Bone marrow metastasis frequently occurs in patients with solid cancers and most often leads to poor outcome. Yet, the composition of bone marrow metastases, including tumor and surrounding cells, has so far not been characterized. Herein, we aimed to investigate the diversity of tumor and surrounding cells, i.e., the microenvironment, in bone marrow metastases, using the childhood tumor neuroblastoma as a model. To this end, we screened genome-wide datasets to define a panel of cell-specific markers for multiplex microscopy of metastatic bone marrow samples, and developed DeepFLEX, a computational pipeline for subsequent image analysis. Thereby, we identified 35,000 single cells covering metastasized tumor cells, and various types of developing immune and bone marrow cells. In parallel, we analyzed the transcriptome, i.e., all genes that are expressed as mRNA, of 38 patients with and without bone marrow metastasis. We found vast tumor cell diversity and identified a marker protein, FAIM2, which can help to identify a broader range of tumor cell variants. In addition we showed that tumor cell metastasis in the bone marrow is associated with an immune response resembling inflammation, and the presence of cells that can repress an immune attack against cancer cells. Our study suggests that metastatic tumor cells are shaping the bone marrow microenvironment and builds the basis to further investigate its clinical relevance.

While the bone marrow attracts tumor cells in many solid cancers leading to poor outcome in affected patients, comprehensive analyses of bone marrow metastases have not been performed on a single-cell level. We here set out to capture tumor heterogeneity and unravel microenvironmental changes in neuroblastoma, a solid cancer with bone marrow involvement. To this end, we employed a multi-omics data mining approach to define a multiplex imaging panel and developed DeepFLEX, a pipeline for subsequent multiplex image analysis, whereby we constructed a single-cell atlas of over 35,000 disseminated tumor cells (DTCs) and cells of their microenvironment in the metastatic bone marrow niche. Further, we independently profiled the transcriptome of a cohort of 38 patients with and without bone marrow metastasis. Our results revealed vast diversity among DTCs and suggest that FAIM2 can act as a complementary marker to capture DTC heterogeneity. Importantly, we demonstrate that malignant bone marrow infiltration is associated with an inflammatory response and at the same time the presence of immuno-suppressive cell types, most prominently an immature neutrophil/granulocytic myeloid-derived suppressor-like cell type. The presented findings indicate that metastatic tumor cells shape the bone marrow microenvironment, warranting deeper investigations of spatio-temporal dynamics at the single-cell level and their clinical relevance.