NEW YORK – Predictive Oncology and Cvergenx said on Thursday that they have teamed up to develop a genomics- and machine learning-based approach to personalized radiation therapy and oncology drug discovery.
For its part of the partnership, Predictive Oncology will bring its Patient-centric Drug discovery by Active Learning (PeDAL) platform, which uses machine learning combined with a proprietary biobank to generate drug response predictions.
Cvergenx, meanwhile, will bring its precision genomics radiation therapy platform (pGRT), which the Moffitt Cancer Center spinout is currently evaluating in a Phase II clinical trial as a tool to inform treatment for patients with triple-negative breast cancer. The informatics-based tool assigns patients to a genomics-adjusted radiation dose, or GARD, based on a gene-expression-based radiation sensitivity index (RSI) and biological information.
The partners seek to use their two platforms to further optimize radiotherapy and to identify targets to develop radiosensitizers and radioprotective drugs, in turn improving patient outcomes. Predictive Oncology and Cvergenx said the collaboration could also result in repurposing existing compounds or developing an entirely new class of drugs.
"Today, radiotherapy is prescribed based on a one-size-fits-all approach, where all tumors are treated with uniform doses of radiation therapy," Javier Torres-Roca, cofounder and acting CEO of Cvergenx, said in a statement. "pGRT provides the first clinically validated approach to optimize radiotherapy prescription dose for each individual patient. The pGRT platform has been shown to correctly identify radiosensitizers and radioprotectors from large pharmacogenomic screens."