Establishment and Characterisation by Expression Microarray of Patient-Derived Xenograft Panel of Human Pancreatic Adenocarcinoma Patients.

Sandra Roche,Justin Geoghegan,Sinead Toomey,Niall Swan,Michael Moriarty,Ninfa L Straubinger,Robert M Straubinger,Martin Clynes,Kevin Conlon,Justine Meiller,Neil T Conlon,Gerard Mcvey,Robert O’Connor,Rozana Rahman,Fiona O’Neill,Ray Mcdermott,Jean Murphy

doi:10.3390/ijms21030962

Abstract

Pancreatic cancer remains among the most lethal cancers worldwide, with poor early detection rates and poor survival rates. Patient-derived xenograft (PDX) models have increasingly been used in preclinical and clinical research of solid cancers to fulfil unmet need. Fresh tumour samples from human pancreatic adenocarcinoma patients were implanted in severe combined immunodeficiency (SCID) mice. Samples from 78% of treatment-naïve pancreatic ductal adenocarcinoma patients grew as PDX tumours and were confirmed by histopathology. Frozen samples from F1 PDX tumours could be later successfully passaged in SCID mice to F2 PDX tumours. The human origin of the PDX was confirmed using human-specific antibodies; however, the stromal component was replaced by murine cells. Cell lines were successfully developed from three PDX tumours. RNA was extracted from eight PDX tumours and where possible, corresponding primary tumour (T) and adjacent normal tissues (N). mRNA profiles of tumour vs. F1 PDX and normal vs. tumour were compared by Affymetrix microarray analysis. Differential gene expression showed over 5000 genes changed across the N vs. T and T vs. PDX samples. Gene ontology analysis of a subset of genes demonstrated genes upregulated in normal vs. tumour vs. PDX were linked with cell cycle, cycles cell process and mitotic cell cycle. Amongst the mRNA candidates elevated in the PDX and tumour vs. normal were SERPINB5, FERMT1, AGR2, SLC6A14 and TOP2A. These genes have been associated with growth, proliferation, invasion and metastasis in pancreatic cancer previously. Cumulatively, this demonstrates the applicability of PDX models and transcriptomic array to identify genes associated with growth and proliferation of pancreatic cancer.

Highlights

Pancreatic cancer remains a cancer of unmet need, with a 5 year survival rate of approximately 7−9% [1]
This study shows the comprehensive development of patient-derived xenograft models, including detailing the tumours that failed to proliferate as Patient-derived xenograft (PDX) models
The ability of a patient tumour to grow a tumour in vivo was not correlated with any single characteristic, though Pancreatic ductal adenocarcinoma (PDAC) samples, which formed the bulk of the samples received, formed tumours at a greater rate than neuroendocrine tumours (78% compared to 20%)

Summary

Introduction

Pancreatic cancer remains a cancer of unmet need, with a 5 year survival rate of approximately 7−9% [1]. Pancreatic ductal adenocarcinoma (PDAC) has no viable screening method, and an absence of early clinical symptoms, leaving it detected usually later in the disease stage. Even with radical surgical resection, the 5 year survival rates are not comparable with other solid cancers. The limitations of conventional cell lines for pancreatic cancer research has been previously discussed, showing to be a poor predictor of clinical trial outcome [2]. In vitro models are often poor in predicting clinical therapeutic response [3]. Pathology - Extensive associated chronic pancreatitis High-grade PanIN PanIN grade 3. Well-differentiated neuroendocrine tumour (NET) Invasive ductal adenocarcinoma.

Objectives

Methods

Results

Discussion

Conclusion