ABSTRACT: Typically diagnosed late, when systemic metastasis has already occurred, pancreatic ductal adenocarcinoma (PDAC) has one of the worst 5-year survival rates of any cancer type. For many patients with advanced disease, current chemotherapy regimens offer only modest benefit despite significant toxicity and surgical resection, the only treatment option with curative potential, is not possible. Therefore, while new treatments are much needed, diagnosing patients at an ‘earlier’ disease stage when surgery remains possible and the window of opportunity for treatment response is longer will be critical to improving patient outcomes. In this regard, the identification of biomarkers from biospecimens that can be easily sampled from patients remains the focus of considerable research, however success has not been forthcoming. Using a suite of novel genetically defined murine isogenic models of early PDAC, engineered using CRISPR-Cas9 gene editing, we sought to address whether loss-of-function mutations in common driver genes, and thus the genetic heterogeneity inherent to the disease, may represent an important confounding factor in the identification of a one-size-fits-all biomarker suitable for early detection. Focussing on the multi-omics analysis of blood, we show that both loss of Cdkn2a and / or Smad4 on the background of a KrasG12D Trp53-/- genotype has profound effects on the profile of differentially expressed RNA species including protein coding RNAs, lncRNAs, snoRNAs, scRNAs, snRNAs and miRNAs, and on plasma protein expression, when compared to both healthy controls and chemically induced pancreatitis. In addition, we find that loss of Smad4, a genomic event that occurs following progression from PanIN to PDAC, substantially limits the availability of blood biomarkers. These findings identify the need to move towards genotype-specific biomarker signatures and uncover a potential role for Smad4 loss in limiting opportunities for the early detection of pancreatic cancer.
