials.
Henry Kaplan, M.D., head of breast medical oncology, and principal investigator of the breast cancer research registry, emphasized that "this project has the potential to optimize use of genomic information in the care of our breast cancer patients, to include through our SCI Molecular Tumor Board."
Indeed, by automating the process of connecting cancer patients with their most effective treatment, the GNS-SCI collaboration demonstrates a framework for scaling precision medicine initiatives. The approach makes it possible to generate the data needed by healthcare payers to prove the value of precision medicine, an outcome that would bring clarity to reimbursement decisions and expand access to life-saving treatments and better outcomes for patients, officials note.
“Few oncologists today have the complete training or time necessary to decipher complex results of a tumor's biologic fingerprint. GNS' REFS-generated models will streamline this process and enhance our PMRP's work to support providers and their patients with treatment recommendations," Thomas Brown, M.D., executive director of the Swedish Cancer Institute, said in a prepared statement. "This unique collaboration is intended to facilitate the practice of personalized, i.e. precision, medicine at the SCI, increasing access of this important resource to our patients, while allowing our doctors to remain patient-focused."
