Next generation sequencing to identify multiple clinically relevant genetic lesions for the diagnosis of acute myeloid leukaemia

Telford, Nicholas (2018) Next generation sequencing to identify multiple clinically relevant genetic lesions for the diagnosis of acute myeloid leukaemia. Doctoral thesis (Degree of Professional Doctorate), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview

Abstract

Routine cytogenetic and molecular genetic investigations aid the diagnosis of acute myeloid leukaemia (AML) and are critical for prognostic stratification to optimise therapy and enhance survival of patients. Advances in the understanding of the genomics of AML by next generation sequencing (NGS) technology have identified a mutational landscape, which has the potential to improve risk assessment and identify new targets for therapy. This project developed a novel, custom-designed NGS panel for the resequencing of genomic DNA (gDNA), to identify multiple types of genetic lesion in AML. Regions of recurrent genetic abnormalities were targeted, to reproduce the output of conventional testing in a single assay. Solution hybrid capture and Illumina-based NGS were used to analyse 36 AML samples, without use of normal control to represent the typical diagnostic workflow. A panel of 42 genes, including those most frequently mutated, was used to test for clinically relevant abnormalities, including common duplications and gene fusions. Sequence data was analysed with a pipeline of relevant bioinformatic tools and the output was compared to standard results and sequencing from an alternative NGS platform. Following variant annotation, a total of 143 likely oncogenic variants were detected across all samples. This included all 13 NPM1 insertions, 10 FLT3-ITD, and the 7 fusion genes found by routine tests. There was strong concordance between NGS platforms for mutation detection. Multiple new findings included two KMT2A-PTD, a TP53 mutation in a patient with a complex karyotype, and a rare NUP98-DDX10 gene fusion. Patients were regrouped by a new prognostic scheme based on genomic features. Eight patients were reclassified; seven changed from the Intermediate group, three to Favourable and four to Adverse. The successful detection of genomic lesions demonstrated the principle that the new NGS assay could reliably detect a variety of genomic abnormalities and that it could be refined for use in the diagnostic laboratory, with the potential to rationalise multidisciplinary workflows. The feasibility of implementation is discussed. A potential clinical utility was inferred and suggests that benefit could be derived for its validation for mainstream diagnosis for the clinical management of AML.