Tempus AI today released clinical validation data showing its Immune Profile Score (IPS) more accurately predicts patient benefit from immune checkpoint inhibitors (ICIs) than established biomarkers such as tumor mutational burden, microsatellite instability and PD-L1 expression. Across four independent validation cohorts encompassing metastatic solid tumors, IPS achieved a hazard ratio of 0.45 for overall survival in ICI-treated patients, compared to less prognostic utility from conventional assays. Notably, IPS identified 13% of microsatellite stable colorectal cancer patients who experienced a marked survival advantage (HR=0.20) on ICIs—an at-risk population previously overlooked by standard testing—and classified 17% of patients with rare metastatic solid tumors as “IPS-High,” demonstrating a significant median survival gain (HR=0.26) despite lacking FDA-approved labels for immunotherapy. These results underscore IPS’s potential to expand appropriate ICI use and personalize treatment decisions based on multimodal DNA and RNA signals rather than single markers alone.

In a strategic move to capture growth in AI-driven pathology, Tempus AI has completed its acquisition of Paige, a digital pathology leader, and concurrently launched Paige Predict, an algorithm suite delivering biomarker insights across 16 cancer types. The combined business now leverages one of the industry’s largest libraries of whole-slide images and genomic profiles, enabling automated detection of clinically actionable features such as tumor infiltrating lymphocytes and specific mutational signatures. Tempus projects that integrating Paige’s image-based analytics with its existing genomic tests will accelerate adoption among academic centers and community oncology practices, potentially addressing an addressable market estimated at over 1 million pathology cases annually. Leadership expects the unified platform to drive new service revenues and enhance gross margins by reducing manual slide review time by up to 40%, while generating richer data sets to fuel future AI model development.