Spatial heterogeneity of PD-1/PD-L1 defined osteosarcoma microenvironments at single-cell spatial resolution.

Abstract

Osteosarcoma, predominantly affecting children and adolescents, is a highly aggressive bone cancer with a 5-year survival rate of 65-70%. The spatial dynamics between TAM and other cellular subtypes, including T cells, osteoblasts and osteoclasts, are critical for understanding the complexities of the osteosarcoma tumor microenvironment (TME) and can provide insights into potential immunotherapeutic strategies. Our study employs a pioneering approach that combines deep learning-based digital image analysis with multiplex fluorescence immunohistochemistry (mfIHC) to accurately implement cell detection, segmentation, and fluorescence intensity measurements for in-depth study of the TME. We introduce a novel algorithm for TAM/osteoclast differentiation, crucial for accurate characterization of cellular composition. Our findings reveal distinct heterogeneity in cell composition and spatial orchestration between PD-1 (-/+) and PD-L1 (-/+) patients, highlighting the role of T-cell functionality in this context. Furthermore, our analysis demonstrates the efficacy of nivolumab in suppressing tumor growth and enhancing lymphocyte infiltration without altering the M1/M2 TAM ratio. This study provides critical insights into the spatial orchestration of cellular subtypes within the PD-1/PD-L1 defined osteosarcoma TME. By leveraging advanced mfIHC and artificial intelligence, we underscore the critical role of TAMs and T-cell interactions, proposing new therapeutic avenues focusing on TAM repolarization and targeted immunotherapies, thus underscoring the study's potential impact on improving osteosarcoma treatment.