Pancreatic ductal adenocarcinoma (PDA) has been a challenging disease, since many randomized controlled trials have failed to demonstrate improved survival with several agents considered effective for other malignancies. In parallel, studies in other neoplasms have shown that heterogeneity in survival is determined, in part, by molecular differences between tumours. As recently demonstrated in breast and lung cancers, treatment outcomes have been improved by targeting drugs to tumour subtypes in which they are selectively effective.
The introduction of genome-wide screening techniques has expanded the numbers of genes linked to pancreatic cancer in general. However, until now, identification of PDA molecular subtypes and its correlation with clinical patterns has been non-integrative.
In this line, Collisson and colleagues recently published interesting data regarding transcriptional profiling of primary PDA samples from two sources (Department of Pathology, University of California – San Francisco, and Bioinformatics Group, National Institute for Research in Informatics – Bucharest, series GSE5471), along with human and mouse PDA cell lines.1 In order to identify subtypes of the disease, microdissected PDA samples from UCSF tumours (n = 27), for which clinical information was available, were merged with previously published GSE15471 tumour tissue samples (n = 36).
A 62-gene signature from merged data sets was developed (PDAssigner). In addition, a non-negative matrix factorization (NMF) analysis supported up to three subtypes of PDA, defined as classical, quasimesenchymal (QM-PDA) and exocrine-like, based on subtype-specific gene expression. The classical subtype had high expression of adhesion-associated and epithelial genes, and the QM-PDA sub¬type showed high expression of mesenchyme-associated genes. The exocrine-like subtype showed relatively high expression of tumour cell–derived digestive enzyme genes, with immunohistochemical staining supporting this observation.
Investigators also found that stratification by PDA transcriptional subtype provided useful prognostic information. Individuals with classical subtype tumours fared better than individuals with QM-PDA subtype tumours after resection. Interestingly, histological grade was also associated with subtype. In a multivariate Cox regression model, including stage and subtype, subtype was an independent predictor of overall survival.
Using PDAssigner, investigators also demonstrated that classical and QM-PDA subtypes were represented in a collection of 19 human and 15 mouse PDA cell lines. Interestingly, human PDA cell lines treated with gemcitabine and erlotinib had subtype-specific responses. QM-PDA subtype lines were, on average, more sensitive to gemcitabine than the classical subtype. Conversely, erlotinib was more effective in classical subtype cell lines.
These results further establish phenotypic differences between the classical and QM-PDA subtypes and suggest that these and perhaps additional drugs will show subtype specificity in PDA, a distinction that could be exploited in clinical trial sensitivity enrichment schemes. More immediately, these results indicate that gemcitabine and erlotinib are preferentially active in different PDA subtypes, so that the current practice of combining them may increase toxicity without increasing efficacy for many patients.
In conclusion, this translational study supports the use of gene-expression subtypes as an independent prognostic indicator in resected PDA. Possibly, PDAssigner subtypes may have utility in preclinical drug development models and in allocation of patients for clinical trials, being useful to identify new subtype-specific therapies.
1. Collisson EA, Sadanandam A, Olson P, et al. Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy. Nat Medicine, 17(4):500-3, 2011.