Deciphering tumour tissue organization by 3D electron microscopy and machine learning

This study analyzed the 3D organization of hepatoblastoma patient-derived xenograft tissues using serial block-face scanning electron microscopy and machine learning. Digital reconstitution of the sample revealed correlations between hepatoblastoma cell size and their nucleus, cytoplasm, and mitochondrial mass. Anatomical connections were found between blood capillaries and the planar alignment and size of tumor cells. A set of tumor cells showed polarization towards a bile canaliculus-like structure. The study identified bioarchitectural parameters shaping tumor organization, furthering the field of onconanotomy.

Baudouin Denis de Senneville, Fatma Zohra Khoubai, Marc Bevilacqua, Alexandre Labedade, Kathleen Flosseau, Christophe Chardot, Sophie Branchereau, Jean Ripoche, Stefano Cairo, Etienne Gontier, Christophe F. Grosset