Spatial transcriptomic atlas of aggressive osteosarcomas reveals shared immune landscape and targetable surface markers.

Osteosarcoma is characterized by extensive inter- and intra-tumoral heterogeneity, contributing to treatment resistance and poor outcomes. Here, we present a comprehensive spatial transcriptomics analysis of osteosarcoma, encompassing primary tumors and local or metastatic relapses across diverse phenotypic subtypes. Despite this heterogeneity, we identify a nine-gene cell surface signature with theranostic potential, validated in independent datasets and shown by immunohistochemistry to be distributed across distinct tumor compartments, supporting multi-targeted therapeutic strategies. Analysis of the tumor immune microenvironment reveals systematic lymphoid exclusion, differential myeloid infiltration patterns, and a type I interferon response signature that may explain the failure of IFN-α supplementation in prior trials. Notably, we provide evidence of monocyte-derived osteoclastic differentiation within human osteosarcoma lung metastases, identifying precursor populations with complex secretory phenotypes representing potential immunomodulatory targets. This study offers biological insights and translational opportunities while providing a resource for the osteosarcoma research community.