Protocol Selection and Optimization
A central issue in current practice is the choice of imaging protocol, particularly the role of biparametric MRI (bpMRI). By omitting dynamic contrast enhancement (DCE), bpMRI offers a pragmatic strategy to reduce costs, acquisition time, and patient burden. Emerging evidence, including prospective trials such as PRIME, supports the non-inferiority of bpMRI for detecting clinically significant prostate cancer in biopsy-naïve patients, provided that image quality is high and interpretation is performed by experienced readers.
Regardless of protocol selection, technical optimization remains critical. While 3-T systems are generally preferred due to their higher signal-to-noise ratio, 1.5-T systems remain a viable alternative when combined with optimized acquisition protocols and, increasingly, deep-learning–based reconstruction techniques. In parallel, patient preparation strategies—such as the management of rectal distension or the selective use of antispasmodic agents—may help mitigate motion artifacts, although their impact varies across clinical settings.
Addressing Image Quality
High-quality image acquisition is the cornerstone of reliable MRI interpretation. However, PI-RADS currently lacks a formal mechanism to classify or exclude non-diagnostic examinations. Unlike other structured reporting systems, there is no equivalent to a “PI-RADS 0” or “PI-RADS X” category. As a result, suboptimal studies may be inadvertently categorized as PI-RADS 3, contributing to unnecessary biopsies and reducing the positive predictive value of equivocal findings.
In this context, the adoption of standardized quality assessment tools—such as PI-QUAL version 2—should be strongly encouraged. Systematic grading of image quality would help ensure that only diagnostically adequate studies guide clinical decision-making, thereby improving both reliability and downstream outcomes.
Scoring Refinements
Several elements of the PI-RADS scoring system warrant further refinement. The current 1.5 cm lesion size threshold distinguishing PI-RADS 4 from 5 remains debated, with emerging evidence suggesting that a lower threshold (e.g., 1.0 cm) may better reflect the biological definition of clinically significant disease.
In addition, the introduction of an adjunctive category—such as a “PI-RADS M” (Malignancy) designation—could address scenarios involving atypical or infiltrative tumor patterns, including mucinous variants, that do not conform to conventional scoring criteria but nonetheless raise strong suspicion for malignancy. Such refinements may help reduce ambiguity and improve clinical decision-making in challenging cases.
Local Staging Challenges
Local staging represents another domain requiring greater standardization. Although PI-RADS v2.1 recommends reporting key features such as extraprostatic extension (EPE) and seminal vesicle invasion (SVI), it does not provide fully standardized imaging criteria for their assessment. This gap contributes to variability in reporting across institutions and readers.
Future iterations of PI-RADS should place greater emphasis on clear, reproducible definitions—particularly in differentiating intra-prostatic infiltrative disease from true extraprostatic spread, a distinction with direct implications for treatment planning and prognosis.
Future Directions
The next phase in the evolution of PI-RADS will likely be driven by two major shifts. First, the integration of artificial intelligence (AI) as a clinical “co-pilot” has the potential to reduce inter-reader variability, improve detection accuracy, and support less experienced readers. However, robust validation, standardization of workflows, and clarification of human–AI interaction models remain essential prerequisites for widespread implementation.
Second, there is a growing transition toward risk-based clinical pathways, moving beyond isolated imaging scores. The adoption of Key Performance Metrics (KPMs)—including grade selectivity, biopsy efficiency, and net clinical benefit—reflects a broader effort to evaluate imaging within the context of its impact on patient outcomes rather than purely diagnostic accuracy.
These principles are central to the development of the PI-RADS Pathway 2026, which aims to integrate imaging findings with clinical variables into a unified, evidence-based framework. Future updates, including a potential PI-RADS 3.0, will be guided by strict governance standards requiring high-level, multicenter evidence to ensure that any modifications translate into meaningful clinical benefit.
Written by: Andrea Ponsiglione,1 Ivo G. Schoots,2,3 and Anwar R. Padhani4
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy.
- Department of Radiology, The Netherlands Cancer Institute, Amsterdam, The Netherlands.
- Department of Radiology and Nuclear Medicine, Erasmus University Medical Center Rotterdam, Rotterdam, The Netherlands.
- Paul Strickland Scanner Centre, Mount Vernon Cancer Centre, Northwood, UK.