High-Resolution 3D Printing of Pancreatic Ductal Microanatomy Enabled by Serial Histology

Pancreatic ductal adenocarcinoma (PDAC) is a deadly cancer that can develop from pancreatic intraepithelial neoplasia (PanIN), a microscopic lesion in the pancreatic ductal system. PanIN and PDAC are conventionally studied in 2D via histological tissue sections. As such, their true structure is poorly understood due to the inability to image them in 3D. CODA, a recently developed technique for reconstruction of tissues at cellular resolution, is used to study the 3D morphology of the pancreas. Here, CODA is extended through 3D printing of normal pancreatic ducts, PanIN, and PDAC at cm-scale and mu m resolution. A methodology is presented to create 3D printable files from anatomical maps generated from serial histological images and to show detailed validation of the accuracy of this method. Existing 3D printing workflows utilizing medical images derived from computerized tomography (CT), X-ray, and magnetic resonance imaging (MRI) are scientifically proven to be useful in printing whole organ-scale structures with sub-mm resolution. Here, using serial histological sections, it is demonstrated that 3D printing of higher-resolution structures is also possible. Finally, with the 3D models of normal ducts, PanIN, and PDAC, marked changes to the structure of the human pancreas during tumorigenesis are revealed. In this work, the authors present and validate a workflow to create 3D printed models of microscopic features of the human pancreatic anatomy. Using CODA, a deep learning based workflow for digital 3D mapping of dense tissues at cellular resolution, 3D printed models of normal pancreatic ducts, pancreatic cancer precursors, and pancreatic cancer, are created.image