Highly multiplexed imaging recovers immune and metabolic niches in multiple myeloma associated with disease progression and bone involvement

Multiple myeloma (MM) is a cancer arising from genetically aberrant plasma cells (PCs) that remain dependent on the bone marrow (BM) microenvironment for disease establishment and progression. Here, we spatially mapped BM biopsies from patients with MM, smoldering MM and monoclonal gammopathy of undetermined significance by imaging mass cytometry. We found that PCs near bone surfaces were quiescent and demonstrated a distance–to-bone-dependent upregulation of IL-32 and HIF1α in patients with bone disease, consistent with their role in promoting bone loss in MM. We further identified two distinct PC neighborhoods termed PC_OXPHOS characterized by focal PC growth, enrichment of endothelial cells and elevated oxidative phosphorylation; and PC_MYELOID, characterized by PCs interspersed with immune cells and featuring a glycolytic phenotype. Spatial interactions between PCs and immune correlated with shorter progression free survival. Notably, a strong neighbor preference between PCs and CD4 + T cells was an independent predictor for disease progression. Our work underscores spatial context as a key factor in understanding MM pathogenesis and the potential for spatial analyses to improve MM risk stratification. Significance This study demonstrates that MM cells in different neighborhoods experience unique metabolic conditions and immune environments, and that neighbor preference of CD4 + T cells and plasma cell is associated with disease progression. These features were not captured by non-spatial metrics, highlighting the value of spatially resolved analyses in uncovering pathophysiological mechanisms of MM.