Expression profiles of MUC5AC glycoforms in neoplastic and non-neoplastic pancreatic tissue.

Abstract

e16008 Background: MUC5AC, secretory mucin, exists in two major glycoforms, heavily-glycosylated mature (MM) and less-glycosylated immature (IM) isoforms. Prior IM-based tissue expression studies could not establish the clinical significance of MUC5AC in pancreatic ductal adenocarcinoma (PDA). Despite having strong pre-clinical evidence of MM’s role in local invasion, metastasis, and drug (gemcitabine) resistance, its impact on PDA outcomes is unclear. In this study, for the first time, we examined the differential expression of MM and IM by immunohistochemistry in primary PDA (pPDA), metastatic PDA (mPDA), normal pancreatic tissue adjacent to tumor (NAJ), neuroendocrine tumors (NET), Pancreatitis (P), normal pancreas (N) using tissue microarrays (TMA). Methods: Commercially available TMA (PA2081c, tissuearray.com, Derwood, MD) constructed from 96 patients, including 38 pPDA (14-stage 1, 22-Stage 2, and 2-stage 3), 5 mPDA, 24 NAJ, 11 NET, 8 P and 10 N was used in this study. Two separate 1 mm cores represented each case. Immunohistochemistry for MUC5AC was performed using CHL2 clone (ab77576, abcam, Boston, MA) for the IM isoform and 45M1 clone (ab369, abcam, Boston, MA) for the MM isoform. The immunostains were optimized as per the manufacturer's recommendations using appropriate controls. The sections are evaluated for the intensity of expression (0-3) and the percentage of reactive cells, and an H-score is derived from the product of these measurements. Cases are categorized as having negative/weak (score <10) or positive (score >10) expression. For every positive case, the pattern of expression in terms of Apical (Ap), cytoplasmic (Cy), and extracellular (EC) expression was recorded. Results: MUC5AC (MM and IM) is not expressed in N, NAJ, and P. MM expression is only noted in 44% of PDAs (19/43; 18 pPDA and 1 omental mPDA lesion). Three different expression patterns were noted, including Ap (63%), Cy (84%), and EC (47%). IM expression was detected in 60% of PDAs (26/43, 25 pPDA and omental mPDA lesion) and 18% NET (2/11). IM expression was noted in cases with EC MM expression (9/43, 21%), suggesting that IM was further glycosylated to MM and expressed extracellularly in some patients. MM and MM were co-expressed in 40% of PDA (17/43). Liver and abdomen metastatic lesions in the TMA did not express IM or MM. Pertinent clinical information was not available to study the impact of MM’s expression on PDA outcomes. Conclusions: Our study shows that MUC5AC is primarily expressed in PDAs and can be a useful biomarker. Larger studies are ongoing to correlate tissue expression patterns in PDAs for evidence of any association with early recurrence or treatment failure. Preliminary data suggests that only MM isoform is expressed extracellularly and could be detected in patient serum and serve as a potential non-invasive biomarker for treatment selection (for gemcitabine resistance) and monitor the response in certain cases.