Every Cure, a nonprofit organization, is leveraging Google Cloud to revolutionize drug repurposing by matching existing drugs with diseases lacking approved treatments. Co-founded by Dr. David Fajgenbaum, who discovered that the drug sirolimus could treat his rare Castleman disease, Every Cure aims to address the 14,000 recognized diseases without FDA-approved treatments. The nonprofit uses Google AI to analyze 75 million drug-disease predictions, identifying potential life-saving treatments more efficiently.
Traditional drug repurposing research typically focuses on one disease or one drug at a time. Every Cure’s approach, called computational pharmacophenomics, examines all drugs and all diseases simultaneously, scoring the strength of evidence for each possible interaction. This method requires a robust data foundation, which Every Cure achieves using Google Cloud’s BigQuery and Cloud Storage. The platform, powered by Google Cloud GPUs, has accelerated the prediction pipeline from 100 days to approximately 17 hours, enabling dozens of experiments per month.
In just one year, Every Cure has launched nine active drug repurposing programs. These include advancing treatments for ultra-rare pediatric disorders and highlighting evidence that perioperative lidocaine may reduce the risk of breast cancer metastasis by up to 29% at five years. Despite its low cost and wide availability, lidocaine remains underutilized due to lack of financial incentives for stakeholders. Every Cure’s goal is to repurpose 15 to 25 treatments for 15 to 25 diseases by 2030, matching the output of larger pharmaceutical companies.
Dr. Fajgenbaum emphasizes the critical role of Google Cloud in scaling their machine learning models. “With a bold mission like ours, we need amazing partners, and we would not be close to where we are without Google Cloud,” he says. By leveraging existing drugs, Every Cure aims to save lives more efficiently and cost-effectively, with advancing an existing drug costing roughly 1% of developing a new molecule.