Valar Labs Publishes Study Validating AI for Predicting Chemotherapy Response in Pancreatic Cancer
Valar Labs, a leader in computational histology and precision oncology, has announced a major breakthrough in the treatment of advanced pancreatic cancer with the development and validation of Vitara Pancreas ChemoPredict, an AI-powered diagnostic tool that helps select the optimal first-line chemotherapy for patients. The findings were published in the Journal of Clinical Oncology (JCO), a top-tier peer-reviewed medical journal, marking a significant milestone in the application of artificial intelligence to cancer care. Pancreatic cancer remains one of the most lethal forms of cancer, with limited treatment options and a poor prognosis. Currently, oncologists must choose between two standard first-line chemotherapy regimens—FOLFIRINOX or Gemcitabine/Nab-Paclitaxel—often based on a patient’s overall health rather than tumor-specific biology. This trial-and-error approach can delay effective treatment and expose patients to unnecessary side effects. For decades, the medical community has sought reliable biomarkers to guide these critical decisions. Valar Labs’ new AI diagnostic, Vitara Pancreas ChemoPredict, addresses this unmet need by analyzing standard hematoxylin and eosin (H&E)-stained pathology slides—widely available in routine clinical practice—using a proprietary algorithm. The system identifies subtle histological patterns in tumor tissue that correlate with response to specific chemotherapies, uncovering predictive information invisible to the human eye. The study, titled “Development and Validation of a Computational Histology Artificial Intelligence-Powered Predictive Biomarker for Selection of Chemotherapy in Advanced Pancreatic Cancer,” was rigorously validated using data from two prospective clinical cohorts: the PanCAN SPARK health data platform and the COMPASS trial. The results demonstrated that the AI model could accurately predict which chemotherapy regimen a patient is most likely to respond to, significantly improving the precision of treatment selection. “Publication in the Journal of Clinical Oncology underscores the scientific rigor and clinical validity of our approach,” said Anirudh Joshi, CEO and co-founder of Valar Labs. “For too long, first-line treatment decisions in pancreatic cancer have been a guessing game. Now, we can provide oncologists with a data-driven tool to personalize care and potentially improve outcomes.” Dr. Andrew Hendifar, lead author of the study and Pancreatic Cancer Medical Director at Cedars Sinai Samuel Oschin Cancer Center, praised the findings: “Biomarkers to guide choice of first-line chemotherapy in advanced pancreatic cancer have been a decades-long goal. We are excited to validate a predictive AI biomarker and look forward to integrating it into clinical practice.” The study also highlights the power of AI in leveraging de-identified patient data from large-scale research platforms. Anna Berkenblit, MD, MMSc, Chief Scientific and Medical Officer at PanCAN, emphasized that this work shows how AI can accelerate research and translate discoveries into real-world benefits. Kawther Abdilleh, PhD, Director of Data Science and Informatics at PanCAN, co-authored the study and noted that such initiatives exemplify the potential of AI to transform cancer care through data-driven insights. Following successful validation, Valar Labs is offering the Vitara Pancreas ChemoPredict test under early access to oncologists treating advanced pancreatic cancer patients. The test enables clinicians to make more informed, personalized treatment decisions based on the unique biology of each patient’s tumor. Valar Labs, already a leader in AI diagnostics for genitourinary cancers with its Vesta bladder cancer portfolio, is now expanding into gastrointestinal cancers with this new tool. The company’s mission is to reduce uncertainty in cancer treatment by turning routine pathology slides into actionable clinical insights. With this innovation, Valar Labs is advancing a new era of precision oncology—where artificial intelligence helps match patients to the right therapy at the right time, potentially improving survival and quality of life for thousands of individuals affected by pancreatic cancer.
