Isogenic cell models for development of personalised cancer therapies

Supervisors: Prof M Knowles and Dr Julie Burns (Molecular Biology of Urological Cancers Group, Leeds Institute of Molecular Medicine)

Although the molecular characteristics of many human cancers have been defined, the development of “targeted” or “personalised” therapies remains a major goal. Some targeted agents are available, but a major barrier to their development has been the lack of relevant cell-based models. Ideally, not only should the relevant oncogenic target be present but comparisons should be made with cells that are identical except for the presence of that target. This allows identification of biomarkers of sensitivity and resistance, identification of targets downstream of otherwise “undruggable” targets and selection of optimal cytotoxic drugs for use in combination with targeted agent.

Efficient methods for AAV-mediated somatic recombination allow cancer-specific events to be modelled accurately in relevant target cells, avoiding artefacts generated by high-level ectopic gene expression or gradual re-expression following stable shRNA knockdown. This generates isogenic models that can include knockout of tumour suppressor genes and knock-in of oncogenic mutations. Immortal human cells expressing knock-in mutant oncogenic alleles show “addiction” to the oncogenic stimulus and provide a unique pharmacogenomic platform for the rational design and assessment of personalised therapies.

Several genes that contribute to the development of bladder cancer have been identified, including mutated or over-expressed oncogenes (PIK3CA, FGFR3, FGFR1, RAS genes, EGFR, ERBB2/3) and tumour suppressor genes (TP53, RB, CDKN2A, PTEN, TSC1). Our group studies bladder cancer genomics and we examine the phenotypic consequences of modulation of specific genes in normal urothelial cells. We have implemented AAV-mediated somatic recombination technology to allow the development of faithful disease models. Knock-in of common PIK3CA and FGFR3 mutations and knock-out of one allele of TSC1 in immortalised normal human urothelial cells is underway.

In this project, we will:
1. Engineer cells to contain both FGFR3 and PIK3CA activating mutations for comparison with single mutant cell lines and isogenic controls. Many bladder tumours contain mutations in both genes.
2. Drugs that have not been tested in bladder cancer may be suitable for use alone or in combination with targeted agents. We will screen with a custom compound library comprising approved chemotherapeutic drugs and targeted agents.
3. Depending on the results of these screens, additional screens may be carried out e.g. RNAi or peptide aptamer screens to identify potential resistance mechanisms and biomarkers.
Once initial isogenic recombinants have been made, there will be much scope for the candidate to drive the project according to their interests. For example, more extensive profiling of isogenic pairs of lines may be carried out e.g. transcriptional or proteomic profiling. Responses to drugs in 2-dimensional (2D) culture commonly do not reflect responses in 3D cultures or in vivo and the development of suitable 3D screening platform for these cells may be appropriate. Results obtained in urothelial cells may be applicable to other cancers with the same mutations and the study could test relevant drug combinations in other tissue types.