Reinterpretation of prostate cancer pathology by Appl1, Sortilin and Syndecan-1 biomarkers

The diagnosis of prostate cancer using histopathology is reliant on the accurate interpretation of stained or labelled tissue sections. Current standards rely on assessment following Haematoxylin and Eosin (H&E) staining, which is often difficult to interpret and introduces inter-observer variability. Here, we present a digital pathology atlas for prostate cancer tissue, using micrographs of both H&E and our novel set of three biomarkers as an interactive tool, where clinicians and scientists can explore high resolution histopathology from various case studies. The digital pathology prostate cancer atlas when used in conjunction with the biomarkers, will greatly assist pathologists to accurately grade prostate cancer tissue samples.

This repository also contains the original image dataset used to generate the case studies for the digital pathology atlas. The data consist of a benign case and cases of ISUP grades from 1 to 5, where initial grading was performed on the hematoxylin and eosin (H&E) stained tissue section, along with three serial sections from the same tissue block, labeled with our biomarker panel - Appl1, Sortilin, and Syndecan-1. This biomarker panel was then used to assist the initial H&E grading, providing a more accurate diagnosis and prognosis from the retrospective patient dataset.

Methods
All images were acquired using a Carl ZEISS AxioScan.Z1 microscope, with a Planapochromat 40x/0.95 objective. Images were orginially stored in the native Carl ZEISS Image format (*.CZI), and then converted to BigTIFF format using Carl ZEISS Zen Blue Software (version 3.8). Lossless compression was also used to minimise file sizes, while retaining image quality. All images can be  using freely available imaging software, such as ImageJ or FiJi.