Long read nanopore DNA sequencing with adaptive sampling to identify tyrosine kinase fusion genes

Diverse haematological neoplasms are driven by tyrosine kinase (TK) fusion genes formed by recurrent or non-recurrent genomic rearrangements. The resulting chimeric proteins often present excellent targets for treatment with kinase inhibitors, and the fusion transcripts or genomic junctions can be used as specific targets for molecular monitoring. Whilst the TK genes involved are generally well characterised (e.g. ABL1, PDGFRA, FGFR1), the fusion partners are very diverse, presenting a challenge for detection and characterisation of these structural variants (SV) using current diagnostic methods. We assessed the ability of targeted nanopore sequencing using adaptive sampling to detect fusion genes in myeloid neoplasms. We sequenced genomic DNA from patients (n = 20) with a known or suspected TK gene fusion and identified rearrangements in 18 cases, including all cases with a known TK fusion, typical and atypical BCR::ABL1 rearrangements, an 843Kb deletion causing a FIP1L1::PDGFRA fusion, novel AGAP2::PDGFRB and NFIA::PDGFRB fusions, and a complex CCDC88C::PDGFRB rearrangement with multiple translocation events. The approach was fast (<72 h/sample from DNA to result), flexible with minimal hands-on laboratory time, and provided accurate, patient-specific characterisation of genomic breakpoints.