Genomic characteristics and prognostic correlations in Chinese multiple myeloma patients

Abstract

Background: Multiple myeloma (MM) is a hematologic malignancy characterized by the proliferation of abnormal clonal plasma cells in the bone marrow. The heterogeneity in Chinese MM populations remains underexplored.

Methods: We conducted whole-exome sequencing (WES) on 241 tumor samples, complemented by RNA sequencing (RNA-seq) on 131 samples from 212 Chinese MM patients.

Results: We identified a novel mutational signature and analyzed molecular differences between newly diagnosed MM (NDMM) and relapsed/refractory MM (RRMM) patients. NFKBIA mutations were notably more frequent in NDMM patients compared to the MMRF-COMMPASS cohort (4/50 vs 22/937, p = 0.048), with additional recurrent mutations in several genes like TTN, IGLL5 and SYNE1. In RRMM patients, UBR5 mutations were more prevalent (4/24 vs 0/50, p = 0.01), alongside frequent mutations in OBSCN, CACNA1H, and HSPG2. Clonal evolution was assessed through multiple time points and locations, identifying genes potentially linked to circulating plasma cell formation. Cox regression analysis revealed that age and mutations in OBSCN and RB1 were significant predictors of progression-free survival (PFS) in NDMM patients. Additionally, albumin, β2-microglobulin, and RB1 mutations were correlated with overall survival (OS).

Conclusions: In summary, we characterized the genomic landscape of MM in diverse Chinese populations, confirmed clonal evolution, and identified prognostic genes.

Keywords: Genomic characteristics; Multiple myeloma; Prognosis.

Fig. 1

Distinct molecular characteristics of NDMM and RRMM. A Mutational signature identified in Chinese MM patients (n = 74). A novel signature termed ‘SBSA’ was identified and showed high similarity (cosine similarity = 0.78) to signature 15 in COSMIC V2. B Comparison of mutation burden between NDMM and RRMM. Each column in the histogram represents a sample, with somatic mutations divided into single nucleotide variant (SNV) and insertion and deletion (INDEL). C Comparison of number of segments with CNVs between NDMM and RRMM. Each column in the histogram represents a sample, with CNVs divided into increased copy number (gain) and decreased copy number (loss). D The landscape of CNVs of NDMM. Left: Significant (q < 0.25) recurrent focal amplified CNVs detected by CNVkit along all autosomes using GISTIC 2.0 are shown. Right: Significant (q < 0.25) recurrent focal deleted CNVs detected by CNVkit along all autosomes using GISTIC 2.0 are shown. The Cancer Gene Census (CGC) genes and MM driver genes located in these regions are highlighted and significant CNV regions present in both NDMM and RRMM are displayed in bold font. E The landscape of CNVs of RRMM

Fig. 2

Recurrent mutated genes in NDMM. A Waterfall of NDMM patients’ gene mutations. All genes are divided into three groups, genes identified as SMGs by MutsigCV (SMG), previously reported SMGs (Public SMG) and some novel mutant genes (Novel). In the third group, six genes also appear in the validation cohort (cohort 1A, n = 77) at a higher frequency (> 5%) are marked with an asterisk. Related clinical characteristics of all patients are shown at the top, including sex, age at diagnosis, DS stage, ISS stage, R-ISS stage and sum of FISH results, which include t(4;14), t(14;16) and 17p-. DS Durie-Salmon, ISS International Staging System, R-ISS Revised International Staging System, FISH fluorescence in situ hybridization, SMG significantly mutated gene. B Frequency comparison of SMGs between cohort 1a and MMRF-COMMPASS. C NFKBIA somatic mutation sites in cohort 1a

Fig. 3

Recurrent mutated genes in RRMM. A Waterfall of RRMM patients’ gene mutations. B Frequency comparison of SMGs between cohort 1a and cohort 1b. C UBR5 somatic mutation sites in cohort 1b

Fig. 4

Clonal evolution in MM. A Clonal evolution of patient 189. Fishplot showing dynamic clonal evolution from BM-PCs to CPCs in patient 189. B Clonal evolution of patient 190. Fishplot showing dynamic clonal evolution from BM-PCs to CPCs in patient 190. C Clonal evolution of patient 193. Fishplot showing dynamic clonal evolution of patient 193 when disease progressed from the initial diagnosis. D Clonal evolution of patient 209. Fishplot showing dynamic clonal evolution of patient 209 when disease progressed from the remission phase. “*” SMGs, “^#” Cancer Gene Census (CGC) genes, “⁺” novel recurrent mutated genes identified in NDMM patients, “⁺⁺” novel recurrent mutated genes identified in RRMM patients, CBD C: cyclophosphamide; B: bortezomib; D: dexamethasone, VP16 etoposide, KPD K: carfilzomib; P: pomalidomide; D: dexamethasone, + positive,—negative, NA not available

Fig. 5

Prognosis analysis of 64 NDMM patients. A Univariate cox logistic regression analysis of PFS in 64 NDMM patients. B Multivariate cox logistic regression analysis of PFS in 64 NDMM patients. C Univariate cox logistic regression analysis of OS in 64 NDMM patients. D Multivariate cox logistic regression analysis of OS in 64 NDMM patients. m_OBSCN mutation of OBSCN, m_RB1 mutation of RB1, IGLL5 expression of IGLL5, Hb hemoglobin, ALB serum albumin, dFLC difference between involved and uninvolved free light chain