Poster Presentation 39th Annual Lorne Genome Conference 2018

Elucidating the molecular signatures underlying ALT activity in tumours (#177)

Michael Lee 1 , Erdahl T Teber 2 , Oliver Holmes 3 , Katia Nones 4 , Ann-Marie Patch 4 , Rebecca A Dagg 5 , Loretta MS Lau 5 , Joyce H Lee 5 , Christine E Napier 5 , Jonathan W Arthur 2 , Sean M Grimmond 6 , Nicholas K Hayward 7 , Peter A Johansson 7 , Graham J Mann 8 , Richard A Scolyer 9 , James S Wilmott 9 , Roger R Reddel 5 , John V Pearson 4 , Nicola Waddell 3 , Hilda A Pickett 10
  1. Children's Medical Research Institute, Westmead, NSW, Australia
  2. Bioinformatics Unit, Children's Medical Research Institute, University of Sydney, Westmead, NSW, Australia
  3. Genome Informatics Group, QIMR Berghofer Medical Research Institute, Herston, Queensland, Australia
  4. Medical Genomics Group, QIMR Berghofer Medical Research Institute, Herston, Queensland, Australia
  5. Cancer Research Unit, Children’s Medical Research Institute, University of Sydney, Westmead, NSW, Australia
  6. Centre for Cancer Research, University of Melbourne, Melbourne, Victoria, Australia
  7. Oncogenomics Group, QIMR Berghofer Medical Research Institute, Brisbane, QLD, Australia
  8. Centre for Cancer Research, Westmead Institute for Medical Research, Westmead, NSW, Australia
  9. Discipline of Pathology, Sydney Medical School, The University of Sydney, Sydney, NSW, Australia
  10. Telomere Length Regulation Unit, Children's Medical Research Institute, University of Sydney, Westmead, NSW, Australia

The replicative immortality of human cancer cells is achieved by activation of a telomere maintenance mechanism (TMM). To achieve this, cancer cells utilise either the enzyme telomerase, or the Alternative Lengthening of Telomeres (ALT) pathway. These distinct molecular pathways are incompletely understood with respect to activation and propagation, as well as their associations with clinical outcomes. We have developed a pipeline capable of accurately quantitating telomere variant repeats from whole genome sequencing (WGS) data. We validated TMM in two tumour sample sets comprising of 86 pancreatic neuroendocrine tumours and 81 melanomas. We then used our telomere variant repeat pipeline to generate a random forest classifier, which was capable of predicting TMM with 91.6% accuracy using telomere length and sequence content alone. This classifier was then applied to 908 WGS datasets across 22 tumour types, for which information regarding the activated TMM was not previously known. We found that aberrations in TMM associated genes, such as loss-of-function mutations in ATRX/DAXX and TERT promoter mutations, were not common across all tumour types nor mutually exclusive to one TMM. Analysis of pathways that were correlated with the activation of ALT identified autophagy and cell cycle control of chromosomal replication pathways as important for the survival of ALT tumours. Our analysis also revealed extensive genetic heterogeneity between tumour types, highlighting the complexities involved in wide scale genome studies of TMMs. Our approach has the potential to be further developed with the addition of more TMM validated datasets and can be applied to larger tumour datasets in order to elucidate the tumour type specific mechanisms of ALT activation.