S-PI-RADS and PI-RRADS for Biparametric MRI in the Detection of Prostate Cancer and Post-treatment Local Recurrence

Biparametric Prostate MRI Protocol in Diagnosis and Post-treatment of Prostate Cancer

Prostate biparametric MRI at 3T without endorectal coil (ERC) and with a phased array body coil for the detection of PCa and the local recurrence of PCa workup includes: axial, sagittal, and coronal T2W, axial DWI using b-values of 0, 500, 1000 and 1500 s/mm² and ADC map calculation, axial pelvic T1W fat-suppression, and T2-weighted SPIR MRI. MRI scanner at 1.5T with updated/optimal protocols would also produce good quality diagnostic images.

The prostate bpMRI protocol at 3T without ERC for diagnosis of PCa and detection of local recurrence after RT and RP is shown in Table I.

Steps for Biparametric MR Images Reading and Interpretation

To simplify and facilitate the use of S-PI-RADS and PI-RRADS, we recommend utilizing a 3-step approach which includes: 1) lesion detection and categorization, 2) prostate volume measurement and 3) lesion localization.

1) Lesion detection and categorization. In the detection of PCa, preliminary the axial T1-weighted fat-suppression images were analyzed to exclude hemorrhage post-biopsy foci. Next, four images (axial DWI with high b-value image, ADC map, sagittal and axial T2W images) are displayed on one or two 21.3-inch standard radiology high resolution 2 megapixel monitors and a contrast ratio of 1400:1. DWI at high b-values at 1500 s/mm² (or inverted) and corresponding ADC maps were analyzed: a) for the detection PCa [both in peripheral zone and transition zone (TZ)] and for its categorization, and b) for detection and assessment of the probability of local recurrence in the irradiated prostate gland (PZ and TZ) and at level of the prostatectomy bed after RT and RP, respectively.

In patients suspicious of PCa, the lesion appears hyperintense on DWI at high b-values and moderate homogeneous or marked homogeneous or inhomogeneous hypointense on ADC maps. The lesions are categorized on ADC map as moderate homogeneous hypointensity (S-PI-RADS category 3) and marked homogeneous or inhomogeneous hypointensity (S-PI-RADS category 4). Moderately homogenous hypointense lesions on the ADC map are classified as S-PI-RADS category 3. These lesions were discriminated based on volume, calculated on DWI at high b-values with the ellipse formula (V=L×H×W×0.52) and in agreement with Epstein criteria (10), using a cut-off value of 0.5 cm³, in the S-PI-RADS category 3a [volume <0.5 cm³ (no biopsy)], and category S-PI-RADS 3b [volume ≥0.5 cm³ (targeted biopsy)]. Otherwise, targeted biopsy is indicated for all S-PI-RADS category 4. In a prostate with multiple lesions that show different restriction of diffusion, the “index lesion” should be considered the largest with marked restriction of diffusion or the smallest with marked restriction of diffusion compared to others that may be even larger but with moderate diffusion restriction.

After RP, the PCa recurrence appears as lobulated, curvilinear, or semi-circumferential, nodular-like, or plaque-like soft tissue thickening hyperintense on DWI at high b-values and moderately homogeneous or markedly homogeneous or inhomogeneous hypointense on ADC maps. DWI/ADC after RP allows detection of lesions suspected of local recurrence ≥5 mm (11).

After RT, the signal pattern of recurrence on DWI/ADC is similar to that of in situ primary PCa (12, 13); it appears as mass-like, focal rounded, lenticular or irregular.

After RT and RP, local recurrence appears as a lesion with SI moderately or markedly hypointense on ADC and classified as category 3 and 4, respectively.

In addition to qualitative analysis of lesions, for accurate PCa risk stratification by S-PI-RADS and categorization of 3 and 4 probability of local recurrence in PI-RRADS, and to improve inter-reader agreement, we proposed quantitative analysis of the SI of lesions using a grayscale ADC map.

We suggest the adoption of ADC quantitative analysis for measurable lesions. In particular, in S-PI-RADS, for differentiating lesions with volume <0.5 cm³ (S-PI-RADS category 3a vs. 4) and to indicate the follow-up for the subcategory 3a, while, in the PI-RRADS to distinguish probability categories (PI-RRADS category 3 vs. 4) and to monitor the treatment response. The grayscale ADC map for quantitative analysis of the SI of lesions in S-PI-RADS and PI-RRADS is reported in Figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Grayscale ADC map for quantitative analysis of the signal intensity (SI) of suspicious areas in S-PI-RADS and PI-RRADS. The apparent diffusion coefficient (ADC) scale measures the SI values of normal tissue and cell density of prostate cancer that increase with a decrease in ADC values. Three sub-parameters are identified: ADC minimum or ADC_min, ADC mean, and ADC maximum or ADC_max. ADC_max value is the water or urine (white) with high signal intensity value and is obtained from average of three circular regions-of-interest (ROIs) placed within the bladder; ADC_max value varies on the same MRI scanner and between different MRI scanners [for example on our 3T MRI unit (Achieva, Philips Medical Systems, Healthcare, Eindhoven, the Netherlands)] equipped with a phased array body coil, the ADC_max value on the ADC map generated using b-values of 0, 500, 1000 and 1500 s/mm² can reach an average value of 4000]. ADC_min value is the bone (black) equal to 0 and is obtained from a circular ROI placed within the femur head; ADC_min value does not vary on the same MRI scanner nor between different MRI scanners. ADCmean is obtained from the formula: ADC_mean=(ADC_max + ADC_min)/2. On grayscale ADC map, measurable and homogeneous areas with ADC values ≥ ADCmean can be considered as moderately hypointense SI areas on ADC map, while those with ADC values < ADCmean are considered as markedly hypointense SI areas on the ADC map.

2) Prostate volume measurement. In the detection of primary PCa and/or local recurrence after RT, the prostate volume is calculated on axial and sagittal T2W imaging using the ellipsoidal formula (V=L×H×W×0.52).

3) Lesion localization. In the detection of PCa and after RT, T2W imaging is able to localize the lesion detected on DWI/ADC and in the assessment of the prostate capsule involvement. In these patients, T2W imaging confirms the lesion detected on DWI/ADC: on T2W imaging PCa appears as a hypointense focus, and recurrence appears after RT local as a mass-like or focal moderate hypointensity compared to the irradiated prostatic tissue, in the site of the lesion detected on DWI/ADC. For lesion localization both in the detection of PCa and after RT, a prostate 41-sectors map is used (14).

On T2W the local recurrence after RP appears as a mass-like or focal lesion slightly hyperintense to the pelvic muscle in the site of the lesion detected on DWI/ADC and that occurs anywhere in the prostatectomy bed, but most commonly at the vesico-urethral anastomosis, the retro-vesical space, the bladder neck, near the seminal vesicles bed, or adjacent to the vas deferens (15, 16). Local recurrence after RP can be located exactly on T2W, using as landmarks, the vesico-urethral clock on the axial T2W and the distance of the lesion from the lower margin of the pubic symphysis on the sagittal T2W images.

S-PI-RADS and PI-RRADS Categories bpMRI-based

S-PI-RADS and PI-RRADS provide minimum acceptable technical standards for MR image acquisition and suggest a dedicated structured method for bpMRI reporting.

S-PI-RADS assesses four categories and indicates the management for each of them. Lesions with extra-prostatic extension (EPE) and/or invasion of the seminal vesicle are included in category 4 in addition to intraglandular lesions (17, 18).

S-PI-RADS categories based bpMRI are reported in Table II.

PI-RRADS allows the detection of local recurrence after RP or RT in men with suspected BCR. Relapse after local therapy is defined by a rising PSA level >0.2 ng/ml following RP and >2 ng/ml above the nadir after RT (19, 20).

PI-RRADS includes four categories of bpMRI assessment of the likelihood of recurrence at the prostatectomy site after RP and in the prostate gland after RT. PI-RRADS assessment criteria, similarly to those of S-PI-RADS, are based firstly on DWI/ADC for lesion detection, size, and shape and then on anatomical T2W imaging for the lesion location in the prostatic bed after RP or intra-glandular site or extra-glandular extension after RT. PI-RRADS assessment categories and probability of recurrence are summarized in Table III.

In the initial diagnostic approach for PCa workup, and in post-treatment, T2W spectral presaturation with inversion recovery sequence of the the entire pelvis, in addition to DWI, is useful in the detection of lymph nodes and bone metastases, and other findings.