Synthetic Lethality In CLL With DNA Damage Response Defect By Targeting ATR Pathway

Ataxia Telangiectasia Mutated (ATM) protein coordinates responses to DNA double strand breaks (DSBs) and the ATM-null status caused by biallelic ATM gene inactivation in chronic lymphocytic leukemia (CLL) results in resistance to p53-dependent apoptosis. Accordingly, alternative strategies to target ATM-null CLL are needed.

ATM is a serine/threonine protein kinase that synchronises rapid DNA damage response (DDR) to DNA double strand breaks (DSBs) with activation of cell cycle checkpoints, DNA repair and apoptosis via p53 activation. ATM-null cells are defective in a type of DSB repair that involves homologous recombination and rely on co-operating and compensatory DNA repair pathways for their survival. Therefore, inhibition of DNA repair pathways to which CLL cells with loss of ATM signalling become addicted could provide ‘synthetic lethality’ and induce tumour specific killing. Indeed, we have recently shown that inhibition of a single strand break protein PARP induces differential killing of ATM-null CLL tumours. Here we expand the concept of synthetic lethality in ATM-null CLL and address the question of whether ATM-null deficient CLL cells can be targeted by inhibition of the ATR protein that governs responses to post-replicative damage and co-operates with ATM.

First, we addressed the status of the ATR pathway in primary CLL cells and consistent with previous findings we observed that initiation of cell cycling is required for both ATR upregulation and activation of ATR target Chk1 in response to replicating stress inducing agent hydroxyurea.

We then proceeded with testing viability of the isogenic CLL cell line CII, with and without stable ATM knock down, in the presence or absence of increasing doses of ATR inhibitor AZD6738. We observed a uniform loss of cellular viability in the presence of 1 or 3 μM of inhibitor in ATM-null cells but not in the ATM-wt counterpart. Similar observation was made in primary CLL cells initiated to cycle in the presence of stimulatory oligonucleotide-ODN2006/IL2 support.

To confirm the cytotoxic effect of AZD6738 in vivo we used an ATM null primary CLL xenograft model. Representative primary CLL tumour cells with 15% bialleic ATM inactivation, as assessed by percentage of 11q deletion and allelic frequency of ATM mutation 4220T>C, was engrafted in the presence of activated autologous T lymphocytes into 10 NOG mice. Upon detection of engraftment in peripheral blood, animals were treated by oral administration of either AZD6738 (50mg/kg) or vehicle alone over a 2 week period, and tumour load measured by FACS analysis of CD45+ CD19+ human cells in infiltrated spleens. We observed a reduction in tumour cell numbers in AZD6738-treated compared to vehicle-treated spleens and current investigations are underway to determine whether this difference can be attributed to the selective disappearance of CLL population with biallelic ATM loss.

We suggest that targeting ATR pathway provides an attractive approach for selective killing of ATM-null CLL cells and that this approach should be considered as a future therapeutic strategy for this CLL subtype.