Abstract SY16-01: Optimizing BH3 mimetic therapy: Insights from BCL2 inhibition in the clinic

Abstract

Abstract For many cancers, evasion of apoptosis is an important component in their pathogenesis. This was first appreciated for lymphoid haematological malignancies, based on the discovery of the recurring t(14;18)(q32;q21) translocation in follicular lymphomas between the IGH and BCL2 genes, and the subsequent elucidation of the function of the previously unknown BCL2 protein. Those seminal scientific breakthroughs in the 1980s generated enormous interest in the field of cell death regulation. We now have a detailed understanding of how normal cells regulate their survival, balancing extra- and intra-cellular signals that either promote cell survival or the triggering of the mitochondrial path to apoptotic death. The BCL2 family of proteins includes both promoters and suppressors of apoptosis, and direct interactions between family members govern whether cells evade or undergo apoptosis. The perturbed physiology of cancer cells can create many different pro-apoptotic stresses that have to be overcome for clinical tumorogenesis. This paradigm is perhaps best understood in hematological malignancies, but can be considered more broadly applicable across many solid tumors as well. BCL2, MCL1 and BCLxL are the most studied pro-survival BCL2 family members. BIM, BAD, PUMA, NOXA and BID are prominent members of the BH3-only subfamily of BCL2-like proteins, and these all function in an apoptosis-promoting way, and so can be considered the physiological antagonists of BCL2, MCL1 and BCLxL. Despite recognition in the late 1980s of the potential for suppression of BCL2 function as an anti-cancer strategy, realization of this potential had to wait until the development of so-called “BH3 mimetics” more than a decade later. BH3 mimetics are small molecules which bind BCL2 or related prosurvival proteins and antagonise their function. They exhibit cytotoxicity solely by triggering the mitochondrial path to apoptosis. Early putative BH3 mimetics that lacked potency or had broader mechansims of action failed in clinical trials. Potent BH3 mimetics that target BCL2 (venetoclax), or BCL2 and BCLxL (navitoclax) have demonstrated significant single agent clinical activity against chronic lymphocytic leukemia. Venetoclax has also demonstrated moderate clinical activity against mantle cell lymphoma, follicular lymphoma and Waldenstom's Macroglobulinemia, and some activity against myeloma and diffuse large B cell lymphoma. In reviewing these recently published and unpublished clinical trial data as well as a rich vein of preclinical data, this presentation will aim to make the case that several conclusions can currently be drawn. Firstly, that high level expression of the target (ie BCL2 or BCLxL) is necessary, but not sufficient for identification of cancers that will be susceptible in vivo. While the evidence is strong that it is the balance among BCL2 family members that determines whether a cancer cell is susceptible to BH3 mimetic therapy, simple profiling of BCL2 family expression has not proved to be useful as a screening tool. Functional assays are likely to be required for us to be able to prospectively identify tumor types most amenable to therapy with newer BH3 mimetics that target only BCLxL or MCL1. Secondly, that other than BCL2-selective inhibitors for CLL, BH3 mimetics are unlikely to be sufficiently clinically active as single agents. While preclinical data can identify primary cancer cells (and normal cells) that are dependent upon either BCL2 or BCLxL or MCL1 for survival, in most circumstances this dependence is context dependent and requires additional stressors to be revealed. Importantly, this means that BH3 mimetics are highly likely to be attractive partners in combination therapies. Clinical and preclinical examples of this principle will be presented. Thirdly, that BH3 mimetics kill cells in a TP53-independent fashion. Emerging evidence will be presented that supports the theoretical concept that these drugs work downstream of TP53, and so have particular attraction for use in TP53 functionally deficient tumors. Fourthly, that as yet the mechanisms for clinical resistance to venetoclax and navitoclax are unknown. These preliminary conclusions highlight the challenges for development of BCL2, BCLxL and MCL1 inhibitors, and the potential that they can deliver as future therapy for many cancers. Citation Format: Andrew Roberts. Optimizing BH3 mimetic therapy: Insights from BCL2 inhibition in the clinic. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr SY16-01.