Identification and characterisation of non-coding driver mutations in chronic lymphocytic leukaemia

The identification of genes that are recurrently mutated in haematological cancers has directly led to novel therapies, biomarkers of tumour behaviour, improved cancer classification and better fundamental understanding of disease and its response to therapy. Large-scale genomics projects have catalogued the somatic landscape of chronic lymphocytic leukaemia (CLL). However, these have focused almost exclusively on coding mutations and despite high-profile, genuine successes, the number of new driver genes identified has been limited. Importantly, significant sub-populations of CLL have no observable “driver” mutation and we still do not know the full complement of molecular lesions that are individually necessary – and together sufficient – to cause malignancy.

The protein-coding component of the genome accounts for <2% of the total sequence. By contrast, The Encyclopedia of DNA Elements (ENCODE) project indicates that ~80% of the genome contains biochemically functional elements. Although there is very little information on how non-coding somatic variation affects oncogenesis, a few examples exist that establish the principle of its importance. TERT mutations are one of the best described examples, occurring at promoter hotspots and leading to TERT over-expression. Furthermore, parallels exist in Mendelian cancers where identical phenotypes can be caused by inactivating mutations in coding regions or by changes that affect transcriptional control (e.g. POT1), or by mutations that are solely non-coding A shared pathogenic mechanism is the misregulation of enhancers. For example, mutations can affect chromatin folding and hence the structure of topologically associating domains (TADs), leading to tumour type-specific super-enhancers and a 'locked' gene regulatory state driving uncontrolled proliferation. The maturation of high-throughput technologies and the availability of whole-genome sequencing (WGS) data from CLL provides an opportunity to both identify key mutations in non-protein coding genomic regions and decipher their downstream consequences. This offers the prospect of identifying novel therapeutic and chemoprevention agents and classifying patients into molecular subgroups to personalise therapy.

Objectives:
The project is aimed at identifying non-protein coding cancer driver mutations for CLL and deciphering their functional consequences. We shall identify cell-type specific oncogenic pathways that contribute to tumorigenesis and investigate the underlying molecular mechanisms. A key premise is that a substantial fraction of such elements act in cis by regulating the expression of nearby genes. The project will employ multiple experimental and computational approaches to address the study aims.