Genomic Analysis Yields Four Subtypes of Pancreatic Cancer
An integrated genomic analysis has found that pancreatic cancer can be divided into four specific subtypes based on molecular characteristics. This division could help guide treatment decisions and future research avenues into this difficult malignancy.
An integrated genomic analysis has found that pancreatic cancer can be divided into four specific subtypes based on molecular characteristics. This division could help guide treatment decisions and future research avenues into this difficult malignancy.
“Despite many decades of research into pancreatic cancer we have faced numerous obstacles in finding new and effective treatments,” said co-lead author Andrew Biankin, MBBS, PhD, of the University of Glasgow in Scotland, in a press release. “But our crucial study sheds light on how the chaotic chromosomal rearrangements cause a huge range of genetic faults that are behind the disease and provide opportunities for more personalized pancreatic cancer treatment.”
The new study involved a comprehensive integrated genomic analysis of 456 pancreatic ductal adenocarcinomas, including a combination of whole-genome and deep-exome sequencing. The results of the analysis were published online ahead of print in Nature.
A total of 32 “significantly mutated” genes were uncovered among the 456 samples, which were aggregated into 10 specific pathways. The most common included activating KRAS mutations in 92%; disruption in the G1/S checkpoint pathway that included TP53, CDKN2A, and TP53BP2 mutations in 78%; mutations in the TGF-beta signaling machinery in 47%; and others.
The analysis revealed four subtypes of pancreatic cancer based on their genomic expression. The subtypes included squamous, pancreatic progenitor, immunogenic, and aberrantly differentiated endocrine exocrine (ADEX). The subtypes were associated with specific histologic characteristics. The squamous subtype was associated with adenosquamous carcinomas, with 6 of 25 in the squamous group matching that histology compared with only 1 of 71 in the rest of the evaluable samples (P = .0011). The pancreatic and the progenitor subtypes were associated with mucinous non-cystic adenocarcinomas and with carcinomas arising from intraductal papillary mucinous neoplasm (P = .0005). And the ADEX subtype was associated with the more rare acinar cell carcinoma histology.
With regard to genomic expression, the squamous subtype of pancreatic cancer had a high rate of TP53 and KDM6A mutations, and featured upregulation of the TP63âN transcriptional network. The progenitor subtype, meanwhile, tended to preferentially express certain transcription factors including PDX1, MNX1, FOXA2/3, and others. The authors wrote that the immunogenic subtype shares some of the progenitor subtype’s characteristics, but “is associated with evidence of a significant immune infiltrate.” This included upregulation of B cell signaling pathways, antigen presentation, and CD4+ and CD8+ T cells, among others.
The ADEX subtype was defined by transcriptional factors important later in pancreatic development. This included upregulation of genes involved in KRAS activation, and of both exocrine (NR5A2 and RBPJL) and endocrine (NERUOD1 and NKX2-2) differentiation.
This subdivision of pancreatic cancers could help guide treatment decisions and research directions in the future. For example, the immunogenic subtype is a novel description for this malignancy, and is characterized by mechanisms that immune modulators could target. Also, certain subtypes may benefit more from certain cytotoxic agents.
“Being able to identify which patients would benefit from platinum-based treatments would be a game-changing moment for treating pancreatic cancer, potentially improving survival for a group of patients,” Biankin said.
