Open AccessSingle‐cell RNA‐seq reveals clonal diversity and prognostic genes of relapsed multiple myeloma
Author(s) -
He Haiyan,
Li Zifeng,
Lu Jing,
Qiang Wanting,
Jiang Sihan,
Xu Yaochen,
Fu Weijun,
Zhai Xiaowen,
Zhou Lin,
Qian Maoxiang,
Du Juan
Publication year - 2022
Publication title -
clinical and translational medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.125
H-Index - 1
ISSN - 2001-1326
DOI - 10.1002/ctm2.757
Subject(s) - multiple myeloma , oncology , clone (java method) , gene expression profiling , disease , gene , cancer research , malignancy , cohort , medicine , computational biology , biology , gene expression , genetics
Background Multiple myeloma (MM) is a clinically and biologically heterogeneous plasma‐cell malignancy. Despite extensive research, disease heterogeneity and relapse remain a big challenge in MM therapeutics. We tried to dissect this disease and identify novel biomarkers for patient stratification and treatment outcome prediction by applying single‐cell technology. Methods We performed single‐cell RNA sequencing (scRNA‐seq) and variable‐diversity‐joining regions‐targeted sequencing (scVDJ‐seq) concurrently on bone marrow samples from a cohort of 18 patients with newly diagnosed MM (NDMM; n = 12) or refractory/relapsed MM (RRMM; n = 6). We analysed the malignant clonotypes using scVDJ‐seq data and conducted data integration and cell‐type annotation through the CCA algorithm based on gene expression profiling. Furthermore, we identified disease status‐specific genes and modules by comparison of NDMM and RRMM datasets and explored the findings in a larger MM cohort from the MMRF CoMMpass study. Results We found that all the myeloma cells in either diagnosed or relapsed samples were dominated by a major clone, with a few subclones in several samples ( n = 5). Next, we investigated the universal transcriptional features of myeloma cells and identified eight meta‐programs correlated with this disease, especially meta‐programs 1 and 8 (M1 and M8), which were the most significant and related to cell cycle and stress response, respectively. Furthermore, we classified the malignant plasma cells into eight clusters and found that the cell numbers in clusters 2/6/7 were exclusively higher in relapsed samples. Besides, we identified several attractive candidates for biomarkers (e.g. SMAD1 and STMN1 ) associated with disease progression and relapse in our dataset and related to overall survival in the CoMMpass dataset. Conclusions Our data provide insights into the heterogeneity of MM as well as highlight the relevance of intra‐tumour heterogeneity and discover novel biomarkers that might be a potent therapy.