NEW YORK – Tempus AI on Wednesday said it signed multiyear, strategic collaboration agreements with AstraZeneca and Pathos AI to develop a multimodal foundation model for oncology.
The firms hope to gain biological and clinical insights, discover novel drug targets, and accelerate development of therapeutics. Researchers will build the model using Tempus' AI-driven platform and repository of de-identified patient data.
Under the terms of the agreement, Chicago-based Tempus will receive $200 million in data licensing and model development fees. All three partners will have access to the model to support their individual programs.
The new agreement expands an existing partnership between AstraZeneca and Tempus, originally signed in 2021. Under that older agreement, Tempus and AstraZeneca have been collaborating on a prospective clinical study to identify biomarkers of response in small cell lung cancer (SCLC) patients and a study in non-small cell lung cancer to create a multiomics dataset for discovering biomarkers and developing precision treatments and diagnostics.
"Generative AI and the emergence of large multimodal models is the final catalyst needed to usher in precision medicine in oncology at scale," Tempus Founder and CEO Eric Lefkofsky said in a statement. "Tempus has spent the last decade investing billions of dollars into collecting the necessary data needed for a foundation model of this kind to take shape."
According to Tempus, its data repository is larger than the Cancer Genome Atlas (TCGA), with more than 8 million de-identified patient records. These include 2 million records with imaging data, 1.5 million records with matched clinical data and genomic information, and over 250,000 records with transcriptome profiles.