Multiple myeloma (MM) is a plasma cell malignancy that is located in the bone marrow (BM) and remains fatal for the majority of patients. Despite an increasing knowledge about the underlying tumor biology of MM, the mechanisms leading to relapse and finally to treatment failure have not been sufficiently identified. One possible explanation could be that previous studies were based almost exclusively on tumor specimens randomly collected from a single site of the skeleton, usually the iliac crest of the pelvis. Imaging studies, however, have shown MM to be a heterogeneously distributed disease with the majority of patients presenting with nodular accumulations of malignant plasma cells scattered throughout the bone marrow containing skeletal system ? so called focal lesions (FLs). Notably, the number of FLs represents an independent, negative prognostic marker for disease outcome. Furthermore, our own sequencing and expression data have revealed an accumulation of poor prognosis mutations in these regions and tumor growth-promoting conditions, which suggests that FLs are hotspots of tumor evolution. Despite their prognostic role, FLs are poorly understood and their cellular and extracellular composition and the interplay between these components remain uncharacterized. The central hypothesis of the proposed work is that specific tumor?microenvironmental (ME) interactions within FLs foster MM evolution, which finally leads to high-risk transformation and, subsequently, to relapse and treatment failure. To address this hypothesis, we will investigate the genotypic and phenotypic states of tumor and ME cells in FLs by single-cell RNA sequencing (Aim 1). BM specimens from non-FL sites from the same patient will be utilized as a control. The analysis of the proliferation and senescence state, variation in oncogenic signaling pathways, and various programs resulting in the exhaustion of T cells will unravel the specific features of FLs compared to other BM sites. Furthermore, we will perform a spatial-temporal DNA sequencing study of specimens from FLs and other sites of the skeletal system to determine the impact of anti-MM treatment on the spatial clonal architecture and the evolutionary history of selected clones (Aim 2). This will elucidate tumor-intrinsic mutational drivers and the mechanisms leading to drug-resistance and high-risk disease. To determine the functional consequences of key findings from this work, we will use our in-house SCID-rab model, a sophisticated murine model that allows for growth of MM cells in implanted rabbit bones (Aim 3). As a first step in these studies, will introduce an activating mutation into the STAT3 gene in the IL-6-dependent MM cell line INA-6 and introduce it into the SCID-rab model. This will reveal whether this variant leads to BM independency of the cell line and facilitates growth as FLs and/or at sites outside the BM. Successful completion of these studies will provide a mechanistic explanation for the prognostic value of the number of FLs. This will allow for the development of more effective therapies and finally lead to the cure of MM, which is the long-standing goal of this work.