Cancer Genomics

Lead Research Organisation: MRC London Institute of Medical Sciences


Our work focuses on the study of non-Hodgkin lymphoma which in the UK is the sixth most common form of cancer and responsible for over 4,000 deaths annually. Cancer arises when normal cells accumulate mutations that drive uncontrollable proliferation. Tumour cells can harbour thousands of these mutations, but only a fraction of them cause cancer. The goal of our studies is to identify which of these mutations contribute to cancer. By creating mutations in mouse models of lymphoma we can study which mutations contribute to the development of a normal white blood cell into lymphoma cells. Similarly by creating mutations in human lymphoma derived cell lines and exposing them to chemotherapeutic agents we can identify which mutations give rise to drug resistance. By identifying the genes that contribute to lymphoma development we can identify new drug targets that can be used to develop novel therapies for patients whose disease is currently incurable.

Technical Summary

Sequencing of cancer genomes has identified large numbers of recurrent mutations from various tumour types, however the background of non oncogenic “passenger” mutations and allelic variation in these studies can obscure the role that more causal events are playing. Another approach to identifying genes that are mutated in cancer is the use of insertional mutagenesis screens in mice. When newborn mice are infected with slow transforming retroviruses, random insertion of proviruses into the genome of infected host cells leads to deregulation of nearby genes. A proportion of these insertion mutations provide a clonal growth and/or survival advantage to infected cells through activation of oncogenes or inactivation of tumour suppressors, and these cells give rise to lymphoid tumours at three to four months of age. The Sleeping Beauty and piggyBac transposons have been adapted for similar use as oncogenic insertional mutagens to perform screens in selected tissues that are not amenable to infection by slow transforming retroviruses. By cloning the insertion sites from these lymphomas it is possible to identify genes that contribute to malignancy. Loci that are recurrently identified in these screens (referred to as Common Integration Sites or Common Insertion Sites (CISs)) show a significant overlap with the set of known human oncogenes and tumour suppressors, validating the use of this screening approach. These screens create well-controlled panels of tumours in which the causal oncogenic mutations are easily defined by cloning and sequencing of the insertion sites. The screens are a useful complement to human tumour analyses, in part because they provide validation for the oncogenicity of human cancer genes in an independent system using a different mutagen, but also because they yield a thorough coverage of the set of mutations present in each tumour at a cost which is lower by orders of magnitude than the study of human tumours. Our recent screens have identified hundreds of genes mutated in mouse lymphoma and identified hundreds of significant genetic interactions between the somatic insertion mutations and between somatic mutations and the predisposing oncogenic germline alleles present in the mice. Our future screens will focus upon mouse models of follicular lymphoma and diffuse large B cell lymphoma. Mutations in these screens will be validated and investigated further using a panel of human B lymphoma cell lines and by the creation of new mouse models to test the requirement of these genes for tumour maintenance and their potential as therapeutic targets.


10 25 50
Title Insertion mutation cloning method 
Description We developed a novel methodology for identifying insertion mutations. The clonality of mutations is quantitative over two orders of magnitude. Libraries can be prepared in a 96 well format suitable for automation. Hundreds of libraries can be pooled, sequenced on a single Illumina HiSeq flow cell and demultiplexed. This degree of multiplexing yields the insertion mutation profile of hundreds of tumours for a reagent cost of £12 per sample. 
Type Of Material Technology assay or reagent 
Provided To Others? No  
Impact We have used this methodology to sequence insertion mutations from hundreds of mouse lymphomas giving us a cohort of 6000 clonal mutations and 1.2 million subclonal mutations. 
Title MuLV Database 
Description We have sequenced insertion mutations from mice with a spectrum of lymphoid malignancies including BCL2 driven B cell lymphomas (akin to follicular lymphoma and diffuse large B cell lymphoma) and T cell acute lymphoblastic leukaemia. From 800 mice we identified approximately 6000 clonal mutations and an additional 1.2 million subclonal mutations. By following a time course of subclonal mutation frequencies across the genome we identified hundreds of loci that are significantly associated with mouse lymphoid malignancies. The database is interactive such that any subset of tissues or mutations can be selected and the frequencies of mutations observed in each cohort. There is also a function to identify cooperating events between mutations, and associations between mutations and tumour phenotypes including detailed flow cytometric profiling. 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Impact Once published the database will be openly available to all researcher. The majority of all previous citations of insertional mutagenesis screen papers have come from researchers who searched these datasets for mutations in a specific gene of interest. 
Title EZH2 and EHMT2 histone methyltransferase inhibitor 
Description We used a panel of human non Hodgkin lymphoma cell lines to test a novel experimental dual inhibitor of histone methyltrasferases EZH2 and EHMT2, developed by the laboratories of Matthew Fuchter and Bob Brown. 
Type Therapeutic Intervention - Drug
Current Stage Of Development Refinement. Non-clinical
Year Development Stage Completed 2015
Development Status Under active development/distribution
Impact This molecule is the first dual inhibitor of EZH2/EHMT2 active in cells and demonstrates the benefit of this dual inhibition in treatment of cancer cells. 
Title GSK J4 
Description GSK J4 and GSK J1 are inhibitors of the demthylases JMJD3 & UTX sourced from the Structural Genomics Consortium. We have screened a panel of non-Hodgkin lymphoma cell lines and found a subset of these are sensitive to GSK J4. A dose escalation study in mice found the compound well tolerated at doses of 200mg/kg IP. We have found inhibition of the growth of subcutaneous NHL xenografts at a daily dose of 25mg/kg IP. 
Type Therapeutic Intervention - Drug
Current Stage Of Development Initial development
Year Development Stage Completed 2013
Development Status On hold
Impact The product has demonstrated some efficacy in a preclinical mouse model of NHL. 
Title Scripts for identifying integration mutations. 
Description Scripts used to identify retroviral sequences from a ligation mediated PCR protocol developed for the Illumina sequencing platform. 
Type Of Technology Software 
Year Produced 2017 
Impact These scripts were used in the following publication.