Abstract B013: Interrogating and modeling progression of early pancreatic intraepithelial neoplasia

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive disease with one of the lowest five-year survival rates of all cancers. Low grade, premalignant pancreatic intraepithelial neoplasias (PanINs) are common in the general population, but despite expression of mutant KRAS, the vast majority of these lesions will never progress to cancer. Understanding how additional genetic or epigenetic changes as well as changes to the tumor microenvironment contribute to PanIN to PDAC progression is critical and has the potential to reveal mechanisms of interception for this deadly disease. To better understand how specific alterations to epithelial or microenvironmental cell states impact progression of human PanINs, we are (1) analyzing the composition, organization, and cellular states of early human PanIN lesions and (2) designing manipulable in vitro co-culture models that mimic early human lesions. To understand the cell types, cell states, and spatial organization of PanIN lesions, we have collected >25 organ donor pancreata from people without cancer diagnoses. We histologically assessed multiple regions from the head, body, and tail of each pancreas to select ROIs for further genomic, transcriptomic, and proteomic analysis. We performed spatial transcriptomics on 72 ROIs selected from 13 donors, representing distinct age, BMI, and other lifestyle characteristics. Ongoing spatial transcriptomic analysis is determining the cellular composition and pathway usage among epithelial, stromal, and immune cell types across these different lesions, and these will be compared to the cell type and state changes observed in later stage lesions and pancreatic cancer. In order to build in vitro models testing hypotheses derived from this analysis, we have also created epithelial organoid and fibroblast lines from many of the normal pancreata. Using these primary derived cell lines along with additional purchased primary or immortalized cell lines, we are generating complex 3D bioprinted co-culture models of early disease progression. We are comparing phenotypic, proteomic, and transcriptomic changes to the epithelial cells and the microenvironment following distinct combinations or genetic perturbations of the different cell types. These models will be used to interrogate and understand the crosstalk that can permit progression of pancreatic cancer and to identify and test targets to intercept this process.