PI-RADS Scoring System for Prostate MRI
PI-RADS (Prostate Imaging-Reporting and Data System) version 2 is a standardized 5-point assessment scale developed by the American College of Radiology, European Society of Urogenital Radiology, and AdMeTech Foundation to categorize the likelihood of clinically significant prostate cancer on multiparametric MRI. 1
Core Purpose and Design
The PI-RADS v2 system was specifically designed to standardize mpMRI interpretation, reduce observer variability, and assist in patient selection for biopsy by summarizing the level of suspicion for clinically significant prostate cancer (csPCa). 1 The system defines csPCa as lesions with Gleason score ≥7, volume >0.5 cc, or extraprostatic extension. 2
Assessment Categories
The PI-RADS scoring uses a 5-point scale that indicates increasing likelihood of csPCa: 1
- PI-RADS 1: Very low probability of csPCa
- PI-RADS 2: Low probability of csPCa (unlikely but does not completely exclude cancer) 1
- PI-RADS 3: Intermediate/equivocal probability
- PI-RADS 4: High probability of csPCa
- PI-RADS 5: Very high probability of csPCa (highly likely but not definitive proof) 1
Zone-Specific Imaging Approach
PI-RADS v2 uses different dominant MRI sequences based on the anatomic zone where the lesion is located, rather than summing component scores as in version 1. 2
Peripheral Zone (PZ) Lesions
- Diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) maps is the dominant sequence 1, 2
- Dynamic contrast enhancement (DCE) plays a secondary clarification role only for equivocal PI-RADS 3 lesions 1, 2
Transition Zone (TZ) Lesions
- T2-weighted (T2W) imaging is the dominant sequence 1, 2
- DWI/ADC images provide supporting information for equivocal lesions 2
Technical Requirements
PI-RADS v2 establishes minimum acceptable technical parameters for prostate mpMRI acquisition, including: 1, 3
- T1-weighted imaging
- T2-weighted imaging
- Diffusion-weighted imaging
- Dynamic contrast-enhanced imaging 2
Notably, PI-RADS v2 eliminated spectroscopic imaging from the protocol, which was included in version 1. 1
Detection Capabilities and Limitations
The system is optimized to detect clinically significant cancer while intentionally having lower sensitivity for low-volume (<0.5 ml) Gleason 3+3 disease. 1 Specifically:
- Gleason 3+3 lesions require volumes ≥0.5 ml (approximately 9-10 mm diameter) to be reliably detected 1
- Index lesions with Gleason ≥3+4 pattern and volume ≥0.2 ml (approximately 7-8 mm diameter) can be identified on modern 3-T scanners with endorectal coils 1
- Larger and higher-grade lesions are more likely to receive PI-RADS 4-5 scores 1
Important caveat: Non-index lesions are frequently overlooked even when high-grade, and lesion sizes are often underestimated. 1
Clinical Application
PI-RADS v2 deliberately does not assign specific management algorithms to the assessment categories, acknowledging that clinical factors (PSA, PSA density, DRE, family history) must be integrated into patient decision-making. 1
When combined with PSA density, PI-RADS scoring significantly improves predictive accuracy: 4
- PI-RADS ≥4 with PSA density ≥0.15 ng/mL/mL yields 76-97% detection of csPCa 4
- PI-RADS ≤3 with PSA density <0.15 ng/mL/mL may allow avoidance of unnecessary biopsies 4
A negative mpMRI (PI-RADS 1-2) serves as an independent predictor for patient suitability for active surveillance and likely downgrading of Gleason 3+4 at biopsy to 3+3 at prostatectomy. 1
Quality Considerations
Image quality and diagnostic performance vary significantly between centers due to equipment capabilities, patient factors (body habitus, motion, rectal gas, post-biopsy hemorrhage), and most critically, radiologist interpretation experience. 1 This heterogeneity underscores the importance of using experienced centers for prostate mpMRI interpretation. 1