It is a hallmark of cancer cells that they acquire the ability to avoid programmed cell death (apoptosis). This apoptotic programme is regulated by the BCL-2 family of proteins, which consists of pro-survival and pro-apoptotic proteins. The balance between these two sub-groups determines whether a cell lives or dies, respectively. One such way that cancer cells can avoid apoptosis is to tip this balance towards survival through overexpression of the pro-survival proteins. A1 (known as BFL-1 in humans) is a pro-survival protein that is mainly expressed in haematopoietic cells, but its physiological function is poorly understood relative to the other BCL-2 family members. Interestingly, A1 overexpression has been noted in several forms of cancer including leukaemia, lymphoma and melanoma. Studies of A1 in vivo have been hampered by the presence of three functional isoforms in mice (A1-a, -b and -d), making gene targeting difficult. However, our lab has successfully generated a novel mouse model that is deficient for all three of these isoforms. Characterisation of these mice has revealed no major defects in the haematopoietic system in the steady state and in various models of infection. Whilst this finding demonstrates redundancy within the BCL-2 family, it also indicates that A1 may be a suitable drug target for cancers that exhibit high levels of A1, as the on-target side effects would be minimal. When generating the A1-deficient mice, we also generated a conditional A1 knockout mouse model that lacks A1-a and A1-d but has A1-b flanked by loxP sites. We have crossed these mice onto a tamoxifen-inducible Cre recombinase-expressing mouse model, enabling temporal deletion of A1-b. We are currently using this mouse model to delete A1 in murine models of AML, T-ALL and melanoma as a proof of concept for the development of an A1-specific inhibitor that could be used as a cancer therapy.