Golcadomide-Mediated Degradation of Aiolos/Ikaros Synergizes with BET Inhibitors through Bidirectional Restructuring of the Directly Regulated Epigenetic Environment in DLBCL

Background:

Golcadomide (GOLCA; CC-99282), a novel cereblon E3 ligase modulator (CELMoD) agent that degrades the B-cell lineage transcription factors Ikaros and Aiolos, has emerged as a potent option for treating DLBCL. This agent exerts significant cell-autonomous anti-DLBCL activity in preclinical models and has demonstrated notable clinical efficacy in combination with Rituximab and R-CHOP in early clinical trials. Given the epigenetic mechanisms through which Aiolos/Ikaros derive their activity, we sought to explore combination strategies with other epigenetic agents, such as BMS-986158 (BMS-158), a selective BET inhibitor, to further enhance GOLCA’s cell-autonomous activity. While each agent has significant single-agent activity, the combination results in a synergistic anti-proliferative effect characterized by a potent G1 cell cycle arrest. Herein, we deployed a series of epigenetic and transcriptomic analyses in a longitudinal series of experiments to characterize the mechanism through which this combination exerts its synergy.

Results:

Utilizing CRISPR/Cas9-mediated gene knockout, we validated that the phenotypic synergy between GOLCA and BMS-158 was dependent upon their on-target inhibition of Ikaros/Aiolos and BRD4, respectively. Gene set variation analysis (GSVA) analysis of RNA-seq data revealed MYC and E2F target genes were the most significant downregulated pathways by the combination treatment in both SU-DHL-4 and WSU-DLCL2 DLBCL cell lines. RNA-seq analysis of longitudinal time points (24, 48, 72 hrs) demonstrated that the combination treatment led to a progressive decrease of MYC and E2F target genes over the course of the three days in GOLCA single-agent resistant SU-DHL-4 cells. These decreases correlated with the observed G1 cell cycle arrest phenotype, as well as the reduced c-MYC and induced p21 mRNA and protein levels. Interestingly, exogenous overexpression of c-MYC in rescue experiments suggested that the downregulation of c-MYC by the combination treatment contributes to the induction of p21.

Profiling global chromatin accessibility by ATAC-seq demonstrated that the chromatin regions closing in response to the combination treatment were significantly enriched with a CTCF motif and several canonical E-box sequences, including c-MYC, N-MYC, and MAX motifs. Chromatin immunoprecipitation sequencing (ChIP-seq) revealed a high concordance of chromatin occupancy of Ikaros, Aiolos and BRD4, with 24,534 shared peaks identified across promoter (51.3%), gene body (32.2%), and intergenic (16.5%) regions. E2F targets and other cell cycle-related pathways (G2/M checkpoint, mitotic spindle, and MYC targets) topped the enrichment list of the shared promoter peaks. Examining the promoters of representative genes (e.g., MYC, AURKA, and PLK1) in these pathways confirmed the co-occupancy of the three factors and revealed a diminished H3K27ac activation mark in response to the combination treatment, mirroring their reduced expression. Genome-wide, BMS-158 further enhanced GOLCA-mediated depletion of Aiolos occupancy at both promoters and enhancers. On the other hand, while a sub-lethal dose of BMS-158 was sufficient to displace BRD4 from enhancers, reducing promoter-bound BRD4 required a combination of the two drugs.

Interestingly, both SU-DHL4 and WSU-DLCL2 harbor the t(3,8) translocation, rendering c-MYC expression dependent upon the BCL6 super-enhancer (SE). Our analysis demonstrates the BCL6-SE exhibits strong binding by Ikaros, Aiolos, and BRD4, which were significantly diminished by the combination treatment. The effects of the combination treatment on contacts between promoters and enhancers globally, including that of MYC promoter and BCL6-SE, are currently under investigation using chromosome conformation capture (Hi-C).

Conclusions:

Through a genome-wide unbiased multi-omics approach, our results demonstrated that the synthetic lethal relationship between Ikaros/Aiolos and BRD4 in DLBCL depends, at least in part, on their ability to collaboratively sustain MYC expression by epigenetically stimulating the activities of its promoter and the MYC-activating enhancer. These findings suggest that combining GOLCA with a BET inhibitor may result in greater clinical efficacy through enhanced cell-autonomous activity in patients with DLBCL.