Stratification of Pancreatic Ductal Adenocarcinomas Based on Tumor and Microenvironment Features.

摘要: Background & Aims Genomic studies have revealed subtypes of pancreatic ductal adenocarcinoma (PDA) based on their molecular features, but different reported classification systems. It is a challenge to obtain high-quality, freshly frozen tissue for clinical analysis and determination PDA subtypes. We aimed redefine using large number formalin-fixed paraffin-embedded samples, which are more amenable routine evaluation. Methods collected samples from 309 consecutive patients who underwent surgery September 1996 through December 2010 at 4 academic hospitals in Europe; nontumor were not included. Samples formalin fixed paraffin embedded. DNA RNA isolated; gene expression, targeted sequencing, immunohistochemical analyses performed. used independent component deconvolute normal, tumor, microenvironment transcriptome patterns samples. devised systems an unsupervised consensus clustering approach our data set after removing normal contamination components. associated with overall survival disease-free Cox proportional hazards regression estimation hazard ratios 95% confidence interval. The Cancer Genome Consortium International sets as validation cohorts. Results validated the previously basal-like classical tumor-specific PDAs. identified features PDA, including expression patterns, that allowed tumors be categorized into 5 subtypes, called pure basal like, stroma activated, desmoplastic, classical, immune classical. These cancer cells could by pharmacologic agents. Tumor patient outcomes, sets. also observed exocrine signal acinar cell (from tissue). Conclusions system tumor microenvironment. This might select therapies predict outcomes. found evidence exocrine-like (called ADEX) subtype resulted cells. ArrayExpress accession number: E-MTAB-6134.