Project Survival: Engineering a phenomic and artificial intelligence driven precision medicine biomarker pipeline for pancreatic adenocarcinomas

Pancreatic cancer is a complex and dynamic disorder necessitating a comprehensive clinical design integrated with robust OMICS technologies and AI analytics to identify potential molecular and clinical signatures of diagnosis, progression, and treatment outcomes. Project Survival is a multisite prospective longitudinal study currently in the 4th year of a 6 year initiative of sampling and analysis of subjects in 6 categories: healthy volunteers with a first degree relative with pancreatic cancer (N=39), pancreatitis (N=34), pancreatic cystic neoplasm (N=52), suspicious pancreatic masses with pathology other than pancreatic cancer (N=22), early stage (N=66), locally advanced (N=123), and metastatic pancreatic cancer (N=99). All diseased patients are longitudinally sampled multiple times per year for sera, plasma, buffy coat, saliva, urine, and tumor/adjacent normal tissue. The BERG Interrogative Biology® platform is employed for multi-omic mass spectrometry analysis (metabolomics, lipidomics and proteomics) and applies artificial intelligence (bAIcis®, BERG Artificial Intelligence Clinical Information System) technologies. bAIcis® is harnessed to align the multi-omic profiles with longitudinal clinical information to infer probabilistic cause-and-effect relationships among molecular and clinical variables in a network-based model. Multiple longitudinal time points continue to be collected during the course of the six-year timeline enabling dynamic modeling. The value of this longitudinal study is in the epidemiological assessment of patient type progression to more advanced stages and identification of biomarkers and clinical features that align with the shifts observed in the patient populations. Collectively, we are incorporating patient progression with longitudinal sampling to investigate predictive signatures of disease advancement. Biomarker panels with AUC > 0.7 will be pursued in a further prospective clinical study with a larger subject number. The integration of multi-omic analysis with artificial intelligence has identified several biomarker panels that meet numerous unmet needs for the identification and clinical stratification of pancreatic adenocarcinoma. Citation Format: Eric Michael Grund, Michael A. Kiebish, Viatcheslav R. Akmaev, Rangaprasad Sarangarajan, John J. Crowley, Amy Stoll-D9Astice, Tori Singer, Corinne Decicco, Wendy Hori, Abena Darkwah, Lixia Zhang, Valerie Bussberg, Leonardo O. Rodrigues, Emily Y. Chen, Tomislav Dragovich, Manuel Hidalgo, Niven R. Narain, A James Moser. Project Survival: Engineering a phenomic and artificial intelligence driven precision medicine biomarker pipeline for pancreatic adenocarcinomas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4945.