Poster Presentation 39th Annual Lorne Genome Conference 2018

Transcriptional Consequences of Cancer Fusion Genes (#166)

Erin E Heyer 1 , Ira W Deveson 1 2 , Timothy R Mercer 1 3 4 , James Blackburn 1 3
  1. Garvan Institute of Medical Research, Darlinghurst, NSW, Australia
  2. School of Biotechnology and Biomolecular Sciences, Faculty of Science, University of New South Wales, Randwick, NSW, Australia
  3. St. Vincent’s Clinical School, Faculty of Medicine, University of New South Wales, Randwick, NSW, Australia
  4. Altius Institute for Biomedical Sciences, Seattle, WA, USA

Chromosomal translocations join together two previously distinct parts of the genome. This genomic shuffling can alter expression levels of canonical genes or join two separate genes to encode a single transcript of novel function. These fusion genes account for approximately 20% of human cancer morbidity and are often tumour-specific and drug-targetable, so precise fusion identification can significantly influence disease treatment. To that end, we have developed Blood FuSeq and Solid FuSeq – two diagnostic tests utilising targeted RNA capture sequencing (RNA CaptureSeq) to identify fusion genes in haematological malignancies and solid tumours, respectively.

During initial validation using cell lines harbouring fusion genes, we precisely identified all known fusions and established the sensitivity of the assays at 1:10,000 cells. Expanding our analysis to patient samples, we successfully diagnosed both known and previously overlooked fusion genes, including those for which approved therapeutic treatments already exist. For example, we used the Blood FuSeq panel to interrogate 29 patient samples previously lacking a molecular diagnosis, and in doing so identified fusion genes in 16 of these patients.

Beyond identifying the genes involved in each fusion rearrangement, our data revealed informative variations in exon usage, expression levels and isoform diversity – all factors that can affect treatment efficacy. In some patients, we detected either multiple fusion gene isoforms or novel intragenic deletions undetected by other diagnostic approaches. Given the high coverage data generated by RNA CaptureSeq, we also expanded our analysis to search for novel transcriptomic elements unique to these cancer genomes.

Overall, we believe these FuSeq panels can be developed into clinical diagnostic tools, delivering medically relevant information on a wide range of fusion genes. In addition, the resulting expression level and exon usage information for general cancer-associated genes can be used to infer further therapeutic treatments, all within a single diagnostic test.