Hypoxia Activated Prodrugs of DNA-PK Inhibitors as Radiosensitisers

Abstract

Radiation therapy is a widely used treatment for many different cancer types. However, efficacy is often limited by tumour regrowth due to radioresistant hypoxic tumour cells and DNA DSB repair, and also by normal tissue toxicity. DNA-dependent Protein Kinase (DNA-PK), is a key repair protein in the Non Homologous End Joining (NHEJ) pathway which is responsible for most of the repair of radiation-induced DSBs. Inhibition of DNA-PK has been identified as a useful way to radiosensitise tumour cells, but concerns about sensitisation of normal tissues within the radiation field have been raised. One approach to selectively target DNA-PK inhibitors to tumour tissue is to use Hypoxia Activated Prodrugs (HAP). The HAP is designed to be activated under hypoxic conditions, therefore providing tumour-selective delivery of the DNA-PK inhibitor, and enhance radiation therapy. This thesis investigates the in vitro behaviour of a newly developed inhibitor of DNA-PK, SN39536, and a corresponding HAP, SN39884. In Chapter 3 we investigated the metabolic activation of SN39884 and its ability to release the effector, SN39536, under oxic and anoxic conditions. We explored the dependence of metabolic activation upon a common one-electron reductase, cytochrome P450 oxidoreductase (POR), using three HCT116 human colorectal carcinoma variants that either, did not express POR (PORko-R), overexpressed POR (POR-G), or constitutively expressed POR (HCT116-G). We investigated cellular uptake of SN39536 and SN39884 by analysing both intracellular and extracellular concentrations under oxic and anoxic conditions. SN39884 was found to be stable under oxic conditions and released the DNA-PK inhibitor, SN39536, selectively under anoxic conditions in all three cell lines. Concentrations of SN39536 were 2-3 fold higher in POR-G cells than PORko-R and HCT116-G, indicating that POR contributed to SN39884 metabolism. However, both PORko-R and HCT116-G metabolised SN39884 to a similar extent, indicating that other reductases contributed to SN39884 metabolism under anoxia. Intracellular concentrations were ~10 fold higher than extracellular concentrations for both prodrug, SN39884, and effector, SN39536, under oxic and anoxic conditions. This suggested the extravascular diffusion of the compounds would not be limited by avid cellular uptake. In Chapter 4 we used regrowth inhibition assays to investigate the growth of cells after exposure to SN39884 and SN39535, in combination with radiation, to evaluate radiosensitisation of the three HCT116 POR variants. The effector SN39536 alone, had little effect on cell growth, except at the highest concentration (10 μM), but demonstrated effective radiosensitisation under oxic conditions in all three cell lines. Prodrug SN39884 was stable under oxic conditions, but at the highest concentration (10 μM), demonstrated some toxicity in cells exposed to radiation and treated with drug alone. Under anoxia, SN39884 demonstrated radiosensitisation of all three cell lines, with POR expression not differentiated in this assay. In Chapter 5 we used clonogenic survival assays to evaluate radiosensitisation of the HCT116 POR variant cells by SN39536 and SN39884 under oxic and anoxic conditions. We determined Sensitiser Enhancement Ratios for 10% survival (SER10) values for comparison. The effector SN39536 radiosensitised all three cell lines at 1 μM (SER10 2.19 – 2.29). SN39884 (1 μM) did not result in radiosensitisation in all cell lines under oxic conditions (SER10 1.01 – 1.12). However, under anoxia, SN39884 radiosensitised all cell lines, but gave a higher SER10 for high POR expressing POR-G cells (SER10 1.55) compared to PORko-R and HCT-116-G cells (SER10 1.34 and 1.36, respectively). The clonogenic survival assay provided a more accurate reflection of the metabolic activation of the prodrug under anoxia and could discriminate the activity of the prodrug more precisely than the growth inhibition assay. Overall, we were able to demonstrate that HAP, SN39884, was stable under oxic conditions, and the corresponding DNA-PK inhibitor, SN39536, was able to robustly radiosensitise the cell lines, regardless of POR expression. SN39884 was shown to be able to release SN39536 under anoxic conditions and this produced radiosensitisation, especially in POR overexpressing cells. Further research into these DNA-PK inhibitors is warranted, with the next steps involving 3-D cellular models to evaluate (pro)drug metabolism and diffusion, and in vivo work to further investigate efficacy and tumour selectivity.