Abstract A078: Predictive Biomarkers for Malignant Progression of Pancreatic Intraductal Papillary Mucinous Neoplasms
Kranthi Kumar Chougoni, Cynthia M. Quintero, Kurtis Liang, Carlos Zuazo, Sourat Darabi, Michael Demeure, David R. Braxton, Robert R. Selby, Joshua Millstein, Matthew Salomon, Daniel D. Von Hoff, Steven R. GrossmanAbstract
Background:
Pancreatic ductal adenocarcinoma (PDAC) is among the most lethal of human cancers and the third leading cause of cancer-related deaths in the US. Pancreatic tumorigenesis is often linked to malignant transformation of mucinous pancreatic cystic lesions (PCL’s), comprised mostly of Intraductal Papillary Mucinous Neoplasms (IPMNs). IPMNs can be anatomically classified as arising from the main pancreatic duct (MD) or side branch ducts (BD). Our unique 20+ year, single-institution retrospective database of 3,236 PCLs from 3,214 patients revealed a striking biological divergence, 57% of resected IPMN-MDs harbored high-grade dysplasia/carcinoma in situ or PDAC at pathology, compared to only 3% of resected IPMN-BDs. Moreover, some presumptive IPMN-BD based on imaging were reclassified as higher risk IPMN-MD at resection. The goal of this study is to identify predictive biomarkers for IPMNs with high malignant potential.
Methodology and Results:
In this study, we deployed the 10X Visium HD spatial transcriptomics (ST) platform to explore the microenvironmental and transcriptomic heterogeneity of high/low grade dysplastic IMPN-MD and IPMN-BD. Using formalin-fixed paraffin-embedded pancreatic resection patient samples histologically verified by a pathologist, we performed ST profiling, and identified differentially regulated gene signatures within cyst epithelial cells, along with unique spatial cellular clusters (epithelial, stromal, and immune types) in the IPMN-MD/BD samples. We then identified differentially expressed genes (DEGs) in IPMN epithelial cells that encode for secreted proteins, to identify potential cyst fluid-measurable biomarkers. Two clinically established genes emerged as reciprocal markers between IPMN-MD and IPMN-BD ductal epithelial cells: CEA (CEACAM5) marked IPMN-MD epithelium, and MUC6 marked IPMN-BD epithelium, as both (high/low grade) IPMN-MD samples were at least 97% CEA-dominant, and both IPMN-BD samples were at least 90% MUC6-dominant per bin. Notably, the within-bin ratio log2((CEA+1)/(MUC6+1)) separates duct type almost completely and independently of grade. To preliminarily validate the value of this biomarker ratio, we determined log2((CEA+1)/(MUC6+1)) in epithelial cells from a publicly deposited single cell RNAseq analysis of IPMN-BD vs. IPMN-MD (GSE316796), noting near complete separation of the CEA/MUC6 ratios between IPMN-MD and IPMN-BD samples.
Conclusions:
Our preliminary findings revealed IPMN ductal epithelial divergent gene expression programs and a preliminary secreted biomarker signature which can distinguish low (IPMN-BD) vs. high (IPMN-MD) risk IPMN subtypes on a molecular rather than clinical imaging basis, potentially improving risk stratification of the rapidly growing IPMN patient population.
Citation Format:
Kranthi Kumar Chougoni, Cynthia M. Quintero, Kurtis Liang, Carlos Zuazo, Sourat Darabi, Michael Demeure, David R. Braxton, Robert R. Selby, Joshua Millstein, Matthew Salomon, Daniel D. Von Hoff, Steven R. Grossman. Predictive Biomarkers for Malignant Progression of Pancreatic Intraductal Papillary Mucinous Neoplasms [abstract]. In: Proceedings of the AACR Conference on Pancreatic Cancer: New Frontiers in Biology and Therapeutic Development; 2026 Sep 25-28; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(18_Suppl_2):Abstract nr A078.