What's Happening?
Tempus AI, Inc., a leader in AI-driven precision medicine, has announced significant advancements in its Multimodal Foundation Models at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting. The company has developed novel biological foundation
models using over 500 petabytes of molecularly grounded data, including 45 million de-identified patient journeys and 1.5 million sequenced data records. These models aim to transform this data into unified patient representations, providing actionable insights to accelerate precision medicine in clinical settings and drug development. Tempus' latest model, trained on 2.5 million longitudinal records, has demonstrated the ability to predict patient outcomes without additional data or model fine-tuning. This includes accurately stratifying responses to standard cancer treatments, such as osimertinib for EGFR-mutant NSCLC patients, showcasing its potential to improve clinical trial design and patient care.
Why It's Important?
The advancements by Tempus AI represent a significant leap in the application of AI in healthcare, particularly in oncology. By leveraging vast amounts of multimodal data, Tempus' models can provide insights that were previously unattainable, potentially transforming how cancer treatments are developed and administered. This could lead to more personalized and effective treatment plans, improving patient outcomes and reducing healthcare costs. The ability to predict patient responses to treatments before clinical trials could also streamline the drug development process, making it more efficient and cost-effective. This development underscores the growing role of AI in precision medicine, offering new tools for physicians and researchers to enhance patient care and accelerate the discovery of new therapeutics.
What's Next?
Tempus plans to continue refining its multimodal foundation models to further enhance their predictive capabilities and clinical utility. The company aims to expand the application of these models across different types of cancer and other diseases, potentially broadening their impact on precision medicine. As these models are integrated into clinical practice, they may prompt changes in how clinical trials are designed and conducted, with a focus on leveraging AI to identify optimal patient cohorts and treatment strategies. The ongoing development of these models will likely attract interest from biopharmaceutical companies seeking to improve their drug development processes and outcomes.
