Genomic and transcriptomic signatures of extramedullary multiple myeloma reveal possible drivers of therapeutic resistance

Abstract

Abstract Background: Extramedullary disease (EMD) is an aggressive manifestation of multiple myeloma (MM) associated with inferior overall survival (OS), defined by ability of a malignant clone to develop independently of the bone marrow (BM) microenvironment. The incidence of EMD has markedly increased in recent years, likely reflecting the prolonged survival of MM patients and emergence of late-stage treatment-resistance. Patients with EMD exhibit inferior progression free survival (PFS) and OS following CAR-T cell treatment, with EMD being a common form of relapse. Thus, EMD is an increasingly relevant and clinically challenging entity. The mechanisms driving EMD development and therapeutic resistance remain poorly understood but are pivotal for targeted treatment development. We performed genomic and transcriptomic profiling of paired BM and EMD (primary and secondary) sites to characterize the molecular phenotype of EMD and identify potential drivers and therapeutic targets. We also investigated if glycosylation, particularly sialylation, is altered at EMD sites and if this might contribute to therapeutic resistance, given that increased ST3GAL1 expression is associated with inferior PFS in MM and implicated in CAR-T resistance. Methods: Biopsy samples from 10 patients (12 EMD sites) were obtained following informed consent for DNA and RNA extraction, with 9 paired BM samples. CD138 staining was used for tumor selection and macrodissection from formalin-fixed paraffin-embedded (FFPE) tissues. DNA was sequenced using the Illumina NovaSeq 6000 platform (paired-end 150bp reads). RNA was sequenced using TWIST Bioscience RNA Exome platform for capture-based transcriptome profiling of FFPE tissues. Sequencing data quality was assessed using FastQC and Trimmomatic. Reads were aligned to the human genome (GRCh38) using BWA-MEM. Post-processing, including indexing, copy number variation (CNV) and variant calling, was performed in accordance with GATK best practices using SAMtools, Picard, GATK4, CNVkit, and Mutect2. CRISPR-Cas9 knockout (KO) of target genes was performed in OPM2 cells using multi-guide RNAs. KO efficiency was evaluated via PCR, amplicon sequencing, and downstream evaluation. Impact of target gene KO on therapeutic efficacy was assessed in vitro. Statistical analyses were conducted using GraphPad Prism. Retrospective chart review correlated molecular findings with clinical characteristics, treatment history, and outcomes. Results: Genomic analysis revealed substantial EMD vs BM heterogeneity. EMD samples exhibited consistent CN gains (log₂(CN/2) >0.3) affecting multiple genes on chromosomes 1 and 7, which were not prevalent in corresponding BM samples. Notably, MACC1 (chr7), a gene implicated in metastasis and epithelial-mesenchymal transition (EMT), demonstrated CN gains in 83% of EMD sites, vs 10% of BM samples. MET, a proto-oncogene transcriptionally regulated by MACC1, also showed CN gains in 83% of EMD samples. There was significant tumor mutational burden within pathways associated with anti-apoptotic and EMT signaling, including WNT, RTK-RAS, and PI3K in 95% of samples. Interestingly, 67% of EMD samples exhibited CN gain of CD274, encoding PD-L1, an established mediator of tumor immune evasion. Chromosome 6 CN losses (log₂(CN/2) <-1) were common in EMD. Within chromosome 1, CN gains were observed in 83% of EMD samples at FCRL5, vs 10% of BM samples. MCL1, an anti-apoptotic gene in the BCL2 family, showed gains in 67% of EMD samples compared to 20% of BM. MUC1, encoding a mucin glycoprotein, demonstrated CN gains in 75% of EMD samples and 10% of BM samples. ST3GAL1, which sialylates MUC1, had CN gain in 33% of EMD samples.CRISPR-Cas9 KO of selected genes in OPM2 cells, informed by EMD analysis, included ST3GAL1 which influenced therapeutic response in vitro, suggesting that hyper-sialylation may contribute to treatment resistance in MM and EMD.Conclusion: This study demonstrates a distinct genomic and transcriptomic profile for EMD in MM, with consistent CN gains in genes linked to proliferation, metastasis, immune evasion, and EMT. Mutational burden was significant in anti-apoptotic and signaling pathways relevant to drug resistance. MACC1 and MET, established EMT drivers in solid tumors, may contribute to clonal escape in EMD. Aberrant glycosylation and elevated sialyltransferase activity, particularly ST3GAL1, may mediate immune escape and therapeutic resistance via Siglec-sialoglycan interactions.