BostonGene to Present AI Advances in Tumor Classification and Microenvironment Profiling at ESMO Congress
BostonGene is set to present its advancements in multimodal AI for tumor-of-origin classification and tumor microenvironment profiling at the ESMO AI & Digital Oncology Congress 2026 in Berlin, Germany. The company has developed a proprietary pan-cancer deep learning platform that analyzes bulk RNA-seq data across 20 tumor microenvironment cell types and 13 tumor diagnoses, generating over 14,000 compartment-specific gene expression profiles. This platform, evaluated in collaboration with Inova Schar Cancer Institute and Endeavor Health, can distinguish gene expression from tumor cells versus the tumor microenvironment directly from bulk RNA-seq. Additionally, BostonGene has clinically validated a tri-modal fusion framework, integrating histopathology, transcriptomics, and genomics, to accurately predict tumor origin in cancers of unknown primary (CUP) even from partial diagnostic profiles. This approach aims to rescue fragmented clinical data, accelerate CUP diagnostic workflows, and streamline precision ...