Spatial transcriptomics unveils novel potential disease mechanisms associated with the microenvironment in Multiple Myeloma

Understanding the spatial organization of the bone marrow (BM) microenvironment at single cell resolution and the niche specific transcriptional regulatory programs remains a challenge. Using formalin-fixed paraffin-embedded (FFPE) samples from our recently described MIcγ1 multiple myeloma (MM) mouse model (Larrayoz et al, Nat Med. 2023), we performed spatial transcriptional profiling of healthy and multiple myeloma BM using the Visium Spatial Gene Expression analysis (10x Genomics). A custom data-analysis framework that combines spatial with single-cell transcriptomic profiles allowed us to identify cell type composition and establish specific cell relations within the bone marrow. As a result, we defined the spatial distribution of transcriptionally heterogenous MM plasma cell (MM-PC) groups. Furthermore, we spatially defined transcriptional programs involved in the pathogenesis of MM within the BM microenvironment that were associated with MM-PC density gradient, such as the spatial location of T cells with effector and exhausted phenotypes. Application of spatial transcriptomics to FFPE human BM biopsies with varying degrees of PC infiltration validated findings from our murine model. The application of spatial transcriptomics to fully mineralized tissues provides a comprehensive spatial overview of the BM microenvironment allowing the identification of deregulated mechanism involved in the pathogenesis of diseases such as MM. This study includes spatial transcriptomics analysis of bone marrow samples from healthy mice (YFPcγ1 model) and Multiple Myeloma (MM) mice (MIcγ1 model), as well as human bone marrow biopsies from healthy individuals and MM patients. The analysis aims to compare the spatial and transcriptional profiles of healthy and diseased bone marrow, focusing on cell type composition and specific transcriptional programs associated with MM pathogenesis. No external treatments were applied, and samples were processed using 10x Genomics Visium Spatial Gene Expression technology.