Validation of a Dysplasia Grading Assay for Precancerous Pancreatic Cystic Lesions

Abstract public domain (NIH RePORTER)

PROJECT SUMMARY Advances in imaging technology have markedly increased the incidental detection rate of pancreatic cystic lesions (PCLs), fluid-filled sacs that signal an elevated risk for pancreatic cancer. Among the PCL classes, clinical management of intraductal papillary mucinous neoplasms (IPMNs) is regarded as particularly challenging: while IPMNs featuring high-grade (HG) dysplasia are associated with up to a 60% risk of developing highly lethal pancreatic ductal adenocarcinoma (PDAC), those with low-grade (LG) dysplasia are far less likely to progress to cancer and can be safely monitored. However, no robust diagnostic tools currently exist to accurately grade dysplasia, leaving clinicians challenged to decide between continued surveillance or surgical resection – with its inherent morbidity and mortality risks – to prevent potential progression of an untreated IPMN to PDAC. As over 40% of surgeries reveal LG dysplasia upon pathological analysis, a predictive test that can confidently exclude HG dysplasia and prevent avoidable surgeries is urgently needed. Pancreatic cyst fluid is frequently collected for follow-up of imaging findings using a minimally invasive technique known as endoscopic ultrasound-guided fine-needle aspiration. Samples are often acellular, making cytology impractical, and low collection volumes, biochemical inhibition and sample viscosity all complicate subsequent analyses. Amplified Sciences developed BioMatraTM, a low-volume surface-enhanced Raman spectroscopy (SERS) platform in operation in the company’s CLIA-certified lab, to overcome these limitations. The company recently reported a SERS-based 4-marker panel able to distinguish HG from LG dysplasia in IPMN fluid samples with 97% NPV and sensitivity, 89% PPV, and 88% specificity. An additional biomarker, the 3’-sulfo-Lewis A epitope recognized by an ELISA constructed with the Das-1 monoclonal antibody, distinguished high-risk and low-risk mucinous PCLs in a previously published multi-site validation study with 88% sensitivity and 99% specificity. This assay has been converted to a low-volume SERS format. These results represent a notable advancement over existing clinical options. This proposal aims to perform analytical and clinical validation of the 5-marker panel. Qualified reagent materials, SOPs, operational know-how and a Testing Plan will be transferred to a CLIA-certified lab for analytical validation. Clinical validation will be performed on multi-site cohorts of retrospective, pathologically-confirmed IPMNs, with Training and Validation cohort screening executed at Amplified Sciences and Test cohort screening executed under CLIA guidelines. Successful completion of the proposed work will produce: 1) finalized SOPs and analytical validation of the marker panel and 2) clinical validation and operational transfer of the resulting HG dysplasia rule-out test to a CLIA-certified lab. The resulting PanHGLG diagnostic tool will be ready for future clinical utility studies and commercialization.

Public health relevance public domain (NIH RePORTER)

NARRATIVE Pancreatic cancer is usually diagnosed at a late and largely incurable stage, but increasing incidental detection of pancreatic cystic lesions (PCL) through routine imaging has created an important opportunity to identify and pre-emptively treat patients at high risk for progression to cancer. Novel molecular diagnostics are required to determine which PCL patients have a higher likelihood of developing cancer, while minimizing unnecessary surgeries for lower risk patients. This UH2/UH3 proposal seeks to address this unmet need by analytically and clinically validating a multidimensional, low volume assay panel that can accurately differentiate between low-grade and high-grade dysplasia, thereby determining which patients are most likely to progress to pancreatic cancer.