The BELLINI Study and the Dawn of Precision Medicine in Multiple Myeloma

Multiple myeloma (MM) is a molecularly complex and heterogeneous plasma cell malignancy as evidenced by various genomic abnormalities such as copy number aberrations (CNAs), IgH translocations, single nucleotide variations (SNVs), and other structural variations involving some known oncogenes.^1-3  Compared to previous decades, clinical outcomes have significantly improved in patients with MM; however, disease relapse and therapeutic resistance are still common during the disease course. Whether or not biomarker-driven therapy is needed to chip away at relapse remains uncertain; however, data published during 2020 could mark the beginning of a precision approach to this heterogeneous disease.

Dr. Shaji K. Kumar and colleagues reported results of a randomized, double-blinded, phase III trial (BELLINI) of the combination of venetoclax or placebo with bortezomib and dexamethasone in relapsed or refractory MM.⁴  Venetoclax is an oral selective inhibitor of the antiapoptotic BCL2 protein and belongs to the class of drugs known as BH3 mimetics. BCL2 is overexpressed in various malignancies, including chronic lymphocytic leukemia, non-Hodgkin lymphoma, and MM.^5,6  BCL2 and other antiapoptotic family members bind and sequester their proapoptotic counterparts to ensure tumor cell survival. When venetoclax binds to BCL2, it displaces the BH3 activator proteins, which subsequently activates proapoptotic proteins and induces mitochondrial membrane permeabilization and cell death.

The median progression-free survival (PFS) was significantly improved in the venetoclax arm compared to placebo (HR, 0.6; p=0.01), as was the depth of response. However, more deaths occurred in the venetoclax arm (HR, 2.03; p=0.034) — a finding that led the U.S. Food and Drug Administration to temporarily halt venetoclax trials in multiple myeloma until additional analysis is conducted. The prespecified subgroup analyses showed that patients with translocation t(11;14) or high BCL2 expression had significant improvement in PFS (HR, 0.11 and 0.5, respectively), without increased mortality. Overall survival was worse in the venetoclax arm; however, within that treatment cohort, there appeared to be improved outcomes in the subgroup of patients with t(11;14) or high BCL2, leading to the conclusion that there was an unfavorable risk-benefit ratio in patients without these molecular features.

Early studies led by Dr. Rafael Fonseca and others^7-10  showed that t(11;14) MM subgroups specifically have a lymphoplasmacytic morphology in 50 percent of cases, which was associated with a higher positive rate of B-lineage–associated antigens by immunohistochemistry (IHC) and expression of B-cell markers and signaling genes by gene expression profiling. Clinically, this subgroup tends to have lower M protein levels, enriched in light chain subtype, and less likelihood of having hyperdiploid DNA content.^1,8  These findings highlight the unique molecular, histological, and clinical features of this subgroup. Nonetheless, more studies are still needed to understand the relationship between the t(11;14) subgroup and BCL2 overexpression, and whether other factors contribute to therapeutic response.

Although this trial was underpowered for understanding the definite effect of venetoclax on the t(11;14) subgroup, it has provided compelling evidence for using this agent in future trials targeting this molecular subgroup, particularly those with high BCL2 expression. These innovative next-generation trials should be accompanied by extensive correlative studies to identify predictive biomarkers of response or resistance to venetoclax. Moreover, using BH3 profiling assays could provide a more comprehensive picture of antiapoptotic dependence in MM, because looking at BCL2 expression alone is unlikely to tell the complete story. It is known that other antiapoptotic proteins such as BCLXL and MCL1 can sequester proapoptotic proteins liberated from BCL2 by BH3 mimetics. Preclinical in vitro studies of MM cells co-expressing BCL2 and BCLXL were resistant to venetoclax but sensitive to navitoclax or BCLXL-selective inhibitor; moreover, xenograft models that co-expressed BCLXL or MCL1 with BCL2 were resistant to venetoclax.¹¹  Moreover, IHC of patient tumors revealed high levels of BCL2 and BCLXL in 62 percent and 43 percent of evaluable samples, respectively, while 34 percent were characterized as BCL2^High/BCLXL^low.¹¹  These findings underscore the fact that dependence on individual antiapoptotic proteins varies greatly among MM patients, and a broader evaluation of predictive biomarkers is essential for a precise and smart therapeutic intervention.

In conclusion, this 2020 publication is the first to provide data of a biomarker-driven intervention in MM rather than the traditional method of a one-size-fits-all approach, and could be the beginning of a new exciting era of precision medicine in MM.