The future direction of imaging in prostate cancer: MRI with or without contrast injection

Multiparametric MRI (mpMRI) is the "state of the art" management tool for patients with suspicion of prostate cancer (PCa). The role of non-contrast MRI is investigated to move toward a more personalized, less invasive, and highly cost-effective PCa diagnostic workup.

To perform a non-systematic review of the existing literature to highlight strength and flaws of performing non-contrast MRI, and to provide a critical overview of the international scientific production on the topic.

Online databases (Medline, PubMed, and Web of Science) were searched for original articles, systematic review and meta-analysis, and expert opinion papers.

Several investigations have shown comparable diagnostic accuracy of biparametric (bpMRI) and mpMRI for the detection of PCa. The advantage of abandoning contrast-enhanced sequences improves operational logistics, lowering costs, acquisition time, and side effects. The main limitations of bpMRI are that most studies comparing non-contrast with contrast MRI come from centers with high expertise that might not be reproducible in the general community setting; besides, reduced protocols might be insufficient for estimation of the intra- and extra-prostatic extension and regional disease. The mentioned observations suggest that low-quality mpMRI for the general population might represent the main shortage to overcome.

Non-contrast MRI future trends are likely represented by PCa screening and the application of artificial intelligence (AI) tools. PCa screening is still a controversial topic; bpMRI has become one of the most promising diagnostic applications, as it is a more sensitive test for PCa early detection, compared to serum PSA level test. Also, AI applications and radiomic have been the object of several studies investigating PCa detection using bpMRI, showing encouraging results.

Today, the accessibility to MRI for early detection of PCa is a priority. Results from prospective, multicenter, multireader, and paired validation studies are needed to provide evidence supporting its role in the clinical practice.