Advancing BH3 mimetics to treat cancers

Abstract

The evasion of apoptosis, one of the hallmarks of cancer, is observed in many cancers. This can also impair the efficacy of many conventional chemotherapies. The BCL2 protein family is the central regulator of the intrinsic apoptotic pathway and plays a vital role during tumor development. In particular, the levels of the pro survival family members are often elevated in some cancers. Venetoclax, a BH3 mimetic inhibitor that mimics the BH3-only proteins, natural inhibitors of the pro-survival BCL2 proteins, has proven to be effective for treating hematological cancers by selectively targeting BCL2. This has translated into regulatory approvals of venetoclax for treating a subset of chronic lymphocytic leukemia and acute myeloid leukemia. In addition to targeting BCL2, potent and specific BH3 mimetic inhibitors of its relatives, BCLxL and MCL1, are now also available. However, their full clinical utility is poorly defined.

This thesis focuses on advancing the utility of the BH3 mimetic compounds as anti-cancer agents. Previous studies have suggested roles for BCLxL and MCL1 in many solid cancers (e.g. colon, breast, lung). In particular, colorectal cancers have elevated levels of the pro survival protein, one usually associated with chemo resistance. Furthermore, colorectal cancer patients with advanced disease or those who carry poor prognostic markers do not respond well to the current stand of care therapies such as surgery and adjuvant chemo/radiotherapy.

Given the pressing need to find better treatments for these patients, we first utilized a panel of validated BH3 mimetics to assess the feasibility of using them for treating colorectal cancer. By using cancer cell lines and patient-derived organoids, we identified and validated BCLxL and MCL1 as the most important survival factors for colorectal cancer. We then validated them as potential targets by pharmacological inhibition in a mouse model in vivo. Moreover, we found that even those tumors that harbor poor prognostic factors respond as avidly as those do not, further highlighting the potential of this approach for treating patients with colorectal cancer.

Even though the targeting BCLxL might be a possible approach to kill cancers that depend on it, the clinical use of BCLxL selective inhibitors is limited due to the toxicity of BCLxL inhibition on platelets. I have screened for novel regulators of BCLxL using the CRISPR/Cas9 technology, which might offer potential approaches to target BCLxL safely.

The final goal of this thesis is to identify biomarkers that predict response to BH3 mimetics, given that there are few reliable tools to stratify patients that might respond well to these novel anti-cancer agents. By using large scale transcriptomic datasets from publicly available RNA sequencing studies, I was able to identify a few candidate genes and achieve reasonable prediction performance.