Prediction of Mismatch Repair Status in Endometrial Cancer from Histological Slide Images Using Various Deep Learning-Based Algorithms.

Simple Summary: In the context of endometrial cancer, molecular classification is becoming increasingly significant as the molecular class determines appropriate treatments and the prognosis. However, performing clinical testing for molecular classification in a large number of patients entails significant financial and time costs. Therefore, there is a need to develop a substantial molecular profile screening method for endometrial cancers. The objective of this study was to explore whether the molecular classification of endometrial cancer could be predicted from digital images of hematoxylin and eosin (H&E)-stained slides using deep learning as a screening tool. After making adjustments to the training data set and hyperparameters, we confirmed the feasibility of estimating the mismatch repair status from histological digital images of endometrial cancer. Deep learning was found to be effective for predicting one aspect of the molecular classification from H&E-stained histological digital images. The application of deep learning algorithms to predict the molecular profiles of various cancers from digital images of hematoxylin and eosin (H&E)-stained slides has been reported in recent years, mainly for gastric and colon cancers. In this study, we investigated the potential use of H&E-stained endometrial cancer slide images to predict the associated mismatch repair (MMR) status. H&E-stained slide images were collected from 127 cases of the primary lesion of endometrial cancer. After digitization using a Nanozoomer virtual slide scanner (Hamamatsu Photonics), we segmented the scanned images into 5397 tiles of 512 × 512 pixels. The MMR proteins (PMS2, MSH6) were immunohistochemically stained, classified into MMR proficient/deficient, and annotated for each case and tile. We trained several neural networks, including convolutional and attention-based networks, using tiles annotated with the MMR status. Among the tested networks, ResNet50 exhibited the highest area under the receiver operating characteristic curve (AUROC) of 0.91 for predicting the MMR status. The constructed prediction algorithm may be applicable to other molecular profiles and useful for pre-screening before implementing other, more costly genetic profiling tests. [ABSTRACT FROM AUTHOR]