Multiomic profiling of human and canine soft-tissue sarcomas reveals extensive molecular homology across species and identifies clinically relevant subgroups
Fuchs, D.; Jarosch, A.; Beebe, E.; Poeschel, A.; Sarver, A. L.; Kauzlaric, A.; Ruiz Buendia, G.; Roh, V.; Fournier, N.; Weber, M.; Opitz, L.; Kunz, L.; Wolski, W.; Guscetti, F.; Floercken, A.; Nolff, M. C.; Markkanen, E.
Show abstract
Soft-tissue sarcoma (STS) are rare and heterogeneous mesenchymal tumours with over 100 recognized human subtypes. Despite advances in cytogenetic and molecular characterization, diagnostic precision and therapeutic options remain limited for most subtypes. Spontaneously occurring canine STS could represent valuable translational models, due to their higher incidence and clinical similarity to human counterparts. However, molecular cross-species comparisons of specific subtypes are largely missing. Here, we performed a tissue-resolved, cross-species analysis of tumour and matched adjacent normal tissue (NT) in human and canine fibrosarcoma (FSA) and myxofibrosarcoma (MFS) by laser-capture microdissection of FFPE specimens combined with RNAseq and LC-MS/MS. Multimodal profiling revealed FSA and MFS to represent a molecular continuum rather than distinct entities in both species, resulted in identification of clinically relevant subgroups based on immune activation, proliferative activity and copy number alterations, and identified a novel canine STS subtype associated with a gene fusion. Moreover, our analyses revealed cross-species conserved transcriptomic and proteomic alterations distinguishing tumour from NT, including pathways linked to extracellular matrix remodelling, immune modulation, and cell proliferation. These data establish the first comprehensive molecular comparison of canine and human FSA and MFS, highlight the translational relevance of canine models, and identify candidate biomarkers for diagnostic refinement and development of targeted therapeutic modalities.
Matching journals
The top 11 journals account for 50% of the predicted probability mass.