A comparison of high resolution melting, allele-specific priming and sanger sequencing for the detection of BRAFV600E mutation in hairy cell leukaemia from different haematological specimens

Abstract

Introduction: Recently, the BRAFV600E mutation was discovered as a highly specific and sensitive marker for hairy cell leukaemia (HCL) and a potential target for drug therapy. However, the quality of clinical specimens is often suboptimal for genotypic analysis due to a small number of hairy cells in blood and bone marrow aspirate. This study aims to assess the usefulness of three molecular techniques for the detection of theBRAFV600E mutation in HCL from different types of haematological samples. Methods: We assessed the performance of high resolution melting (HRM), allele-specific priming (ASP) and Sanger sequencing (SS) for BRAFV600E detection in 17 unenriched HCL samples: blood (n=7), marrow aspirate (n=3), ethylenediaminetetraacetic acid (EDTA)-decalcified trephine biopsy (n=2), formic acid (FA)- decalcified trephine biopsy (n=5).

Results: In peripheral blood and marrow aspirate samples where DNA was well preserved, all three molecular techniques showed analysable results in all 10 HCL except for one sample in which HRM result was not analysable. In formalin-fixed, paraffinembedded and decalcified trephine biopsies, DNA preservation was less optimal. High resolution melting analysis failed to produce an analysable result in any of the seven trephine biopsies tested. The poor performance of HRM on trephine biopsies was not related to the method of decalcification, the length of storage or the extent of involvement. Allele-specific priming and SS demonstrated a better analytical ability for the mutation. The results showed that for blood and marrow aspirate, both HRM and ASP had a very high analytical sensitivity (1% tumour load), diagnostic sensitivity (100%) and specificity (100%) in analysable samples. Sanger sequencing had a lower analytical sensitivity (4% tumour load), resulting in false-negative analysis in some cases. High resolution melting was technically the simplest and had the shortest turn-around-time of 2 hours. In decalcified trephine biopsies, HRM was not useful, while SS was least demanding on sample DNA quality for a successful analysis. However, none of the three techniques showed satisfactory diagnostic performance for trephine biopsies. Conclusions: High resolution melting is a cost-effective technique for initial screening of the BRAFV600E mutation in blood and marrow aspirate samples of suspected HCL patients. Atypical cases can be confirmed by ASP or SS. A robust detection method for this mutation on decalcified trephine biopsies still awaits development and validation.