Using the zebrafish as a pre-clinical animal tumour model for drug discovery and the evaluation of anti-leukemia agents

T-cell acute lymphoblastic leukemia (T-ALL) is a high risk subtype of the most common leukemia affecting children. Although intensive chemotherapy treatments have improved survival in this disease, a significant number of patients do not respond to therapy or relapse. For these patients, a better understanding of the molecular pathways underlying the development of this disease holds the promise of personalized therapy with improved outcomes and fewer side effects. Two common pathways affected in T-ALL are NOTCH and PTEN/PI3K. Drugs targeting NOTCH signalling, called gamma secretase inhibitors (GSIs), have been effective against subtypes of T-ALL with NOTCH mutations. However, GSIs have had limited success in clinical trials due to side-effects including gut toxicity, and a significant proportion of leukemias are insensitive to GSIs. Recently drugs targeting the PTEN/PI3K pathway have shown promise for the treatment of T-ALL in both cell culture and mouse models of T-ALL. T-ALL cell lines are typically used for drug screening but they do not accurately represent the complexity of interactions between the leukemia cells and the human body. Transplanting human leukemia cells into mice may better represent human disease but is a costly and slow method of screening new anti-cancer compounds or to evaluate the drug sensitivity of an individual patient's cancer to a particular therapy. We have developed an innovative approach to transplant human cancer cells into zebrafish embryos, which can then be used for rapid pre-clinical drug testing for anti-leukemia agents. We propose a two-pronged approach testing new drugs developed by GlaxoSmithKline that block PI3K signalingin human T-ALL cell lines and primary patient samples, using both zebrafish and mice. The results of this study will provide validation of the zebrafish model as a cost-effective and rapid means of identifying new agents for the treatment of T-ALL, as well as demonstrating the potential of this model system to personalize cancer therapy.