Radiogenomics of Prostate Cancer: Current State and Future Directions
摘要
Radiomic analysis, which has developed rapidly in recent years, has become extremely popular as a promising subject for the future. Radiomic data obtained from prostate-specific membrane antigen positron emission tomography/computed tomography (PSMA-PET/CT), prostate-specific membrane antigen positron emission tomography/magnetic resonance (PSMA-PET/MR), ultrasonography (USG), sonoelastography, computed tomography (CT), CT perfusion images, and multiparametric magnetic resonance imaging (mpMRI), which is the most frequently used imaging technique in prostate cancer, have been evaluated in various studies. The radiomic data obtained from the prostate gland and/or tumour has been used alone or combined with clinical, genomic, epigenomic, transcriptomic, proteomic, and metabolic data, and models have been developed through interpretation with artificial intelligence–supported statistical methods. It has been attempted to associate the developed models with many parameters such as Gleason Score prediction, prostate imaging-reporting and data system (PI-RADS) prediction, clinically important prostate cancer diagnosis, differentiation of benign and malignant lesions, prediction of low and high-risk lesions, prediction of extra-prostatic extension, follow-up of patients under active observation, need for repeated biopsy, prediction of treatment efficacy, overall survival prediction, biochemical recurrence prediction, metastatic progression risk prediction, prediction of toxicity developing associated with radiotherapy, and prediction of the need for adjuvant radiotherapy and androgen deprivation therapy (ADT). Despite the rapid and successful results, there is a need for the automisation of prostate gland segmentation and the standardisation of radiomic signatures obtained and the models developed, and for these studies to be confirmed with large data sets. With increasing experience in these fields and completion of deficiencies, it is expected that radiogenomic approaches will have an important place in clinical practice in the not too distant future.