Morphological Features Extracted by AI Associated with Spatial Transcriptomics in Prostate Cancer

Simple Summary Prostate cancer has very varied appearances when examined under the microscope, and it is difficult to distinguish clinically significant cancer from indolent disease. In this study, we use computer analyses inspired by neurons, so-called 'neural networks', to gain new insights into the connection between how tissue looks and underlying genes which program the function of prostate cells. Neural networks are 'trained' to carry out specific tasks, and training requires large numbers of training examples. Here, we show that a network pre-trained on different data can still identify biologically meaningful regions, without the need for additional training. The neural network interpretations matched independent manual assessment by human pathologists, and even resulted in more refined interpretation when considering the relationship with the underlying genes. This is a new way to automatically detect prostate cancer and its genetic characteristics without the need for human supervision, which means it could possibly help in making better treatment decisions. Prostate cancer is a common cancer type in men, yet some of its traits are still under-explored. One reason for this is high molecular and morphological heterogeneity. The purpose of this study was to develop a method to gain new insights into the connection between morphological changes and underlying molecular patterns. We used artificial intelligence (AI) to analyze the morphology of seven hematoxylin and eosin (H & E)-stained prostatectomy slides from a patient with multi-focal prostate cancer. We also paired the slides with spatially resolved expression for thousands of genes obtained by a novel spatial transcriptomics (ST) technique. As both spaces are highly dimensional, we focused on dimensionality reduction before seeking associations between them. Consequently, we extracted morphological features from H & E images using an ensemble of pre-trained convolutional neural networks and proposed a workflow for dimensionality reduction. To summarize the ST data into genetic profiles, we used a previously proposed factor analysis. We found that the regions were automatically defined, outlined by unsupervised clustering, associated with independent manual annotations, in some cases, finding further relevant subdivisions. The morphological patterns were also correlated with molecular profiles and could predict the spatial variation of individual genes. This novel approach enables flexible unsupervised studies relating morphological and genetic heterogeneity using AI to be carried out.Peer reviewe