Introduction: The development of patient-specific preclinical models to assess therapy response in HNSCC is an ambitious goal ("personalised medicine"). However isolated cells from patient material are affected by isolation/expansion methods and change their properties. This study aims to show whether gene expression profiling can be used to analyse the reliability of the chosen model.

Methods: Cells from primary material of HNSCC patients (n=3) were isolated by outgrowth culture (OC) or enzymatic digestion (ED). Cells were expanded in 2D cell culture up to 2nd passage. 2nd passage (2D) cells were used to form spheroids (25000 cells/spheroid) or to isolate mRNA (Isolate II RNA Mini Kit). The mRNA of spheroids was isolated after 7 days. mRNA was analysed using nCounter technology (Nanostring Technology Inc.), which allows direct analysis of mRNA via fluorescently labeled oligonucleotides (100 bases/oligo) that specifically recognise the mRNA of certain genes - a specific cancer panel (Nanostring Technology Inc.) with 750 genes was used. Data analysis was performed using n-Solver Analysis Software 4.0.

Results: In cells isolated by ED, 12 genes involved in a signaling pathway termed "EGFR tyrosine kinase inhibitor resistance", or 26 genes of the mTOR signaling pathway are less expressed than in cells isolated by OC. In spheroids, the expression of these genes is even lower than in 2D-expanded cells, but the expression is comparable between isolation techniques.

Discussion: In this study, we have shown that the expression of genes involved in drug response is influenced by cell isolation/expansion methods. Different gene expression levels influence the results of preclinical drug testing. Therefore, we propose to characterise the model used by gene expression profiling.

Die Autorinnen/Autoren geben an, dass kein Interessenkonflikt besteht.