Abstract
The most recent edition of the prostate imaging reporting and data system (PI-RADS version 2) was developed based on expert consensus of the international working group on prostate cancer. It provides the minimum acceptable technical standards for MR image acquisition and suggests a structured method for multiparametric prostate MRI (mpMRI) reporting. T1-weighted, T2-weighted (T2W), diffusion-weighted (DWI), and dynamic contrast-enhanced (DCE) imaging are the suggested sequences to include in mpMRI. The PI-RADS version 2 scoring system enables the reader to assess and rate all focal lesions detected at mpMRI to determine the likelihood of a clinically significant cancer. According to PI-RADS v2, a lesion with a Gleason score ≥7, volume >0.5 cc, or extraprostatic extension is considered clinically significant. PI-RADS v2 uses the concept of a dominant MR sequence based on zonal location of the lesion rather than summing each component score, as was the case in version 1. The dominant sequence in the peripheral zone is DWI and the corresponding apparent diffusion coefficient (ADC) map, with a secondary role for DCE in equivocal cases (PI-RADS score 3). For lesions in the transition zone, T2W images are the dominant sequence with DWI/ADC images playing a supporting role in the case of an equivocal lesion.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
Prostate adenocarcinoma, with an estimated 180,890 new cases in 2016, is the most common noncutaneous malignancy and the second most common cause of cancer-related mortality in men in the United States [1]. There is significant biologic heterogeneity of prostate adenocarcinoma; some tumors are aggressive with high morbidity and mortality, while others remain indolent [2]. Given this range of clinical behavior, it is beneficial to the patient to attempt to prospectively differentiate clinically significant tumors (Gleason score ≥7, volume >0.5 cc, or extraprostatic extension) that would be candidates for therapy from clinically insignificant tumors that can safely undergo active surveillance.
Prostate adenocarcinoma is typically diagnosed using nontargeted random transrectal ultrasound (TRUS)-guided biopsy in the setting of an abnormal digital rectal examination or elevated prostate-specific antigen (PSA). Since TRUS-guided biopsy randomly samples the gland, it is possible that the biopsy results may not accurately detect the most aggressive tumor within the gland [3, 4]. Because of this potential for not detecting clinically significant cancer, new approaches for MR-defined targeted biopsy based upon improved MR imaging techniques have been proposed [5].
Traditionally, pelvic MRI has been used for treatment planning and monitoring with loco-regional staging of a biopsy-proven prostate cancer [6]. However, physiologic and functional sequences [diffusion-weighted imaging (DWI), dynamic contrast-enhanced imaging (DCE), and MR spectroscopy] are now added to the traditional sequences (T1- and T2-weighted images) to create current state of the art multiparametric MRI (mpMRI). This multiparametric approach expanded the role of MRI from only loco-regional staging to also include prostate adenocarcinoma detection, localization, and characterization. As mpMRI became more widely validated, the European Society of Urogenital Radiology (ESUR) developed the prostate imaging reporting and data system version 1 (PI-RADS v1) to provide standardized guidelines for image acquisition and interpretation [7]. PI-RADS v1 allowed for equal weighting for all acquired sequences (including MR spectroscopy), with a total maximum score of 20 (5 points for each sequence). PI-RADS v1 was demonstrated to have good accuracy, and moderate interobserver agreement with superior performance in the peripheral zone compared to the transition zone [8]. However, several investigators have noted that the use of MR spectroscopy and DCE sequences may not add significant value to the interpretation of mpMRI [9–11]. In part to address the shortcoming of PI-RADS v1 in the transition zone, and to reflect the concept of a dominant sequence to assess a lesion based on primary anatomic location within the prostate gland, an updated version of PI-RADS was proposed in early 2015 [12, 13]. This new version of PI-RADS has been validated in several studies and shown to have moderate interobserver variability [14, 15].
In this review, we provide an overview of PI-RADS v2 using a case-based approach and discuss and illustrate common pearls in the interpretation of mpMRI.
Background
PI-RADS v2 is designed to improve focal lesion detection, localization, characterization, and risk stratification in patients with suspected cancer and consists of technical recommendation for MRI acquisition and a scoring system for image interpretation. PI-RADS v2 uses a scale of 1–5 to report the overall probability of clinically significant prostate cancer on mpMRI. The use of PI-RADS v2 is limited to treatment naïve patients and it should not be used for staging, assessment of treatment outcome, recurrence, or progression during surveillance. The scoring algorithm is summarized in Table 1.
PI-RADS v2 recommends including T1W, T2W, high b-value DWI (≥1400) with corresponding ADC map, and DCE sequences in mpMRI (Table 2). Although a magnetic field strength of 3T is suggested, scanning at 1.5T with updated/optimal protocols would also result in acceptable image quality, and therefore, is not discouraged. The use of an endorectal coil is an institutional decision with the goal of obtaining optimal images.
Steps in lesion identification and scoring
To simplify the use of PI-RADS v2, we recommend utilizing a three-step approach: (1) Lesion localization based on PI-RADS v2 39-sector map (Fig. 1), (2) lesion measurement, and (3) location-based scoring (Table 1, Fig. 2). PI-RADS v2 uses an algorithmic approach to arrive at the overall score based on the zonal location of the lesion. DWI with corresponding ADC map is the dominant sequence in the peripheral zone that will determine the overall suspicion score (Fig. 3). There is a secondary role for DCE in the peripheral zone in equivocal cases (PI-RADS 3), where a positive DCE score can upgrade the lesion to a PI-RADS 4 (Fig. 4). T2W imaging is the dominant sequence in the transition zone that will determine the overall suspicion score (Fig. 5). There is a secondary role for DWI in the transition zone in equivocal cases (PI-RADS 3), where a large corresponding diffusion signal abnormality (PI-RADS 5, >1.5 cm) can upgrade the lesion to an overall PI-RADS 4. A lesion in the anterior fibromuscular stroma or central zone should be scored using the transition zone scoring algorithm. Additional example cases of PI-RADS 2 and 3 lesions are shown below (Figs. 6, 7, 8).
PI-RADS pearls
Measurement
The maximum lesion diameter is the key feature in discriminating between an overall PI-RADS suspicion score of 4 or 5. The ADC map is the preferred sequence for measuring peripheral zone lesions, whereas T2-weighted images are the suggested sequence to measure lesions in the transition zone. However, if a lesion is difficult to measure or obscured on the recommended sequence, measurement should be carried out using the sequence or plane with the best lesion conspicuity (Fig. 9).
DWI scoring
In the peripheral zone, evaluating the signal intensity of a lesion on high b-value DWI allows the reader to distinguish between PI-RADS DWI scores of 3 and ≥4. To assign a score of ≥4 for DWI, the lesion must be hyperintense on DWI with corresponding hypointensity on the ADC map (Fig. 3B, C). A lesion with low signal on ADC map but with isointensity or mild hyperintensity on DWI will receive a PI-RADS score of 3 (Fig. 4B, C).
T2W scoring
The dominant sequence in the transition zone is T2-weighted imaging. Clinically significant cancers (PI-RADS ≥4) are typically lenticular or noncircumscribed, homogeneous, and moderately hypointense on T2-weighted images. In contrast, equivocal tumors (PI-RADS 3) may be more heterogeneous on the T2-weighted images and have obscured margins. Benign findings in the transition zone (benign prostatic hyperplastic nodules) are circumscribed, while tumors traditionally have been described as having a “smudged charcoal” appearance. In fact, obscure margins, lack of capsule, and homogenous hypointensity are characteristics of a high score lesion (Fig. 10).
Conclusion
By using this simplified algorithmic approach, readers of varying levels of experience with prostate MRI should be able to confidently provide an overall suspicion score for each focal lesion detected at mpMRI.
PI-RADS v2 represents an improvement over PI-RADS v1 as it has increased the standardization of prostate MRI reporting by providing an overall assessment based on the appearance of a lesion on T2-weighted, DWI, and DCE sequences [15]. This clinically useful information, along with simplified image interpretation utilizing a binary score for DCE and the concept of a dominant sequence based on zonal anatomy, should hopefully increase utilization and lead to improved patient care [15]. Although it has only recently been introduced into routine clinical practice, several studies have evaluated the performance characteristics of PI-RADS v2 and these compare favorably to what has been achieved with PI-RADS v1 [16, 17]. In a single-center study of fiver readers, PI-RADS v2 had moderate agreement (κ = 0.46) for the overall suspicion score in a cohort of treatment naïve patients [14]. This same study also demonstrated an equivalent sensitivity and specificity in the transition zone and the peripheral zone, an improvement over PI-RADS v1 [14]. In a two-reader retrospective cohort study, Kasel-Siebert et al.' study had substantial inter-reader agreement (κ = 0.68) with slightly superior performance in the peripheral zone compared to the transition zone [18]. Finally, a more recent study utilizing six experienced readers at multiple centers demonstrated moderate reproducibility with improved agreement in the peripheral zone compared to the transition zone [19].
Similar to the other RADS (i.e., BI-RADS, LI-RADS, etc.), PI-RADS is a living document that is designed to be edited and revised in response to changes in technology and clinical practices. BI-RADS is currently in its 6th version and it is fully expected that version 3 of PI-RADS will appear within the next few years. One current issue already under discussion is that of gadolinium. Specifically—is it really needed, what is the risk, benefit, and is it worth the extra cost? In PI-RADS v1, the analysis of the gadolinium sequence involved full DCE curve evaluation (wash-in, wash out, and other PK parameters such as Ktrans) in all lesions. In PI-RADS v2, this has been considerably simplified to a binary decision of whether the lesion enhances before the normal adjacent prostatic tissue or not. As described in this review, the role of the DCE sequence in PI-RADS v2 is limited to assessment of focal lesions in the peripheral zone which are initially felt to be assessment score 3 by DWI. In this situation, if the lesion enhances (DCE+), it will lead to a change of assessment to an overall assessment 4. DCE is not used in any other situation, so it has a limited and thus questionable long-term viability as a critical must obtain sequence. A second area of increased interest is the option to include quantitative data to the current qualitative assessment. The most reported quantitative variable is the ADC from the b1400 or higher sequence. Hassanzadeh et al. have shown in early work the value of calculating and reporting the mean ADC [20].
PI-RADS v2 is the most recent guideline for prostate cancer mpMRI image acquisition and interpretation. The overall PI-RADS suspicion score of a lesion should convey the probability of a clinically significant cancer. Interpreting radiologists should be aware of common mimics of clinically significant cancer as well as imaging pearls to aid in cancer detection and diagnosis. Optimization of technical considerations and image acquisition will also improve study quality. In summary, PI-RADS v2 can be used to aid prostate mpMRI interpretation for cancer detection and risk stratification in treatment naïve patients with suspicious cancer.
References
Cancer Facts and Figures (2016) Atlanta: American Cancer Society; 2016 [cited 2016 2/11/2016]
Etzioni R, Penson DF, Legler JM, et al. (2002) Overdiagnosis due to prostate-specific antigen screening: lessons from US prostate cancer incidence trends. J Natl Cancer Inst 94(13):981–990
Schoots IG, Roobol MJ, Nieboer D, et al. (2015) Magnetic resonance imaging-targeted biopsy may enhance the diagnostic accuracy of significant prostate cancer detection compared to standard transrectal ultrasound-guided biopsy: a systematic review and meta-analysis. Eur Urol 68(3):438–450
Epstein JI, Feng Z, Trock BJ, Pierorazio PM (2012) Upgrading and downgrading of prostate cancer from biopsy to radical prostatectomy: incidence and predictive factors using the modified Gleason grading system and factoring in tertiary grades. Eur Urol 61(5):1019–1024
Penzkofer T, Tuncali K, Fedorov A, et al. (2015) Transperineal in-bore 3-T MR imaging-guided prostate biopsy: a prospective clinical observational study. Radiology 274(1):170–180
Rifkin MD, Zerhouni EA, Gatsonis CA, et al. (1990) Comparison of magnetic resonance imaging and ultrasonography in staging early prostate cancer. Results of a multi-institutional cooperative trial. N Engl J Med 323(10):621–626
Barentsz JO, Richenberg J, Clements R, et al. (2012) ESUR prostate MR guidelines 2012. Eur Radiol 22(4):746–757
Hamoen EH, de Rooij M, Witjes JA, Barentsz JO, Rovers MM (2015) Use of the prostate imaging reporting and data system (PI-RADS) for prostate cancer detection with multiparametric magnetic resonance imaging: a diagnostic meta-analysis. Eur Urol 67(6):1112–1121
Hansford BG, Peng Y, Jiang Y, et al. (2015) Dynamic contrast-enhanced MR imaging curve-type analysis: is it helpful in the differentiation of prostate cancer from healthy peripheral zone? Radiology 275(2):448–457
Platzek I, Borkowetz A, Toma M, et al. (2015) Multiparametric prostate magnetic resonance imaging at 3 T: failure of magnetic resonance spectroscopy to provide added value. J Comput Assist Tomogr 39(5):674–680
Tan CH, Hobbs BP, Wei W, Kundra V (2015) Dynamic contrast-enhanced MRI for the detection of prostate cancer: meta-analysis. AJR Am J Roentgenol 204(4):W439–W448
Weinreb JC, Barentsz JO, Choyke PL, et al. (2016) PI-RADS prostate imaging— reporting and data system: 2015, version 2. Eur Urol 69(1):16–40
Barentsz JO, Weinreb JC, Verma S, et al. (2016) Synopsis of the PI-RADS v2 guidelines for multiparametric prostate magnetic resonance imaging and recommendations for use. Eur Urol 69(1):41–49
Muller BG, Shih JH, Sankineni S, et al. (2015) Prostate cancer: interobserver agreement and accuracy with the revised prostate imaging reporting and data system at multiparametric MR imaging. Radiology 277(3):741–750
Rosenkrantz AB, Oto A, Turkbey B, Westphalen AC (2016) Prostate imaging reporting and data system (PI-RADS), version 2: a critical look. AJR Am J Roentgenol 206(6):1179–1183
Rosenkrantz AB, Kim S, Lim RP, et al. (2013) Prostate cancer localization using multiparametric MR imaging: comparison of prostate imaging reporting and data system (PI-RADS) and likert scales. Radiology 269(2):482–492
Schimmöller L, Quentin M, Arsov C, et al. (2013) Inter-reader agreement of the ESUR score for prostate MRI using in-bore MRI-guided biopsies as the reference standard. Eur Radiol 23(11):3185–3190
Kasel-Seibert M, Lehmann T, Aschenbach R, et al. (2016) Assessment of PI-RADS v2 for the detection of prostate cancer. Eur J Radiol 85(4):726–731
Rosenkrantz AB, Ginocchio LA, Cornfeld D, et al. (2016) Interobserver reproducibility of the PI-RADS version 2 Lexicon: a multicenter study of six experienced prostate radiologists. Radiology. doi:10.1148/radiol.2016152542
Hassanzadeh E, Olubiyi OI, Fedorov A, et al. (2016) Comparison of performance of quantitative ADC versus PI-RADS v2 assessment for differentiating high-grade from low-grade prostate cancer. ISMRM 24th Annual Meeting. Singapore
Acknowledgements
Grant funding provided by the National Institutes of Health Grant Number: U01CA151261 (FMF), the National Institutes of Health Grant Number: R25CA89017 (DIG), Massachusetts Department of Public Health Grant Number: DPH403516 (EH), and National Institutes of Health Grant Number: P41 EB015898 (CT, FMF).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Funding
This study was funded by Grant numbers U01CA151261 (FMF), R25CA89017 (DIG), DPH403516 (EH), and P41 EB015898 (CT, FMF).
Conflict of interest
The authors declare no conflict of interest.
Ethical approval
The study was HIPAA compliant and approved by the Institutional Review Board. Informed consent was waived because of the retrospective nature of the study.
Rights and permissions
About this article
Cite this article
Hassanzadeh, E., Glazer, D.I., Dunne, R.M. et al. Prostate imaging reporting and data system version 2 (PI-RADS v2): a pictorial review. Abdom Radiol 42, 278–289 (2017). https://doi.org/10.1007/s00261-016-0871-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00261-016-0871-z