Novel Biomarkers of Invasive IPMN

Abstract

Purpose: With the increased utilization of cross-sectional imaging, pancreatic cystic lesions (PCL) are common incidental findings, detected in up to 20% of abdominal MRI scans in adults. While PCL, especially mucinous PCL, have the capacity to develop into invasive carcinoma, it is increasingly being recognized that their malignant potential is neither uniform nor certain. Indeed, while data from long-term cohorts of PCL have identified a small but ongoing risk to the development of carcinoma, this must be balanced against the increasing data demonstrating low yield and high potential morbidity of surgical intervention in elderly patients with non-worrisome PCL. Furthermore, ongoing surveillance of PCL of little or no risk represents a large cost and strain on healthcare systems. Current guidelines that rely on clinical and imaging criteria to stratify risk of malignancy while sensitive, are highly non-specific. Given the morbidity and mortality associated with pancreatic resections, there remains an unmet need for molecular tools to stratify high-risk/malignant from low-risk lesions.

The biomarkers currently in clinical practice are CEA and cytology. Cyst fluid CEA has utility in differentiating mucinous from nonmucinous cysts but has little use in detecting advanced neoplasia in IPMN. Cytology has high specificity in detecting advanced neoplasia, but sensitivity is low due to hypocellular cyst fluid aspirates and high operator dependence. Over the last 20 years, there has been a large number of studies examining various potential biomarkers which are difficult to critically assess due to heterogeneity in definitions of “high-risk” lesions, application of evolving clinical guideline criteria, lack of large pathologically validated cohorts (and the biases associated with these cohorts), and lack of prospective validation. Several DNA-based biomarkers (DNA quantity, loss of heterozygosity, as well as mutations in KRAS, GNAS, TP53, P16, PI3K, PTEN, SMAD4, and BRG1) have been suggested to identify high-risk pancreatic cysts. In addition, promising studies have been reported with microRNAs (miRNA-21, miRNA-155, miRNA-221, miRNA-101, miRNA-216, miRNA-217) and several proteins (MUC proteins, plectin 1, amphiregulin, prostaglandin E2, cytokines (IL5, IL8, and IL1β), mAb Das-1). In addition to cyst fluid analysis, a few studies have examined analyses of blood, circulating cells, and pancreatic juice with preliminary, promising results. We review these studies systematically in the following chapter.

The risk stratification of pancreatic cystic lesions remains a clinical challenge. However, a number of promising biomarkers, DNA, RNA, and protein based, are being developed to aid in segregating high- and low-risk lesions. In conjunction with clinical and imaging parameters and validated in large, prospective cohorts, these biomarkers are poised to improve clinical practice.