OA – 28: MM progression and evolution of refractory disease explained by integrative multiomics in Moffitt Cancer Center's patient cohort

Introduction: We present Moffitt’s MM cohort, consisting of 1,005 patients with 1,627 bone marrow biopsies with RNA-seq/WES data, including precursor stages (MGUS/SMOL,16%), therapy naïve (18%) and relapsed disease (66%). We have applied an integrative approach to identify mechanisms driving progression and emergence of multi-drug resistance.


Methods: In addition to multivariate analysis on mutations, cytogenetic abnormalities and transcriptomic pathways associated with progression and refractory disease, we have conducted dNdS analysis to identify low-frequency mutated genes with imbalance between synonymous and non-synonymous mutations. Dimensionality reduction and clustering analysis recreated a transcriptional topology of MM. Geneset enrichment analysis identified biology associated with MM progression and refractory disease, as well as putative mechanisms driving aforementioned transcriptional dysregulation. Furthermore, we have conducted sc-multiomics in healthy donor plasma cells, primary MM cells from MGUS to refractory disease, as well as H3K27ac /CUT&TAG. shRNA-silencing of YY1, one of the identified putative drivers of multidrug resistance, in MM cell lines demonstrated in vitro re-sensitization to proteasome inhibitors (PI).


Results: In addition to previously identified MM frequently mutated genes (e.g. KRAS, NRAS, DIS3, etc.) dNdS analysis identified new loss of function (e.g. SUZ12, etc.) and gain-of-function mutated genes. Transcriptional topology consisted of two mega-clusters of co-expressing genes (beta and alpha). Beta contained the hallmarks EMT, inflammatory response, TNFalpha signaling via NFkB, IL6/STAT3 signaling, apical junction, etc., and was under-expressed in the transition from MGUS to SMOL. Alpha overlapped with hallmarks of oxidative phosphorylation, G2M checkpoint, MYC/E2F targets, etc. and was over-expressed in the transition from therapy naïve to refractory disease. Interestingly, no significant difference was observed in transcriptome between SMOL and therapy naïve MM. Enrichment analysis suggested that transcription of genes in beta and alpha were epigenetically regulated by H3K27me3 and H3K27ac, respectively. ScATAC/RNA-seq demonstrated correlation between differential expression and chromatin accessibility of genes from beta in precursor state transition (MGUS- >SMOL). CUT&TAG analysis confirmed H3K27ac signal in alpha genes, but not beta, as well as correlation between gene expression and H3K27ac signal. Genes in regular and super-enhancers in refractory disease, and genes involved in transcriptional activation domains (e.g. BRD4, CDK9, CTCF, and YY1) were over-expressed compared to therapy naïve MM. Functional validation by silencing YY1 in human MM cell line 8226 and a PI resistant lineage, led to re-sensitization of the latter.


Conclusions: We demonstrate that MM progression and refractory disease are driven by a multifactorial dynamic, with multiple initiating genomic events causing genome-wide epigenetic and transcriptomic dysregulation.