The Molecular Twin Precision Oncology Platform developed at Cedars-Sinai Cancer has been used by investigators to identify biomarkers that surpass the standard test for predicting pancreatic cancer survival. The study, published in the journal Nature Cancer, showcases the potential of this precision medicine and artificial intelligence (AI) tool to enhance cancer treatment.
Dr. Dan Theodorescu, director of Cedars-Sinai Cancer and senior author of the study, emphasizes the versatility of Molecular Twin, stating that it can be employed to study various tumor types, including challenging ones like pancreatic cancer.
The research involved analyzing blood and tissue samples from 74 patients with pancreatic ductal adenocarcinoma, the most common and aggressive form of pancreatic cancer. Using Molecular Twin, the investigators integrated over 6,363 biological data points, including genetic and molecular information, to create a model predicting disease survival with 87% accuracy. Subsequently, they employed AI to streamline the model to just 589 data points, with proteins in the blood emerging as the best single predictor of pancreatic cancer survival.
The Molecular Twin platform outperformed the FDA-approved pancreatic cancer test, CA 19-9, in both the full and streamlined models, as well as the blood-protein test. The findings were validated in independent datasets from The Cancer Genome Atlas, Massachusetts General Hospital, and Johns Hopkins University. Dr. Arsen Osipov, the lead author of the study, highlights the significance of Molecular Twin in addressing the unmet need for biomarkers in guiding pancreatic cancer treatment.
Jennifer Van Eyk, an expert in the study of proteins and a key member of the Molecular Twin team, emphasizes the role of proteins in predicting patient survival, stating that proteins act as the body’s first responders and play a crucial role in assessing how a patient’s body is reacting to cancer. Dr. Theodorescu notes that while the initial focus is on developing tests for pancreatic cancer, the Molecular Twin platform will continue to expand by incorporating data from additional patients and diverse sources such as medical imaging, gut microbiome samples, tumor microenvironment, and wearable device feedback.
The study’s funding was supported by the National Institutes of Health, the Conquer Cancer Foundation ASCO Career Development Award, and the U.S. Department of Defense Congressionally Directed Medical Research Programs.
The Molecular Twin platform, launched in 2021, is expected to evolve into a robust tool applicable across various cancer types. The rich pool of data collected will contribute to the discovery of biomarkers for additional cancer types and the development of new treatments, potentially identifying at-risk patients before cancer develops to prevent it altogether.
