Abstract
Magnetic resonance imaging (MRI) has shown a great potential in the evaluation and management of prostate cancer. In this study, we would like to evaluate the benefit of multiparametric MRI in the detection and localization of prostate cancer by comparing it with the gold standard of histopathology from radical prostatectomy. In this single-centre prospective study, 90 consecutive patients underwent radical prostatectomy from November 2016 to May 2018. All patients first underwent multiparametric (mp)-MRI, and all suspicious regions of interest were delineated and recorded on a 5-point scale as defined in prostate imaging reporting and data system version 2 (PI-RADS V2) score. All radical prostatectomy specimens, acquired after robotic radical prostatectomy with extended pelvic lymphadenectomy, were sent for histopathological examination (HPE). The mean age of the 90 patients was 65.3 years, and the mean serum prostate-specific antigen (PSA) was 16.9 ng/ml. The sensitivity and specificity of mp-MRI in the detection of the corresponding region of interest (ROI) on HPE were 67.4% and 89.3% respectively. Positive predictive value (PPV), negative predictive value (NPV), and accuracy of mp-MRI in the detection of corresponding ROI on HPE were 86.3%, 73.3%, and 78.3% respectively. The mp-MRI detected 96.8% solitary lesions and 61.7% multifocal lesions on the corresponding ROI on HPE. Multiparametric MRI has an excellent specificity and reasonable sensitivity for the diagnosis of prostate cancer. It is a good modality for the detection of solitary tumours, higher-grade tumours, detection of seminal vesicle invasion and extracapsular extension and helps in the decision-making process before radical prostatectomy, focal therapy or selecting an appropriate candidate for active surveillance.
Keywords: Multiparametric, Magnetic resonance imaging, Radical prostatectomy, Prostate cancer, PI-RADS V2 (prostate imaging reporting and data system version 2)
Introduction
Worldwide, more than 650,000 men are diagnosed with carcinoma prostate every year, accounting for a tenth of all new male cancers [1]. Traditionally, the diagnosis and staging of prostate cancer are based on transrectal ultrasound (TRUS) biopsy and computed tomography for pelvic lymphadenopathy. However, these diagnostic methods have several limitations like lower sensitivity and specificity for the detection and delineation of malignant foci and local staging of prostate cancer [2, 3].
Multiparametric magnetic resonance imaging (mp-MRI) is particularly helpful in areas poorly sampled by biopsies (part of base, extreme apex and preurethral compartment). It is postulated that by using mp-MRI for evaluation, the presence of a significant cancer can be eliminated due to the excellent specificity of the technique [4]. Although better than ultrasonography (USG), the use of MRI-guided biopsy (cognitive or TRUS-MRI fusion) for gathering information is based upon the premise that all significant carcinoma prostate will be visible on MRI, and thus, these lesions are targeted using MRI [5].
However, the experience with radical prostatectomy histology has shown that carcinoma prostate is often a multifocal disease and even small foci of high-grade tumours may have clinical implications in decision-making [6]. Therefore, it is important to evaluate whether MRI is able to show all the clinically significant cancer foci. This complete and accurate information about all the cancer foci is necessary to improve the treatment decision and their management [7].
In this study, we evaluated the benefit of mp-MRI (sensitivity, specificity and accuracy) in the detection and localization of prostate cancer by comparing it with the gold standard of histopathology from radical prostatectomy.
Material and Methods
In this single-centre prospective study, 90 consecutive patients underwent radical prostatectomy from November 2016 to May 2018. Inclusion criteria were patients with biopsy-proven carcinoma prostate, who opted for radical prostatectomy. Patients with metastatic disease and any contraindication to MRI (pacemaker, metallic implants, claustrophobia, known allergies to gadolinium-based contrast agents) were excluded. Informed consent was taken from all patients. This study was approved by the institute’s ethics committee.
All patients first underwent mp-MRI using 3 T Siemens Magnetom Trio (Siemens Medical Systems, Malvern, Pennsylvania, USA) system, following standard institutional protocol, which included T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) map and dynamic contrast-enhanced imaging (DCE). Intravenous injection of gadolinium contrast was given at a standard dose of 0.2 mg/kg. Sequences were performed in an orientation perpendicular to the rectal surface and in a serial order of 4 mm intervals from prostate apex to prostate base. An experienced radiologist with expertise in uroradiology reported all the cases blinded by any clinical or pathological information except patient age. The prostate gland was divided into 4 quadrants (right and left half then anterior and posterior), and each quadrant was divided into base, mid and apex. Thus, in each case, we had 12 regions of interest (ROI) within the whole prostate. All suspicious ROI delineated and recorded on a 5-point scale as defined in prostate imaging reporting and data system version 2 (PI-RADS V2) score: (1) probably normal, (2) low level of suspicion, (3) equivocal, (4) cancer probable and (5) definitely cancer. Areas with the same location on adjacent images are considered to belong to the same tumour.
All radical prostatectomy specimens acquired after robotic radical prostatectomy with extended pelvic lymphadenectomy were sent for histopathological examination (HPE). Specimens were inked for laterality and fixed in 40% buffered formalin overnight at room temperature. One pathologist with expertise in genitourinary pathology examined all the specimens. The pathologist kept unaware of mp-MRI report. Both seminal vesicles and vas were dissected from the specimen, and sections were taken separately. The weight of the prostate was measured, and then the specimen were sectioned into four quadrants (right anterior, right posterior, left anterior and left posterior). Each quadrant was then sectioned from apex to base at 4 mm intervals. The specimen was examined, and tumour mapping was done in 12 regions as recorded in MRI reporting (four quadrants into right and left half then anterior and posterior and each quadrant was divided into base, mid and apex). The site and percentage of tumour of each tumour foci were given separately. Gleason score and grade group were given for index and secondary tumours. Tumour with the highest Gleason grade is defined as an index tumour. If multiple tumours had the same grade, the largest tumour is considered an index tumour. The specimens were also examined for seminal vesicle invasion (SVI), extracapsular extension (ECE) and lymph nodes involvement (LNI).
On the MRI, the whole gland was divided into 3 equal transverse blocks from apex to base into base, mid and apex. The axial slices from every sequence were grouped accordingly in each of these segments, guided from the reference lines in the coronal plane. Each axial slice was further divided into right and left, and anterior and posterior. This resulted in 4 segments for the basal, middle and apical blocks respectively, that is, 12 evaluable segments per patient in total. To be considered a match, a focus of carcinoma must have been in the same quadrant of the right or left and anterior or posterior zone and must have been at the same apical, middle or basal level of the prostate in histopathology specimen. The lesions were characterized by the reviewers as true positives if the suspicious lesions were confirmed by histopathology as cancerous. Accordingly, lesions with no characteristics indicative of cancer in both imaging and histopathology were marked as true negatives, lesions evaluated as cancerous by imaging and proved to be benign were marked as false positives, and lesions considered benign by imaging and confirmed as cancerous by histopathology were marked as false negatives. Finally, for ECE, SVI and LNI, the correlation was based on the histopathological report, that is, whether there was a true match.
Statistical Analysis Plan
The descriptive analysis of quantitative data was expressed as means and standard deviation. Categorical/ordinal data was expressed as percentage, mean and range. Cross tables were generated to illustrate the relationship of results by different methods. Regions of interest on mp-MRI and histopathology for each patient were computed, and the percentage matching of scores was estimated. The association between the two measures was tested using the chi-square test. Sensitivity, specificity, positive predictive value and negative predictive value were calculated. P-value < 0.05 was considered statistically significant. SPSS software, version 24.0, was used for analysis.
Results
The mean age of the 90 patients was 65.3 ± 6.9 (range, 36–77) years, and the mean serum PSA was 16.9 ± 14.4 (range 2–77.2) ng/ml. The majority (84.4%) of patients were in D’Amico high-risk and locally advanced groups, and 14 (15.6%) patients were in intermediate-risk (Table 1).
Table 1.
Patients’ characteristics
| Clinical characteristics | Mean ± SD (range)/n (%) |
|---|---|
| Number of patients (n) | 90 |
| Mean age ± SD (range) years | 65.3 ± 6.9 (36–77) |
| Mean serum PSA ± SD (range) ng/ml | 16.9 ± 14.4 (2–77.2) |
| Family history of prostate cancer (%) | 3 (3.3%) |
| Abnormal DRE (%) | 43 (47.8%) |
|
D’Amico risk stratification groups Intermediate risk High risk Locally advanced |
14 (15.6%) 38 (42.2%) 38 (42.2%) |
| Mean time from mp-MRI to radical prostatectomy (days) | 36.63 ± 28.001 (1–129) |
| Mean volume of prostate on mp-MRI (gm) | 45.4 ± 26 (12.4–170.4) |
| Mean volume of prostate on radical prostatectomy HPE (gm) | 44.8 ± 23.3 (9.4–140.9) |
DRE digital rectal examination, PSA prostate specific antigen, mp-MRI multiparametric magnetic resonance imaging, SD standard deviation
The mean time from mp-MRI to radical prostatectomy was 36.6 ± 28 (range 1–129) days. Of the 90 patients, 64 (71.1%), 22 (24.4%), 3 (3.3%) and 1 (1.2%) patients had 1, 2, 3 and 4 MRI suspicious lesions respectively. Of the 120 MRI suspicious lesions, 90 were index lesions, and 30 were non-index lesions. PIRADS score 2, 3, 4 and 5 lesions accounted for 9 (7.5%), 33 (27.5%), 36 (30%) and 42 (35%) respectively. On mp-MRI, seminal vesicle invasion was present in 20 (22.2%), extra-capsular extension in 27 (30%) and lymph node involvement in 9 (10%) patients.
Mean prostate volume on mp-MRI was 45.4 ± 26 g (range, 12.4–170.4), and on radical prostatectomy, HPE was 44.8 ± 23.3 g (range, 9.4–140.9). The correlation coefficient (r = 0.969) and p-value (p < 0.0001) were statistically significant.
Out of total 120 PIRADS lesions identified on mp-MRI, 104 (86.67%) were detected on the corresponding ROI on HPE. The detection of PIRADS scores 2, 3, 4 and 5 lesions were 44.4% (4/9), 78.8% (26/33), 88.9% (32/36) and 100% (42/42) respectively on corresponding ROI on HPE (Fig. 1).
Fig. 1.
PIRADS score wise detection of lesions on HPE (PI-RADS prostate imaging reporting and data system, HPE histopathological examination)
Of the total 132 lesions on HPE, 102 (77.27%) were of Gleason grade 2 or more. Out of these 102 lesions, 97 (95.1%) were detected on the corresponding ROI on mp-MRI. The detection of PIRADS 2, 3, 4 and 5 lesions for Gleason grade 2 or more were 80% (4/5), 89.3% (25/28), 96.8% (30/31) and 100% (38/38) respectively. Of the 132 lesions on HPE, 64 were solitary, and 68 were multifocal. The mp-MRI detected 96.8% (62/64) solitary lesions and 61.7% (42/68) multifocal lesions on the corresponding ROI on HPE. Overall, 85.6% (77/90) patients had all tumour foci detected on mp-MRI, including 96.8% (62/64) solitary and 57.7% (15/26) multifocal lesions.
Out of 1080 regions of interest (ROI), mp-MRI was positive for 422 ROI and negative for 658 ROI, while radical prostatectomy histopathology was positive for 540 ROI and negative for 540 ROI (Table 2). The sensitivity and specificity of mp-MRI in the detection of corresponding ROI on HPE were 67.4% (95% CI: 63.3–71.4%) and 89.3% (95% CI: 86.3–91.7) respectively. PPV, NPV and accuracy of mp-MRI in the detection of corresponding ROI on HPE were 86.3% (95% CI: 83.0–88.9), 73.3% (95% CI: 70.7–75.6) and 78.3% (95% CI: 75.8–80.8) respectively.
Table 2.
Correlation between mp-MRI and HPE in the detection of lesion in ROI
| Multiparametric magnetic resonance imaging | Radical prostatectomy HPE | |
|---|---|---|
| Positive for cancer (n = 540) | Negative for cancer (n = 540) | |
| Positive for cancer (n = 422) | 364 | 58 |
| Negative for cancer (n = 658) | 176 | 482 |
Kappa = 0.567; chi-square = 364.190; p-value = < 0.0001
HPE histopathological examination, mp-MRI multiparametric magnetic resonance imaging, ROI region of interest
Out of 1080 ROI, the mp-MRI index lesion was positive for 329 ROI and negative for 751 ROI, while radical prostatectomy histopathology was positive for 297 ROI and negative for 783 ROI (Table 3). The sensitivity and specificity for index lesion detection on mp-MRI were 77.4% (95% CI: 72.3–82.0) and 87.4% (95% CI: 84.8–89.6) respectively. The PPV, NPV and accuracy of mp-MRI for index lesion detection were 69.9% (95% CI: 65.7–73.8), 91.0% (95% CI: 89.2–92.7) and 84.6% (95% CI: 82.3–86.7) respectively.
Table 3.
Correlation between mp-MRI and HPE in detection of index lesion in ROI
| Multiparametric magnetic resonance imaging | Radical prostatectomy HPE | |
|---|---|---|
| Positive for cancer (n = 297) | Negative for cancer (n = 783) | |
| Positive for cancer (n = 329) | 230 | 99 |
| Negative for cancer (n = 751) | 67 | 684 |
Kappa = 0.627; chi-square = 426.797; p-value = < 0.0001
HPE histopathological examination, mp-MRI multiparametric magnetic resonance imaging, ROI region of interest)
On radical prostatectomy HPE, seminal vesicle invasion, extra-capsular extension and lymph node involvement were seen in 26 (28.9%), 27 (30%) and 13 (14.4%) patients respectively.
Of the 26 SVI seen on HPE, mp-MRI detected 19 involvements and out of 64 patients where SVI was not in HPE, mp-MRI detected 1 lesion. The mp-MRI correctly identified no SVI in 63 patients. The strength of agreement is considered to be ‘good’ (Kappa = 0.567, 95% CI: 0.617–0.919). The sensitivity, specificity, PPV, NPV and accuracy of mp-MRI for the detection of SVI were 73.08% (95% CI: 52.2–88.4), 98.4% (95% CI: 91.6–99.9), 95% (95% CI: 72.8–99.3), 90% (95% CI: 82.7–94.4) and 91.1% (95% CI: 82.2–96.0) respectively.
Of the 27 ECE detected on HPE, mp-MRI detected 21 and missed 6. Of the 63 negative ECE on HPE, mp-MRI correctly ruled out 57 and falsely detected 6 ECE. The strength of agreement is considered to be ‘good’ (Kappa = 0.683, 95% CI: 0.517–0.848). The sensitivity, specificity, PPV, NPV and accuracy of mp-MRI for the detection of ECE were 77.8% (95% CI: 57.7–91.4), 90.5% (95% CI: 80.4–96.4), 77.8% (95% CI: 61.4–88.5), 90.5% (95% CI: 82.4–95.0) and 86.7% (95%CI: 77.8–92.9) respectively.
LNI on final HPE was present in 14.4% of patients. There was a positive weak correlation of LNI between mp-MRI and radical prostatectomy with bilateral extended pelvic lymphadenectomy HPE findings (r = 0.244, p value = 0.021).
Discussion
MRI is a non-invasive diagnostic tool and has shown a great potential in the evaluation and management of prostate cancer [8–12]. However, the experience with radical prostatectomy histology has shown that carcinoma prostate is often a multifocal disease and even small foci of high-grade tumours may have clinical implications in decision-making [6]. Therefore, it is important to evaluate whether MRI is able to show all the clinically significant cancer foci. This complete and accurate information about all the cancer foci is necessary to improve treatment decision-making and their management [7].
In 2015, ESUR proposed PI-RADS V2 for the detection of prostate cancer, in which MR spectroscopy is excluded from the previous version, and report is given on a 5 points scale by using a dominant parameter, i.e. DWI for peripheral zone and T2WI for transition zone, while DCE as a positive or negative. In PI-RADS V2, prostate is divided into 39 sectors/regions, in which 36 sectors are for prostate, 2 for the seminal vesicles and 1 for the external urethral sphincter [13]. Typical picture of carcinoma prostate on mp-MRI is hypointense on T2WI and restricted diffusion with high signal intensity on DWI with hypointense on ADC map [8].
A higher proportion of preoperative Gleason grade group, clinical stage T and D’Amico risk stratification groups in our study are most probably due to a lack of awareness for PSA screening, late detection and delayed presentation of cancer prostate in developing countries. More patients with solitary tumours and higher PIRADS scores in our study impacted the final results, as the ability of mp-MRI increases significantly in the detection of solitary lesions and high PIRADS scores [4, 14].
In present study, on the evaluation of corresponding ROI, 86.7% of the lesions identified on mp-MRI showed prostate cancer on HPE. The detection of PIRADS scores 2, 3, 4 and 5 lesions were 44.4%, 78.8%, 88.9% and 100% respectively on corresponding ROI on HPE. On comparison between mp-MRI and histopathology, it was found that 91.11% index lesions and 40% non-index lesions on mp-MRI showed tumour on corresponding ROI on HPE. On comparison of ROI with HPE, mp-MRI correctly identified 95.1% lesions of clinically significant tumours with the Gleason grade group ≥ 2. Le DJ et al. retrospectively correlated multifocality detection of cancer prostate with prostatectomy HPE in 122 patients. In their study, the detection of PIRADS 3, 4 and 5 lesions on radical prostatectomy histopathology were 65%, 96% and 95% respectively [4]. The higher detection of PIRADS 5 lesion on mp-MRI in our study is most probably due to large and more diffusely involving lesions.
In our study, sensitivity and specificity of mp-MRI in the detection of corresponding ROI on radical prostatectomy HPE were 67.4% and 89.3% respectively. The PPV, NPV and accuracy of mp-MRI in the detection of corresponding ROI on HPE were 86.3%, 73.3% and 78.3% respectively. We observed higher sensitivity of mp-MRI as compared to other studies; it may be due to the use of 1.5 T MRI with or without endorectal coil in other studies. Loggitsi D et al. studied results of mp-MRI on 27 patients and correlated with prostatectomy specimen according to ROI and found out sensitivity, specificity, PPV and NPV as 53%, 90%, 58% and 88% respectively for combined use of T2WI, DWI and DCE [15]. Peuch P et al. evaluated 83 patients by dividing prostate into 8 ROI and correlated mp-MRI findings with histopathology and observed that sensitivity, specificity, PPV and NPV of mp-MRI for the detection of cancer foci were 32%, 95%, 76% and 75% respectively [16].
In present study, sensitivity, specificity, PPV, NPV and accuracy of mp-MRI for the detection of SVI were 73.08%, 98.4%, 95%, 90% and 91.1% respectively. Loggitsi D et al. found out sensitivity, specificity and NPV of mp-MRI for SVI were 79%, 100% and 96% respectively [15]. Grivas N et al. did a retrospective study of 527 patients for staging accuracy of mp-MRI and concluded that SVI was present in 12.5% of patients on mp-MRI. The sensitivity, specificity, PPV and NPV for the detection of SVI on mp-MRI were 75.9%, 94.7%, 62% and 97% respectively [17].
The sensitivity, specificity, PPV, NPV and accuracy of mp-MRI for the detection of ECE were 77.8%, 90.5%, 77.8%, 90.5% and 86.7% respectively. De Rooij M et al. did a meta-analysis of 45 studies to assess the diagnostic accuracy of MRI for the detection of ECE. Pooled data sensitivity and specificity for ECE (5681 patients) were 57% (95% CI, 49–64%) and 91% (95% CI, 88–93%) respectively [18].
Lymph node involvement on final HPE was present in 14.4% of patients. There was a positive weak correlation of LNI between mp-MRI and radical prostatectomy with bilateral extended pelvic lymphadenectomy HPE findings (r = 0.244, p value = 0.021). Hovels AM et al. also got similar results on a meta-analysis of 24 studies to assess the diagnostic accuracy of CT and MRI in the staging of pelvic lymph nodes in patients with prostate cancer. For MRI, the pooled sensitivity was 0.39 (0.22–0.56 95% CI), and the pooled specificity was 0.82 (0.79–0.83 95% CI). They concluded that MRI demonstrates poor performance in the detection of lymph node metastases [19].
The limitation of this study was single-centre and non-randomized in nature with a relatively small sample size. The majority of our patients had high volume and high-stage disease; therefore, the finding needs confirmation in a situation where there is low volume and early prostate cancer. Our study did not use whole mount section for comparison and determination of concordance between MRI finding and histopathology. This study is only a direct comparison of MRI with histopathology and has no follow-up information about the oncologic outcome, thus does not address whether tumour detection is associated with oncologic outcome.
Conclusion
Multiparametric MRI has an excellent specificity (89.3%) and reasonable sensitivity (67.4%) for diagnosis of prostate cancer. It is a good modality for the detection of solitary tumours, higher-grade tumours, detection of seminal vesicle invasion and extracapsular extension and helps in the decision-making process before radical prostatectomy, focal therapy or selecting an appropriate candidate for active surveillance.
Declarations
Conflict of Interest
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin. 2014;64:9. doi: 10.3322/caac.21208. [DOI] [PubMed] [Google Scholar]
- 2.Guneyli S, Erdem CZ, Erdem LO. Magnetic resonance imaging of prostate cancer. Clin Imaging. 2016;40:601–609. doi: 10.1016/j.clinimag.2016.02.011. [DOI] [PubMed] [Google Scholar]
- 3.Barentsz JO, Richenberg J, Clements R, Choyke P, Verma S, Villeirs G, et al. ESUR Prostate MR Guidelines 2012. Eur Radiol. 2012;22(4):746–757. doi: 10.1007/s00330-011-2377-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Le JD, Tan N, Shkolyar E, Lu DY, Kwan L, Marks LS, et al. Multifocality and prostate cancer detection by multiparametric magnetic resonance imaging: correlation with whole-mount histopathology. Eur Urol. 2015;67:569–576. doi: 10.1016/j.eururo.2014.08.079. [DOI] [PubMed] [Google Scholar]
- 5.Sarkar D. The role of multi-parametric MRI and fusion biopsy for the diagnosis of prostate cancer – a systematic review. J Pros Canc. 2016;1:2. doi: 10.1007/978-3-319-95693-0_7. [DOI] [PubMed] [Google Scholar]
- 6.Jesse DL, Nelly T, Eugene S, David YL, Lorna K, Leonard SM, et al. Multifocality and prostate cancer detection by multiparametric magnetic resonance imaging: correlation with whole-mount histopathology. Eur Urol. 2015;67:569–576. doi: 10.1016/j.eururo.2014.08.079. [DOI] [PubMed] [Google Scholar]
- 7.Moosavi B, Flood TA, Al-Dandan O, Breau RH, Cagiannos I, Morash C, et al. Multiparametric MRI of the anterior prostate gland: clinical-radiological-histopathological Correlation. Clin Radiol. 2016;71:405–417. doi: 10.1016/j.crad.2016.01.002. [DOI] [PubMed] [Google Scholar]
- 8.Ghai S, Haider MA. Multiparametric-MRI in diagnosis of prostate cancer. Indian J Urol. 2015;31(3):194–201. doi: 10.4103/0970-1591.159606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Turkbey B, Mani H, Shah V, Rastinehad AR, Bernardo M, Pohida T, et al. Multiparametric 3T prostate magnetic resonance imaging to detect cancer: histopathological correlation using prostatectomy specimens processed in customized magnetic resonance imaging based molds. J Urol. 2011;186:1818–1824. doi: 10.1016/j.juro.2011.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Turkbey B, Albert PS, Kurdziel K, Choyke PL. Imaging localized prostate cancer: current approaches and new developments. AJR Am J Roentgenol. 2009;192:1471. doi: 10.2214/AJR.09.2527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Roethke MC, Kuru TH, Schultze S, Tichy D, Fenchel M, Schlemmer HP, et al. Evaluation of the ESUR PI-RADS scoring system for multiparametric MRI of the prostate with targeted MR/TRUS fusion-guided biopsy at 3.0 Tesla. Eur Radiol. 2014;24:344–352. doi: 10.1007/s00330-013-3017-5. [DOI] [PubMed] [Google Scholar]
- 12.Thompson JE, Van LPJ, Moses D, Shnier R, Brenner P, Delprado W, et al. The diagnostic performance of multiparametric magnetic resonance imaging to detect significant prostate cancer. J Urol. 2016;195:1428–1435. doi: 10.1016/j.juro.2015.10.140. [DOI] [PubMed] [Google Scholar]
- 13.Prostate Imaging Reporting and Data System (PI-RADS) [Internet] Reston (VA): American college of radiology; [cited 2015 Mar 5]. Available from: http://www.acr.org/Quality-Safety/Resources/PIRADS/
- 14.Huang S, Mann S, O’Sullivan R, Ryan A, Landau A, Snow R, et al. Correlation between multiparametric MRI findings and radical prostatectomy specimens in 162 patients. J Urol. 2016;185(4s):e393. [Google Scholar]
- 15.Loggitsi D, Gyftopoulos A, Economopoulos N, Apostolaki A, Kalogeropoulos T, Thanos A, et al. Multiparametric magnetic resonance imaging of the prostate for tumour detection and local staging: imaging in 1.5T and histopathologic correlation. Can Assoc Radiol J. 2017;68:379–386. doi: 10.1016/j.carj.2017.02.003. [DOI] [PubMed] [Google Scholar]
- 16.Peuch P, Potiron E, Lemaitre L, Leroy X, Haber GP, Crouzet S, et al. Dynamic contrast-enhanced-magnetic resonance imaging evaluation of intraprostatic prostate cancer: correlation with radical prostatectomy specimens. J Urol. 2009;74(5):1094–1099. doi: 10.1016/j.urology.2009.04.102. [DOI] [PubMed] [Google Scholar]
- 17.Grivas N, Hinnenb K, Jongc JD, Heemsbergend W, Moonend L, Witteveen T, et al. Seminal vesicle invasion on multi-parametric magnetic resonance imaging: correlation with histopathology. Eur J Radiol. 2018;98:107–112. doi: 10.1016/j.ejrad.2017.11.013. [DOI] [PubMed] [Google Scholar]
- 18.De Rooij M, Hamoen EH, Witjes JA, Barentsz JO, Rovers MM. Accuracy of magnetic resonance imaging for local staging of prostate cancer: a diagnostic meta-analysis. Eur Urol. 2016;70:233. doi: 10.1016/j.eururo.2015.07.029. [DOI] [PubMed] [Google Scholar]
- 19.Hövels AM, Heesakkers RA, Adang EM, Jager GJ, Strum S, Hoogeveen YL, et al. The diagnostic accuracy of CT and MRI in the staging of pelvic lymph nodes in patients with prostate cancer: a meta-analysis. Clin Radiol. 2008;63:387. doi: 10.1016/j.crad.2007.05.022. [DOI] [PubMed] [Google Scholar]