Targeting Bcl-2 Proteins in Myeloma

Escape from apoptosis is a hallmark of malignant cells that constitute an important mechanism associated to therapeutic resistance. Hematologic malignant cells are described as primed for apoptosis, a state in which pro-survival BCL2 family proteins sequester elevated levels of pro-apoptotic BH3-only proteins or even pre-activated BAX/BAK to ensure survival. Then, mimicking BH3-only proteins might preferentially turn on apoptosis in malignant cells. The understanding of BCL2 biology combined to drug discovery efforts have succeeded to the development of BH3-mimetics that bind with high affinity to a specific anti-apoptotic protein, promoting the release of pro-apoptotic members, leading to BAX and/or BAK activation and apoptosis. Venetoclax (ABT-199) is the first-in-class oral BH3-mimetic targeting selectively BCL2 and approved by the FDA for the treatment of CLL patients with 17p deletion. More recently, BH3-mimetics targeting MCL1 have been also described and are currently undergoing early phase clinical evaluation. Multiple myeloma (MM) displays a molecular heterogeneity, which includes hyperdiploid patients and patients with an Ig translocation with different chromosomes (4, 11, 6 or 16) leading to an over-expression of MMSET, CCND1, CCND3 or MAF genes, respectively. Analysis of apoptotic machinery diversity supports the notion that MM heterogeneity is extended to the BCL2 family content. The combined profile of BCL2, BCLXL and MCL1 mRNA expression in the main MM molecular subgroups was sufficient to discriminate each of them, suggesting that the composition of anti-apoptotic proteins should be taken into account for targeted therapies. MCL1 is over-expressed in MM and its strongest expression was found in the MMSET and MAF subgroups associated with an adverse prognosis. In contrast, the CCND1 subgroup expresses high levels of BCL2 and low level of MCL1 and BCLXL. In accordance to the heterogeneity of BCL2 member expression, different approaches such as MCL1 gene editing, BH3 profiling and in vitro sensitivity to specific BH3-mimetic demonstrate that myeloma cells are dependent on BCL2, MCL1 or BCLXL or co-dependent on two or three anti-apoptotic proteins. A plasticity of myeloma cell dependency was also reported according to disease status. Preclinical evaluation of venetoclax reveals that the sensitivity to venetoclax was mainly found in the CCND1 subgroup but not exclusively. The ratio of BCL2/BCLXL appears as a predictor of venetoclax sensitivity. A phase 1 clinical trial of venetoclax in relapsed/refractory myeloma patients demonstrate its efficacy mainly in t(11;14) patients with 40% of overall response. These results validate that targeting BCL2 as a very promising strategy for some patients, nevertheless stratification of patients based on t(11 ;14) and/or BCL2/BCLXL ratio is not sufficient to predict venetoclax response. Recent findings suggest that ex vivo pretreatment sensitivity could predict the clinical response to venetoclax. The reliability of the different approaches used to determine BCL2 dependency will be further discussed. A better comprehension of how to increase BCL2 priming is one of the most interesting perspectives to find the best combination regimens with venetoclax. Indeed, dexamethasone increases the BH3-only Bim protein and shifts it's binding towards BCL2. Phase 1 study of the combination of dexamethasone/venetoclax confirms the interest of this combination for t(11;14) patients. Finally, the possibility to overcome the resistance to venetoclax by targeting directly or indirectly the BCXL and MCL1 resistant factors could be of particular relevance. In conclusion, how basic research should work hand-in hand with clinical practice to design pertinent protocol to evaluate BCL2 targeted therapy will be examined.