Advanced diagnostics are enabling cancer treatments to become increasingly tailored to the individual through developments in immunotherapies and targeted therapies. However, long turnaround times and high costs of molecular testing hinder the widespread implementation of targeted cancer treatments. Meanwhile, gold-standard histopathological assessment carried out by a trained pathologist is widely regarded as routine and mandatory in most cancers. Recently, methods have been developed to mine hidden information from histopathological slides using deep learning applied to scanned and digitized slides; deep learning comprises a collection of computational methods which learn patterns in data in order to make predictions. Such methods have been reported to be successful in a variety of cancers for predicting the presence of biomarkers such as driver mutations, tumour mutational burden, and microsatellite instability. This information could prove valuable to pathologists and oncologists in clinical decision making for cancer treatment and triage for in-depth sequencing. In addition to identifying molecular features, deep learning has been applied to predict prognosis and treatment response in certain cancers. Despite reported successes, many challenges remain before the clinical implementation of such diagnostic strategies in the clinical setting is possible. This review aims to outline recent developments in the field of deep learning for predicting molecular genetics from histopathological slides, as well as to highlight limitations and pitfalls of working with histopathology slides in deep learning.

Prostate Cancer (PCa) is a leading cause of morbidity and mortality among men worldwide. For most men with PCa, their disease will follow an indolent course. However, advanced PCa is associated with poor outcomes. There has been an advent of new therapeutic options with proven efficacy for advanced PCa in the last decade which has improved survival outcomes for men with this disease. Despite this, advanced PCa continues to be associated with a high rate of death. There is a lack of strong evidence guiding the timing and sequence of these novel treatment strategies. This paper focuses on a review of the strategies for diagnostic and the current evidence available for treatment selection in advanced PCa.

Circular RNAs (circRNAs), a recently discovered non-coding RNA, have a number of functions including the regulation of miRNA expression. They have been detected in a number of malignancies including prostate cancer (PCa). The differential expression pattern of circRNAs associated with PCa and androgen receptor (AR) status was investigated in this study. circRNA profiling was performed using a high throughout microarray assay on a panel of prostate cell lines, which consisted of normal, benign, and malignant cells (n = 9). circRNAs were more commonly significantly up-regulated (p < 0.05) than downregulated in malignant cell lines (n = 3,409) vs. benign cell lines (n = 2,949). In a grouped analysis based on AR status, there were 2,127 down-regulated circRNAs in androgen independent cell lines compared to 2,236 in androgen dependent cell lines, thus identifying a potential circRNA signature reflective of androgen dependency. Through a bioinformatics approach, the parental genes associated with the top 10 differentially expressed circRNAs were identified such as hsa_circ_0064644, whose predicted parental gene target is RBMS3, and hsa_circ_0060539, whose predicted gene target is SDC4. Furthermore, we identified three circRNAs associated with the parental gene Caprin1 (hsa_circ_0021652, hsa_circ_0000288, and hsa_circ_0021647). Other studies have shown the importance of Caprin1 in PCa cell survival and drug resistance. Given the modified circRNA expression signatures identified here, these hypothesis generating results suggest that circRNAs may serve as potential putative diagnostic and predictive markers in PCa. However, further validation studies are required to assess the true potential of these markers in the clinical setting.