Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

False positives in PIRADS (V2) 3, 4, and 5 lesions: relationship with reader experience and zonal location

False positives in PIRADS (V2) 3, 4, and 5 lesions: relationship with reader experience and zonal... Purpose To investigate the effect of reader experience and zonal location on the occurrence of false positives (FPs) in PIRADS (V2) 3, 4, and 5 lesions on multiparametric (MP)-MRI of the prostate. Materials and methods This retrospective study included 139 patients who had consecutively undergone an MP-MRI of the prostate in combination with a transrectal ultrasound MRI fusion-guided biopsy between 2014 and 2017. MRI exams were prospectively read by a group of inexperienced radiologists (cohort 1; 54 patients) and an experienced radiologist (cohort 2; 85 patients). Multivariable logistic regression analysis was performed to determine the association of experience of the radiologist and zonal location with a FP reading. FP rates were compared between readings by inexperienced and experienced radiologists according to zonal location, using Chi-square (χ ) tests. Results A total of 168 lesions in 139 patients were detected. Median patient age was 68 years (Interquartile range (IQR) 62.5–73), and median PSA was 10.9 ng/mL (IQR 7.6–15.9) for the entire patient cohort. According to multivariable logis- tic regression, inexperience of the radiologist was significantly (P = 0.044, odds ratio 1.927, 95% confidence interval [CI] 1.017–3.651) and independently associated with a FP reading, while zonal location was not (P = 0.202, odds ratio 1.444, 95% CI 0.820–2.539). In the transition zone (TZ), the FP rate of the inexperienced radiologists 59% (17/29) was significantly higher (χ P = 0.033) than that of the experienced radiologist 33% (13/40). Conclusion Inexperience of the radiologist is significantly and independently associated with a FP reading, while zonal location is not. Inexperienced radiologists have a significantly higher FP rate in the TZ. Keywords MP-MRI · Prostate cancer · PIRADS V2 · Reader experience · Zonal location · False positives Introduction Multiparametric (MP) MRI of the prostate is the best option for local diagnosis of prostate cancer (PCa) [1–3]. It can be useful in various clinical settings, such as for detection or staging purposes, guiding biopsies, detection of local * Derya Yakar recurrence, and as a tool to select candidates for active [email protected] surveillance. A negative transrectal-ultrasound (TRUS)-guided pros- Department of Radiology, University Medical Center Groningen, Hanzeplein 1, 9700 CC Groningen, tate biopsy with persistent clinical suspicion of PCa is still The Netherlands the most frequent reason to perform MP-MRI in clinical Department of Urology, University Medical Center practice. Together with the introduction of the commer- Groningen, Hanzeplein 1, 9700 CC Groningen, cially available TRUS-MRI fusion-guided prostate biopsy The Netherlands systems, the use of MP-MRI of the prostate has increased Department of Urology, Erasmus MC, Dr. Molewaterplein tremendously over the last years. However, along with 40, 3015 GD Rotterdam, The Netherlands the increasing popularity of this diagnostic test, serious Department of Urology, University Medical Center Utrecht, concerns on quality issues have been raised [4, 5]. These Heidelberglaan 100, 3584 CX Utrecht, The Netherlands Vol:.(1234567890) 1 3 Abdominal Radiology (2019) 44:1044–1051 1045 quality concerns include image acquisition, interpretation, Materials and methods and reporting. The Prostate Imaging Reporting and Data System (PIRADS) has been introduced to address some of Patients these aforementioned issues. Nonetheless, even with the use of the PIRADS version (V) 2 system, interreader vari- This retrospective study was approved by the local ethics ability is still a common present-day problem [6]. Reader committee, and the need for informed consent was waived experience is probably an important issue in this setting. (registration number 201700780). All patients (n = 186) Previous studies have already emphasized the importance who had consecutively undergone MP-MRI of the prostate of subspecialty reading in prostate MRI [7, 8]. Strongly in combination with a TRUS-MRI fusion-guided prostate related to this subject are the pitfalls encountered on MP- biopsy between 2014 and 2017 were potentially eligible MRI of the prostate. Awareness of the causes of false for inclusion. MP-MRI scans of the prostate were either positives (FPs) can theoretically improve the diagnostic obtained from in-house records or from referring centers. performance of the radiologist and decrease interreader Patients whose MRI scans were reported before the intro- variability. Besides the experience of the radiologist, zonal duction of standard reporting according to PIRADS version location can also be a source of FPs [9]. For example, 2 [14] (n = 38) were excluded. An additional number of 9 detection of PCa in the transition zone (TZ) is often per- patients were excluded because of missing crucial informa- ceived as a difficult task. Pictorial reviews and case series tion, such as a PIRADS score (n = 7), inconclusive histopa- have touched on the topic of FPs [10–12], but systematic thology report (n = 1), and a history of radiotherapy (n = 1). studies specifically investigating FPs in a pure clinical set- Finally, 139 unique patients were included in this study, of ting are scarce and from the pre-PIRADS era [13]. We whom 54 patients were prospectively read by the inexpe- hypothesized that, when not educated properly, the novice rienced radiologists (cohort 1), and 85 patients were read reader will, in daily clinical practice, cause unnecessary by the experienced radiologist (cohort 2) (Fig. 1). 99/139 biopsies especially in the TZ. patients had undergone prostate MP-MRI because of a pre- Therefore, the purpose of this study was to investigate vious history of negative TRUS-guided prostate biopsy and the effects of reader experience and zonal location on the persistent clinical suspicion (based on PSA levels and/or occurrence of FPs in PIRADS (V2) 3, 4, and 5 lesions on abnormal digital rectal examination). 41/139 had undergone MP-MRI of the prostate, with targeted TRUS-MRI fusion- prostate MP-MRI for staging purposes because of increasing guided prostate biopsy as reference standard. PSA levels, while being on an active surveillance program (inclusion of only low-grade cancers according to PRIAS Fig. 1 Flowchart showing the numbers of potentially eligible patients, excluded patients, included patients, and patients scored by inexperienced and experienced radiologists 1 3 1046 Abdominal Radiology (2019) 44:1044–1051 guidelines (Prostate cancer Research International: Active Targeted biopsies were performed in all patients, while surveillance) [15]). two patients also received an additional systematic biopsy. Biopsy procedures were undertaken by four different urolo- MRI acquisition and analysis gists (with 4, 4, 1, and 1 year(s) of experience with fusion biopsy, respectively). The number of cores per lesion was at 96 patients were referred from outside hospitals (total of the discretion of the urologist performing the biopsy. 9 different institutions), while 43 patients were scanned in our institution. In total, 88 exams were performed on a 1.5T Histopathology scanner (of which none was performed with an endorectal coil) and 51 exams were performed on a 3T MRI scanner The targeted biopsies were analyzed by a specialist uro- (of which 10 with an endorectal coil). Of the 51 exams per- pathologist according to the International Society of Urolog- formed on a 3T scanner, five were primarily scanned on a ical Pathology (ISUP) 2014 recommendations [16]. Biopsy 1.5T scanner in outside hospitals. Due to technical limita- results containing cancer (Gleason ≥ 3 + 3) were categorized tions (e.g., motion artifacts, inadequate quality), these were as true positives, while the ones containing no cancer were re-obtained in our institution. The MP-MRI examinations of categorized as FPs. the prostate of 131 patients comprised an axial T1-weighted image, a high-resolution multiplanar T2-weighted image Statistical analysis (slice thickness of 3 mm), a diffusion-weighted image (DWI) with at least three b-values (varying between 0-2000 s/mm 2 The Shapiro–Wilk’s test was used to test if the continuous and with a minimum highest value of 800 s/mm ) and a variables age, PSA, PSA density, prostate volume, number calculated apparent diffusion coefficient (ADC) map, and of cancer-suspicious lesions per patient, and number of a dynamic contrast-enhanced sequence. The remaining cores taken per patient, were normally distributed. These 8 patients had undergone the same sequences, except for variables were then compared between the patient cohort a dynamic contrast-enhanced (DCE) sequence. All MRI read by the inexperienced radiologists (cohort 1) and the examinations were read at our tertiary referral center. Radi- patient cohort read by the experienced radiologist (cohort ologists were not blinded to clinical data or external reports, 2), using the unpaired t test for normally distributed data, completely in line with clinical practice. MRI data were read and the Mann–Whitney U test for not normally distributed prospectively by four different radiologists, with different data. The FP rate was calculated as the number of biopsied levels of experience in MP-MRI prostate reporting. One lesions containing no PCa divided by the total number of trained genitourinary radiologist (with 5 years of experi- biopsied lesions. FP rate readings were compared between ence and > 500 case readings with histopathologic correla- inexperienced and experienced radiologists according to tion) was defined as the experienced reader, while all other zonal location, using Chi-square (χ ) tests. Multivariable three radiologists (with 1–2 years of experience and < 100 logistic regression analysis was performed to determine case readings without histopathologic correlation) were the association of the experience of the radiologist (inex- categorized as inexperienced readers. The MRI scans were perienced vs. experienced) and zonal location (TZ vs. PZ) reported by using PIRADS V2, with PIRADS 1–5 represent- with a FP reading. P-values less than 0.05 (two-sided) were ing an incremental scoring system of very low likelihood to regarded statistically significant. All statistical analyses were very high likelihood of clinically significant PCa [14]. All performed using MedCalc Statistical Software version 18.5 prostate MRI reports were analyzed for reader experience, (MedCalc Software bvba, Ostend, Belgium). PIRADS score, and zonal location (peripheral zone (PZ), TZ, or central zone (CZ)). Furthermore, the FP lesions were retrospectively investigated by the experienced radiologist for other sources of error such as known mimickers of PCa Results (e.g., prostatitis, anatomic pitfalls) or possible technical ina- curacies (e.g., needle delivery, fusion misregistration). The 139 included patients had a total of 168 PIRADS 3-5 lesions. Median patient age was 68 years (interquartile range TRUS‑MRI fusion‑guided prostate biopsy (IQR) 62.5–73) and median PSA was 10.9  ng/mL (IQR 7.6–15.9) for the entire patient cohort. Age, PSA, PSA den- All patients with at least one lesion with a PIRADS score sity, prostate volume, number of cancer-suspicious lesions of 3 or higher were biopsied according to our institutional per patient, and number of cores taken per patient were not standard, with up to a maximum of three biopsied lesions significantly different (P > 0.101) between the patient cohort per patient. The Dynacad Uronav fusion biopsy system read by the inexperienced radiologists and the patient cohort (Invivo, Gainesville, Florida, USA) was used for all biopsies. read by the experienced radiologist (Table 1). 1 3 Abdominal Radiology (2019) 44:1044–1051 1047 Table 1 Characteristics of the c Patient cohort 1 read by inexpe- Patient cohort 2 read by experi- P value included patients for the two rienced radiologists enced radiologist groups of radiologists Total number of patients 54 Total number of patients 85 Total number of lesions 66 Total number of lesions 102 a b Age (years) 70 (65–73) 67 ± 7.0 0.101 a a PSA (ng/mL) 11.6 (7.7–17.0) 10 (7.3–14.2) 0.155 b b PSA density 0.20 ± 0.10 0.19 (0.14–0.28) 0.448 a a Prostate volume (cc) 69 (51–78) 50 (40–82) 0.303 a a Number of cancer-suspi- 1 (1–1) 1 (1–1) 0.191 cious lesions per patient a a Number of cores taken per 3 (3–4) 3 (3–3) 0.824 suspicious lesion Data presented as median (interquartile range) (not normally distributed according to Shapiro–Wilk test) Data presented as mean ± standard deviation (normally distributed according to Shapiro–Wilk test) According to Mann–Whitney U test (for not normally distributed independent samples) Table 2 Multivariable logistic regression: relationship of experience Table 3 FP rate according to reader experience for zonal location and and zonal location with FP reading PIRADS category Variable Odds ratio (95% CI) P value Variable FP rate inexpe- FP rate experi- P value rienced readers enced reader Inexperience 1.927 (1.017-3.651) 0.044 Zone Zonal location (transition and 1.444 (0.820-2.539) 0.202  Transition 59% (17/29) 33% (13/40) 0.033 peripheral zone)  Peripheral 43% (16/37) 29% (17/58) 0.164 CI confidence interval PIRADS  3 100% (11/11) 71% (17/24) 0.049  4 44% (16/36) 27% (13/49) 0.105 The total numbers of PIRADS 3–5 lesions called by the  5 32% (6/19) 10% (3/29) 0.059 inexperienced radiologists and the experienced radiologist Overall (PIRADS 3–5) 50% (33/66) 32% (33/102) 0.020 were 66 (in 54 patients, cohort 1) and 102 (in 85 patients, a 2 According to χ test cohort 2), respectively. 110 patients had one lesion, 23 patients had two lesions, and 4 patients had three lesions. The inexperienced radiolo- gists called 29 lesions in the TZ (7 PIRADS 3, 16 PIRADS the inexperienced radiologists, and 32% for the experienced 4, 6 PIRADS 5) and 37 lesions in the PZ (4 PIRADS 3, radiologist (P = 0.020). The FP rate in the TZ for the inex- 20 PIRADS 4, 13 PIRADS 5). The experienced radiologist perienced radiologists (59%, 17/29) was significantly higher called 40 lesions in the TZ (13 PIRADS 3, 11 PIRADS 4, (P = 0.033) than that for the experienced radiologist (33%). 16 PIRADS 5) and 58 lesions in the PZ (7 PIRADS 3, 38 On the other hand, the FP rate in the PZ for the inexperi- PIRADS 4, 13 PIRADS 5). Only the experienced radiologist enced radiologists (43%, 16/37) was not significantly dif - called four lesions in the central zone (CZ). Therefore, the ferent (P = 0.164) from that for the experienced radiologist CZ was excluded from further analysis. In total, 102 biopsies (29%). Corresponding results are also displayed in Table 3. (61%) showed PCa (of which 60 (59%) contained clinically After a further breakdown of the 17 FP lesions in the significant cancer, defined as ≥ 3 + 4 Gleason score), while TZ for the inexperienced radiologists, all were, retrospec- 66 biopsies (39%) showed no cancer or benign pathology (of tively, perceived to be misclassifications of benign prostatic which 52 (79%) with normal prostate tissue, 7 (11%) with hyperplasia (BPH) nodules (Fig. 2). Of the 16 FP lesions prostatitis, 5 (8%) with prostatic intraepithelial neoplasia in the PZ, 10 could have retrospectively been avoided (e.g., (PIN), and 2 (3.0%) with PIN and prostatitis). anatomic variants of the CZ (Fig. 3), low-grade prostatitis According to multivariable logistic regression analysis, with very little-to-no diffusion restriction, BPH nodule in inexperience of the radiologist was significantly (P = 0.044, PZ). In six cases, a FP reading was perceived as not avoid- odds ratio 1.927, 95% confidence interval [CI] 1.017–3.651) able (3 cases of prostatitis with substantial diffusion restric- and independently associated with a FP reading, while tion mimicking PCa, and 3 cases with, also in retrospect, zonal location was not (P = 0.202, odds ratio 1.444, 95% MR imaging abnormalities but with normal histopathology CI 0.820–2.539) (Table 2). Overall, FP rate was 50% for results). These latter three cases (9%, 3/33) were interpreted 1 3 1048 Abdominal Radiology (2019) 44:1044–1051 Fig. 2 74-year-old patient, with a PSA of 28 ng/mL and a history of was retrospectively perceived as an example of a misclassification of previous negative transrectal-ultrasound-guided biopsy. PSA density a benign prostatic hyperplasia nodule in the transition zone, which of 0.13. Pitfall of benign prostatic hyperplasia nodule in the transition should have been classified as a PIRADS 2 lesion. b Apparent diffu- zone. Histopathology result revealed benign changes, no malignancy. sion coefficient map with the false positive lesion (straight arrow). c a T2-weighted image with a PIRADS 4 lesion (straight arrow) on Dynamic contrast-enhanced image, slightly asymmetric enhancement the initial read, reported by an inexperienced radiologist. This lesion (straight arrow) Fig. 3 72-year-old patient, with a PSA of 12 ng/mL and a history of arrow) on the initial read, reported by an inexperienced radiologist. previous negative transrectal-ultrasound-guided biopsy. PSA density This lesion was retrospectively perceived as an example of a misclas- of 0.19. Classic pitfall of the central zone compressed between the sification of an anatomic variant of the central zone. b Apparent dif- transition zone and peripheral zone. Histopathology result revealed fusion coefficient map with the false positive lesion (straight arrow). normal tissue, no malignancy, prostatic intraepithelial neoplasia or c Dynamic contrast-enhanced image, with no contrast enhancement prostatitis. a T2-weighted image with a PIRADS 4 lesion (straight (straight arrow) as likely due to technical inaccuracies (e.g., needle delivery, as MR imaging abnormalities but with normal histopa- inaccurate fusion of MR, and TRUS images). thology results. Of the 17 FP lesions in the PZ, five were For the experienced radiologist, of the 13 FP lesions high-grade PIN and/or prostatitis, and two were perceived in the TZ, 6 were perceived as misclassifications (of BPH as misclassifications (very little diffusion restriction). nodules and prominent anterior fibromuscular stroma). Ten were perceived, also in retrospect, as MR imaging Two were prospectively called as granulomateus prostati- abnormalities, but with normal histopathology results. tis (however, biopsy was still advised), five were perceived 1 3 Abdominal Radiology (2019) 44:1044–1051 1049 Altogether for both the TZ and PZ, 15 (50%, 15/30) FP to be missed at biopsy. The findings of this study have two lesions were likely due to technical inaccuracies. potential clinical implications. First, the relative underper- formance of inexperienced readers underlines that reading MP-MRI of the prostate should be reserved to experienced Discussion radiologists. The importance of subspecialty training is a well-acknowledged issue in breast and cardiac imaging. This To our knowledge, this study is the first to investigate the should not be different for MRI of the prostate. After internal FP findings of PIRADS (V2) 3-5 lesions in a pure clini- evaluation of the results of this study, we decided to reduce cal setting with emphasis on reader experience and zonal the number of radiologists reporting MP-MRI of the prostate location, in a patient population undergoing TRUS MRI and educate internally. Since then, only two of six special- fusion-guided targeted prostate biopsy. Our results show that ized abdominal radiologists have been reporting MP-MRI of inexperience is independently and significantly associated the prostate at our institution. The second potential implica- with a FP reading while zonal location is not. Furthermore, tion of our findings is that training sessions for inexperi- the evaluation of the TZ is mainly what makes the differ - enced radiologists should, along with a sufficient caseload ence between the inexperienced and experienced radiologist. with histopathological correlation, pay special attention to In general, the TZ is an anatomic area that is considered the TZ [24, 25], CZ, and low-grade prostatitis. difficult when interpreting MP-MRI of the prostate. The The overall detection (or true positive) rate of the expe- challenging aspect of detecting PCa in the TZ is probably rienced reader in this study (68% overall detection rate because of the heterogeneous appearance of the TZ which in PIRADS 3-5 lesions) was slightly higher than other is mainly due to the BPH it contains. Our study shows that expert centers performing MRI targeted prostate biopsy the evaluation of the TZ is even more difficult for the inex- (59%–62%) [26, 27]; however, both studies were performed perienced radiologist. We also found that, to a lesser degree, using PIRADS V1 which could also have accounted for other potential sources of a FP reading for the inexperienced the slight differences. In 2013, Bratan et al. [13] were the reader are anatomic variants of the CZ and low-grade prosta- first to investigate the FPs. However, this publication was titis. One of the main advantages of MP-MRI is its potential from the pre-PIRADS era, and they did not assess different to reduce unnecessary biopsies [17, 18]. Nonetheless, in case levels of experience. In another study published in 2017, of an inexperienced radiologist and a suspected TZ lesion, the issue of experience was addressed [5]. They reported this could lead to more unnecessary biopsies. The PIRADS that reader experience may help to reduce overcalling and 3 classification had a high number of FPs, for both the inex- avoid over targeting of lesions, which is in concordance with perienced and the experienced radiologists. Even though this our results. Nevertheless, none of these studies specifically is likely to be expected as PIRADS is a Likert scale, efforts looked at the combination of reader experience and zonal for reducing the number of unnecessary biopsies in this spe- location in FP lesions. cific group should be made. A possible solution could be Our study had several limitations. First, because of its the addition of PSA density [19]. Consequently, all lesions retrospective design, MRI protocols were heterogeneous with a PIRADS score of 3 and a cut off value of for example (i.e., different magnetic field strengths and slightly vary - below 0.15 could be refrained from biopsy. However, before ing MRI sequence settings), and possibly of suboptimal implementation this needs to be investigated further with quality (e.g., majority of scans were obtained with a 1.5T special attention to the number of missed significant cancers scanner; and no endorectal coil or the inability to control with this approach. Also noteworthy, FP rates do not only imaging parameters due to referrals from outside hospi- depend on factors accountable to the reader, but also on inac- tals). However, this is completely in line with routine clini- curacies related to the biopsy technique itself, most likely in cal practice, and this issue is also frequently encountered case of smaller lesions [20–22]. Nonetheless, when analyz- in multicenter studies [5, 18], which in fact increases the ing the FPs in the patient cohort of this study that were read generalizability of our results. Second, we were not able to by the inexperienced radiologists the majority appeared to be analyze false negatives and their relation to reader experi- classical examples of misclassification for the TZ as well as ence, which could have been done if the reference stand- for the PZ. In our study, we found that only in 9% of the FP ard would have been prostatectomy specimens or if long- cases for the inexperienced radiologists and in 50% of the FP term follow-up data would have been available. Third, this cases for the experienced radiologist technical issues could study did not investigate other factors potentially asso- have been the reason for a FP lesion. A study by Sheriden ciated with a FP reading, such as clinical variables and et al. [23] reported that 28% of the FP PIRADS 5 lesions apex-base location [23]. Fourth, this study consisted of could have been missed at biopsy. The difference with our patients who had received prior biopsy (either negative or study is that they only investigated PIRADS 5 lesions that with cancer but on active surveillance) possibly reducing are usually large and therefore technique-wise less likely the future generalizability of the results. It is expected 1 3 1050 Abdominal Radiology (2019) 44:1044–1051 OpenAccess This article is distributed under the terms of the Creative that as we move forward, more biopsy-naïve patients will Commons Attribution 4.0 International License (http://creativecom- undergo MP-MRI and targeted biopsy. Fifth, we did not mons.org/licenses/by/4.0/), which permits unrestricted use, distribu- include Gleason 3 + 3 lesions in our FP definition. In the tion, and reproduction in any medium, provided you give appropriate era of the increasing number of active surveillance candi- credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. dates, we are well aware of the importance of being able to discriminate between clinically significant (currently regarded as Gleason ≥ 3 + 4) and insignificant PCa (cur - References rently regarded as Gleason ≤ 3 + 3). However, including 3 + 3 in the FP definition on MP-MRI would, unfairly, 1. M.A. Haider, X. Yao, A. Loblaw, A. Finelli, Multiparametric Mag- imply that there are proven MRI features that are prospec- netic Resonance Imaging in the Diagnosis of Prostate Cancer: A tively able to discriminate Gleason 3 + 3 from 3 + 4. There Systematic Review, Clin. Oncol. (2016). https://doi.or g/10.1016/j. are retrospective and validation studies that have investi- clon.2016.05.003 2. B. Turkbey, A.M. Brown, S. Sankineni, B.J. Wood, P.A. Pinto, gated the relationship between ADC values and Gleason P.L. Choyke, Multiparametric Prostate Magnetic Resonance grades, and they have shown promising results [28]. Yet Imaging in the Evaluation of Prostate Cancer, CA Cancer J Clin. substantial overlaps between the different grades and ADC (2016). https ://doi.org/10.3322/caac.21333 values exist [28], making it not very useful in a clinical 3. D. Yakar, O.A. Debats, J.G.R. Bomers, M.G. Schouten, P.C. Vos, E. Van Lin, J.J. Fütterer, J.O. Barentsz, Predictive value of MRI in setting. Due to the lack of robust discriminatory imaging the localization, staging, volume estimation, assessment of aggres- features, publications addressing pitfalls on prostate MRI siveness, and guidance of radiotherapy and biopsies in prostate usually only contain examples of benign disease or FPs cancer, J. Magn. Reson. Imaging. (2012). https://doi.or g/10.1002/ related to more technical issues [10, 11]. Likewise, the jmri.22790 4. G.A. Sonn, R.E. Fan, P. Ghanouni, N.N. Wang, J.D. Brooks, A.M. clinically important decision between Gleason 3 + 3 and Loening, B.L. Daniel, K.J. To’o, A.E. Thong, J.T. Leppert, Pros- 3 + 4 can also be challenging on histopathology slides in tate Magnetic Resonance Imaging Interpretation Varies Substan- biopsy specimens. There is substantial interobserver vari- tially Across Radiologists, Eur. Urol. Focus. (2018). https ://doi. ability together with shortcomings of the Gleason grading org/10.1016/j.euf.2017.11.010 5. N.L. Hansen, B.C. Koo, F.A. Gallagher, A.Y. Warren, A. Doble, scoring system itself [29]. This is probably also the reason V. Gnanapragasam, O. Bratt, C. Kastner, T. Barrett, Compari- why some supposedly Gleason 3 + 3 tumors, even though son of initial and tertiary centre second opinion reads of mul- not very common, can metastasize or show aggressive tiparametric magnetic resonance imaging of the prostate prior to behavior (such as extraprostatic extension or even seminal repeat biopsy, Eur. Radiol. (2017). https://doi.or g/10.1007/s0033 0-016-4635-5 vesical invasion) [30, 31]. Furthermore, targeted biopsies 6. S. Müller, G. Lilleaasen, T.E. Sand, L. Løfsgaard, M. Estop- obtained from the cancer-suspicious lesions detected on Garanto, D. Helgø, P. Sund, V. Mygland, Poor reproducibility of MP-MRI might not accurately represent the true aggres- PIRADS score in two multiparametric MRIs before biopsy in men siveness found in a prostatectomy specimen [32–34], with with elevated PSA, World J. Urol. (2018). https://doi.or g/10.1007/ s0034 5-018-2252-4 some studies even mentioning percentages of upgrading 7. K. Garcia-Reyes, N.M. Passoni, M.L. Palmeri, C.R. Kauffman, from 3 + 3 to 3 + 4 in 67.4% of the cases [32]. So even K.R. Choudhury, T.J. Polascik, R.T. Gupta, Detection of pros- though there is an absolute clinical need for discriminat- tate cancer with multiparametric MRI (mpMRI): effect of dedi- ing significant from insignificant PCa, it is radiologically cated reader education on accuracy and confidence of index and anterior cancer diagnosis, Abdom. Imaging. (2015). https ://doi. not meaningful when assessing FPs. Future studies should org/10.1007/s0026 1-014-0197-7 focus on this unsolved issue. 8. B.G. Muller, J.H. Shih, S. Sankineni, J. Marko, S. Rais-Bahrami, In conclusion, inexperience of the radiologist is sig- A.K. George, J.J.M.C.H. de la Rosette, M.J. Merino, B.J. Wood, nificantly and independently associated with a FP read- P. Pinto, P.L. Choyke, B. Turkbey, Prostate Cancer: Interobserver Agreement and Accuracy with the Revised Prostate Imaging ing, while zonal location is not. Inexperienced radiologists Reporting and Data System at Multiparametric MR Imaging, have a significantly higher FP rate in the TZ. Radiology. (2015). https ://doi.org/10.1148/radio l.20151 42818 9. J. Yu, A.S. Fulcher, S.G. Winks, M.A. Turner, R.D. Clayton, M. Brooks, S. Li, Diagnosis of typical and atypical transition zone prostate cancer and its mimics at multiparametric prostate MRI, Compliance with ethical standards Br. J. Radiol. (2017). https ://doi.org/10.1259/bjr.20160 693 10. A.B. Rosenkrantz, S.S. Taneja, Radiologist, be aware: Ten pitfalls Conflict of interest All authors declare that they have no disclosures that confound the interpretation of multiparametric prostate MRI, or conflicts of interest. Am. J. Roentgenol. (2014). https ://doi.org/10.2214/ajr.13.10699 11. X. Kitzing, A. Prando, C. Varol, G.S. Karczmar, F. Maclean, A. Ethical approval This retrospective study was approved by the local Oto, Benign Conditions That Mimic Prostate Carcinoma: MR ethics committee and the need for informed consent was waived (Reg- Imaging Features with Histopathologic Correlation., Radiograph- istration Number 201700780). ics. 36 (2016) 162–175. https ://doi.org/10.1148/rg.20161 50030 12. V. Panebianco, F. Giganti, Y.X. Kitzing, F. Cornud, R. Campa, G. De Rubeis, A. Ciardi, C. Catalano, G. Villeirs, An update of 1 3 Abdominal Radiology (2019) 44:1044–1051 1051 pitfalls in prostate mpMRI: a practical approach through the lens multiparametric prostate MRI, Am. J. Roentgenol. (2018). https of PI-RADS v. 2 guidelines, Insights Imaging. (2018). https://doi. ://doi.org/10.2214/ajr.17.18680 org/10.1007/s1324 4-017-0578-x 24. A.P.S. Kirkham, P. Haslam, J.Y. Keanie, I. McCafferty, A.R. 13. F. Bratan, E. Niaf, C. Melodelima, A.L. Chesnais, R. Souchon, Padhani, S. Punwani, J. Richenberg, G. Rottenberg, A. Sohaib, F. Mège-Lechevallier, M. Colombel, O. Rouvière, Influence of P. Thompson, L.W. Turnbull, L. Kurban, A. Sahdev, R. Clem- imaging and histological factors on prostate cancer detection and ents, B.M. Carey, C. Allen, Prostate MRI: Who, when, and how? localisation on multiparametric MRI: A prospective study, Eur. Report from a UK consensus meeting, Clin. Radiol. (2013). https Radiol. (2013). https ://doi.org/10.1007/s0033 0-013-2795-0 ://doi.org/10.1016/j.crad.2013.03.030 14. J.C. Weinreb, J.O. Barentsz, P.L. Choyke, F. Cornud, M.A. 25. O. Akin, C.C. Riedl, N.M. Ishill, C.S. Moskowitz, J. Zhang, H. Haider, K.J. Macura, D. Margolis, M.D. Schnall, F. Shtern, C.M. Hricak, Interactive dedicated training curriculum improves accu- Tempany, H.C. Thoeny, S. Verma, PI-RADS Prostate Imaging - racy in the interpretation of MR imaging of prostate cancer, Eur. Reporting and Data System: 2015, Version 2, Eur. Urol. (2016). Radiol. (2010). https ://doi.org/10.1007/s0033 0-009-1625-x https ://doi.org/10.1016/j.eurur o.2015.08.052 26. T. Hambrock, D.M. Somford, C. Hoeks, S.A.W. Bouwense, H. 15. PRIAS guidelines. https ://www.prias -proje ct.org/. Accessed 21 Huisman, D. Yakar, I.M. van Oort, J.A. Witjes, J.J. Fütterer, J.O. Jan 2019 Barentsz, Magnetic Resonance Imaging Guided Prostate Biopsy in 16. J.I. Epstein, L. Egevad, M.B. Amin, B. Delahunt, J.R. Srigley, P.A. Men With Repeat Negative Biopsies and Increased Prostate Specic fi Humphrey, The 2014 international society of urological pathology Antigen, J. Urol. (2010). https://doi.or g/10.1016/j.juro.2009.10.022 (ISUP) consensus conference on gleason grading of prostatic car- 27. G. Gaziev, K. Wadhwa, T. Barrett, B.C. Koo, F.A. Gallagher, E. cinoma definition of grading patterns and proposal for a new grad- Serrao, J. Frey, J. Seidenader, L. Carmona, A. Warren, V. Gnana- ing system, Am. J. Surg. Pathol. (2016). https ://doi.org/10.1097/ pragasam, A. Doble, C. Kastner, Defining the learning curve pas.00000 00000 00053 0 for multiparametric magnetic resonance imaging (MRI) of the 17. H.U. Ahmed, A. El-Shater Bosaily, L.C. Brown, R. Gabe, R. prostate using MRI-transrectal ultrasonography (TRUS) fusion- Kaplan, M.K. Parmar, Y. Collaco-Moraes, K. Ward, R.G. Hindley, guided transperineal prostate biopsies as a validation tool, BJU A. Freeman, A.P. Kirkham, R. Oldroyd, C. Parker, M. Ember- Int. (2016). https ://doi.org/10.1111/bju.12892 ton, Diagnostic accuracy of multi-parametric MRI and TRUS 28. T. Hambrock, D.M. Somford, H.J. Huisman, I.M. van Oort, J.A. biopsy in prostate cancer (PROMIS): a paired validating con- Witjes, C.A. Hulsbergen-van de Kaa, T. Scheenen, J.O. Barentsz, firmatory study, Lancet. (2017). https ://doi.org/10.1016/s0140 Relationship between Apparent Diffusion Coefficients at 3.0-T -6736(16)32401 -1 MR Imaging and Gleason Grade in Peripheral Zone Prostate Can- 18. V. Kasivisvanathan, A.S. Rannikko, M. Borghi, V. Panebi- cer, Radiology. (2011). https ://doi.org/10.1148/radio l.11091 409 anco, L.A. Mynderse, M.H. Vaarala, A. Briganti, L. Budäus, G. 29. L. Egevad, A.S. Ahmad, F. Algaba, D.M. Berney, L. Boccon- Hellawell, R.G. Hindley, M.J. Roobol, S. Eggener, M. Ghei, A. Gibod, E. Compérat, A.J. Evans, D. Griffiths, R. Grobholz, G. Villers, F. Bladou, G.M. Villeirs, J. Virdi, S. Boxler, G. Robert, Kristiansen, C. Langner, A. Lopez-Beltran, R. Montironi, S. P.B. Singh, W. Venderink, B.A. Hadaschik, A. Ruffion, J.C. Hu, Moss, P. Oliveira, B. Vainer, M. Varma, P. Camparo, Standardi- D. Margolis, S. Crouzet, L. Klotz, S.S. Taneja, P. Pinto, I. Gill, zation of Gleason grading among 337 European pathologists, C. Allen, F. Giganti, A. Freeman, S. Morris, S. Punwani, N.R. Histopathology. (2013). https ://doi.org/10.1111/his.12008 Williams, C. Brew-Graves, J. Deeks, Y. Takwoingi, M. Ember- 30. H. Wenger, A.B. Weiner, A. Razmaria, G.P. Paner, S.E. Eggener, ton, C.M. Moore, MRI-Targeted or Standard Biopsy for Prostate- Risk of lymph node metastases in pathological gleason score≤6 Cancer Diagnosis (PRECISION study), N. Engl. J. Med. (2018). prostate adenocarcinoma: Analysis of institutional and population- https ://doi.org/10.1056/nejmo a1801 993 based databases, Urol. Oncol. Semin. Orig. Investig. (2017). https 19. F.A. Distler, J.P. Radtke, D. Bonekamp, et al., The value of PSA ://doi.org/10.1016/j.urolo nc.2016.08.004 density in combination with PI-RADS™ for the accuracy of pros- 31. O. Hassan, M. Han, A. Zhou, A. Paulk, Y. Sun, A. Al-Harbi, tate cancer prediction, J. Urol. (2017). https ://doi.org/10.1016/j. A. Alrajjal, F. Baptista dos Santos, J.I. Epstein, Incidence of juro.2017.03.130 Extraprostatic Extension at Radical Prostatectomy with Pure 20. A. Borkowetz, I. Platzek, M. Toma, T. Renner, R. Herout, M. Gleason Score 3 + 3 = 6 (Grade Group 1) Cancer: Implications Baunacke, M. Laniado, G. Baretton, M. Froehner, S. Zastrow, M. for Whether Gleason Score 6 Prostate Cancer Should be Renamed Wirth, Direct comparison of multiparametric magnetic resonance “Not Cancer” and for Selection Criteria for Active Surveillance, imaging (MRI) results with final histopathology in patients with J. Urol. (2018). https ://doi.org/10.1016/j.juro.2017.11.067 proven prostate cancer in MRI/ultrasonography-fusion biopsy, 32. T. Fukagai, T. Namiki, H. Namiki, R.G. Carlile, M. Shimada, H. BJU Int. (2016). https ://doi.org/10.1111/bju.13461 Yoshida, Discrepancies between Gleason scores of needle biopsy 21. C.P. Filson, S. Natarajan, D.J.A. Margolis, J. Huang, P. Lieu, F.J. and radical prostatectomy specimens, Pathol. Int. (2001). https :// Dorey, R.E. Reiter, L.S. Marks, Prostate cancer detection with doi.org/10.1046/j.1440-1827.2001.01207 .x magnetic resonance-ultrasound fusion biopsy: The role of system- 33. N.M. Corcoran, C.M. Hovens, M.K.H. Hong, J. Pedersen, R.G. atic and targeted biopsies, Cancer. (2016). https://doi.or g/10.1002/ Casey, S. Connolly, J. Peters, L. Harewood, M.E. Gleave, S.L. Gold- cncr.29874 enberg, A.J. Costello, Underestimation of Gleason score at prostate 22. J.P. Radtke, T.H. Kuru, S. Boxler, C.D. Alt, I. V. Popeneciu, biopsy ree fl cts sampling error in lower volume tumours, BJU Int. C. Huettenbrink, T. Klein, S. Steinemann, C. Bergstraesser, (2012). https ://doi.org/10.1111/j.1464-410x.2011.10543 .x M. Roethke, W. Roth, H.P. Schlemmer, M. Hohenfellner, B.A. 34. S. Mesko, L. Marks, O. Ragab, S. Patel, D.A. Margolis, D.J. Hadaschik, Comparative Analysis of Transperineal Template Demanes, M. Kamrava, Targeted Prostate Biopsy Gleason Score Saturation Prostate Biopsy Versus Magnetic Resonance Imaging Heterogeneity and Implications for Risk Stratification, Am. J. Targeted Biopsy with Magnetic Resonance Imaging-Ultrasound Clin. Oncol. Cancer Clin. Trials. (2018). https://doi.or g/10.1097/ Fusion Guidance, J. Urol. (2015). https ://doi.or g/10.1016/j.coc.00000 00000 00030 8 juro.2014.07.098 23. A.D. Sheridan, S.K. Nath, S. Aneja, J.S. Syed, J. Pahade, M. Publisher’s Note Springer Nature remains neutral with regard to Mathur, P. Sprenkle, J.C. Weinreb, M. Spektor, MRI-ultrasound jurisdictional claims in published maps and institutional affiliations. fusion targeted biopsy of prostate imaging reporting and data system version 2 category 5 lesions found false-positive at 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Abdominal Radiology Springer Journals

False positives in PIRADS (V2) 3, 4, and 5 lesions: relationship with reader experience and zonal location

Download PDF
8 pages

Loading next page...
 
/lp/springer-journals/false-positives-in-pirads-v2-3-4-and-5-lesions-relationship-with-M3bR7UpUuo

References (44)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Medicine & Public Health; Imaging / Radiology; Gastroenterology; Hepatology
ISSN
2366-004X
eISSN
2366-0058
DOI
10.1007/s00261-019-01919-2
Publisher site
See Article on Publisher Site

Abstract

Purpose To investigate the effect of reader experience and zonal location on the occurrence of false positives (FPs) in PIRADS (V2) 3, 4, and 5 lesions on multiparametric (MP)-MRI of the prostate. Materials and methods This retrospective study included 139 patients who had consecutively undergone an MP-MRI of the prostate in combination with a transrectal ultrasound MRI fusion-guided biopsy between 2014 and 2017. MRI exams were prospectively read by a group of inexperienced radiologists (cohort 1; 54 patients) and an experienced radiologist (cohort 2; 85 patients). Multivariable logistic regression analysis was performed to determine the association of experience of the radiologist and zonal location with a FP reading. FP rates were compared between readings by inexperienced and experienced radiologists according to zonal location, using Chi-square (χ ) tests. Results A total of 168 lesions in 139 patients were detected. Median patient age was 68 years (Interquartile range (IQR) 62.5–73), and median PSA was 10.9 ng/mL (IQR 7.6–15.9) for the entire patient cohort. According to multivariable logis- tic regression, inexperience of the radiologist was significantly (P = 0.044, odds ratio 1.927, 95% confidence interval [CI] 1.017–3.651) and independently associated with a FP reading, while zonal location was not (P = 0.202, odds ratio 1.444, 95% CI 0.820–2.539). In the transition zone (TZ), the FP rate of the inexperienced radiologists 59% (17/29) was significantly higher (χ P = 0.033) than that of the experienced radiologist 33% (13/40). Conclusion Inexperience of the radiologist is significantly and independently associated with a FP reading, while zonal location is not. Inexperienced radiologists have a significantly higher FP rate in the TZ. Keywords MP-MRI · Prostate cancer · PIRADS V2 · Reader experience · Zonal location · False positives Introduction Multiparametric (MP) MRI of the prostate is the best option for local diagnosis of prostate cancer (PCa) [1–3]. It can be useful in various clinical settings, such as for detection or staging purposes, guiding biopsies, detection of local * Derya Yakar recurrence, and as a tool to select candidates for active [email protected] surveillance. A negative transrectal-ultrasound (TRUS)-guided pros- Department of Radiology, University Medical Center Groningen, Hanzeplein 1, 9700 CC Groningen, tate biopsy with persistent clinical suspicion of PCa is still The Netherlands the most frequent reason to perform MP-MRI in clinical Department of Urology, University Medical Center practice. Together with the introduction of the commer- Groningen, Hanzeplein 1, 9700 CC Groningen, cially available TRUS-MRI fusion-guided prostate biopsy The Netherlands systems, the use of MP-MRI of the prostate has increased Department of Urology, Erasmus MC, Dr. Molewaterplein tremendously over the last years. However, along with 40, 3015 GD Rotterdam, The Netherlands the increasing popularity of this diagnostic test, serious Department of Urology, University Medical Center Utrecht, concerns on quality issues have been raised [4, 5]. These Heidelberglaan 100, 3584 CX Utrecht, The Netherlands Vol:.(1234567890) 1 3 Abdominal Radiology (2019) 44:1044–1051 1045 quality concerns include image acquisition, interpretation, Materials and methods and reporting. The Prostate Imaging Reporting and Data System (PIRADS) has been introduced to address some of Patients these aforementioned issues. Nonetheless, even with the use of the PIRADS version (V) 2 system, interreader vari- This retrospective study was approved by the local ethics ability is still a common present-day problem [6]. Reader committee, and the need for informed consent was waived experience is probably an important issue in this setting. (registration number 201700780). All patients (n = 186) Previous studies have already emphasized the importance who had consecutively undergone MP-MRI of the prostate of subspecialty reading in prostate MRI [7, 8]. Strongly in combination with a TRUS-MRI fusion-guided prostate related to this subject are the pitfalls encountered on MP- biopsy between 2014 and 2017 were potentially eligible MRI of the prostate. Awareness of the causes of false for inclusion. MP-MRI scans of the prostate were either positives (FPs) can theoretically improve the diagnostic obtained from in-house records or from referring centers. performance of the radiologist and decrease interreader Patients whose MRI scans were reported before the intro- variability. Besides the experience of the radiologist, zonal duction of standard reporting according to PIRADS version location can also be a source of FPs [9]. For example, 2 [14] (n = 38) were excluded. An additional number of 9 detection of PCa in the transition zone (TZ) is often per- patients were excluded because of missing crucial informa- ceived as a difficult task. Pictorial reviews and case series tion, such as a PIRADS score (n = 7), inconclusive histopa- have touched on the topic of FPs [10–12], but systematic thology report (n = 1), and a history of radiotherapy (n = 1). studies specifically investigating FPs in a pure clinical set- Finally, 139 unique patients were included in this study, of ting are scarce and from the pre-PIRADS era [13]. We whom 54 patients were prospectively read by the inexpe- hypothesized that, when not educated properly, the novice rienced radiologists (cohort 1), and 85 patients were read reader will, in daily clinical practice, cause unnecessary by the experienced radiologist (cohort 2) (Fig. 1). 99/139 biopsies especially in the TZ. patients had undergone prostate MP-MRI because of a pre- Therefore, the purpose of this study was to investigate vious history of negative TRUS-guided prostate biopsy and the effects of reader experience and zonal location on the persistent clinical suspicion (based on PSA levels and/or occurrence of FPs in PIRADS (V2) 3, 4, and 5 lesions on abnormal digital rectal examination). 41/139 had undergone MP-MRI of the prostate, with targeted TRUS-MRI fusion- prostate MP-MRI for staging purposes because of increasing guided prostate biopsy as reference standard. PSA levels, while being on an active surveillance program (inclusion of only low-grade cancers according to PRIAS Fig. 1 Flowchart showing the numbers of potentially eligible patients, excluded patients, included patients, and patients scored by inexperienced and experienced radiologists 1 3 1046 Abdominal Radiology (2019) 44:1044–1051 guidelines (Prostate cancer Research International: Active Targeted biopsies were performed in all patients, while surveillance) [15]). two patients also received an additional systematic biopsy. Biopsy procedures were undertaken by four different urolo- MRI acquisition and analysis gists (with 4, 4, 1, and 1 year(s) of experience with fusion biopsy, respectively). The number of cores per lesion was at 96 patients were referred from outside hospitals (total of the discretion of the urologist performing the biopsy. 9 different institutions), while 43 patients were scanned in our institution. In total, 88 exams were performed on a 1.5T Histopathology scanner (of which none was performed with an endorectal coil) and 51 exams were performed on a 3T MRI scanner The targeted biopsies were analyzed by a specialist uro- (of which 10 with an endorectal coil). Of the 51 exams per- pathologist according to the International Society of Urolog- formed on a 3T scanner, five were primarily scanned on a ical Pathology (ISUP) 2014 recommendations [16]. Biopsy 1.5T scanner in outside hospitals. Due to technical limita- results containing cancer (Gleason ≥ 3 + 3) were categorized tions (e.g., motion artifacts, inadequate quality), these were as true positives, while the ones containing no cancer were re-obtained in our institution. The MP-MRI examinations of categorized as FPs. the prostate of 131 patients comprised an axial T1-weighted image, a high-resolution multiplanar T2-weighted image Statistical analysis (slice thickness of 3 mm), a diffusion-weighted image (DWI) with at least three b-values (varying between 0-2000 s/mm 2 The Shapiro–Wilk’s test was used to test if the continuous and with a minimum highest value of 800 s/mm ) and a variables age, PSA, PSA density, prostate volume, number calculated apparent diffusion coefficient (ADC) map, and of cancer-suspicious lesions per patient, and number of a dynamic contrast-enhanced sequence. The remaining cores taken per patient, were normally distributed. These 8 patients had undergone the same sequences, except for variables were then compared between the patient cohort a dynamic contrast-enhanced (DCE) sequence. All MRI read by the inexperienced radiologists (cohort 1) and the examinations were read at our tertiary referral center. Radi- patient cohort read by the experienced radiologist (cohort ologists were not blinded to clinical data or external reports, 2), using the unpaired t test for normally distributed data, completely in line with clinical practice. MRI data were read and the Mann–Whitney U test for not normally distributed prospectively by four different radiologists, with different data. The FP rate was calculated as the number of biopsied levels of experience in MP-MRI prostate reporting. One lesions containing no PCa divided by the total number of trained genitourinary radiologist (with 5 years of experi- biopsied lesions. FP rate readings were compared between ence and > 500 case readings with histopathologic correla- inexperienced and experienced radiologists according to tion) was defined as the experienced reader, while all other zonal location, using Chi-square (χ ) tests. Multivariable three radiologists (with 1–2 years of experience and < 100 logistic regression analysis was performed to determine case readings without histopathologic correlation) were the association of the experience of the radiologist (inex- categorized as inexperienced readers. The MRI scans were perienced vs. experienced) and zonal location (TZ vs. PZ) reported by using PIRADS V2, with PIRADS 1–5 represent- with a FP reading. P-values less than 0.05 (two-sided) were ing an incremental scoring system of very low likelihood to regarded statistically significant. All statistical analyses were very high likelihood of clinically significant PCa [14]. All performed using MedCalc Statistical Software version 18.5 prostate MRI reports were analyzed for reader experience, (MedCalc Software bvba, Ostend, Belgium). PIRADS score, and zonal location (peripheral zone (PZ), TZ, or central zone (CZ)). Furthermore, the FP lesions were retrospectively investigated by the experienced radiologist for other sources of error such as known mimickers of PCa Results (e.g., prostatitis, anatomic pitfalls) or possible technical ina- curacies (e.g., needle delivery, fusion misregistration). The 139 included patients had a total of 168 PIRADS 3-5 lesions. Median patient age was 68 years (interquartile range TRUS‑MRI fusion‑guided prostate biopsy (IQR) 62.5–73) and median PSA was 10.9  ng/mL (IQR 7.6–15.9) for the entire patient cohort. Age, PSA, PSA den- All patients with at least one lesion with a PIRADS score sity, prostate volume, number of cancer-suspicious lesions of 3 or higher were biopsied according to our institutional per patient, and number of cores taken per patient were not standard, with up to a maximum of three biopsied lesions significantly different (P > 0.101) between the patient cohort per patient. The Dynacad Uronav fusion biopsy system read by the inexperienced radiologists and the patient cohort (Invivo, Gainesville, Florida, USA) was used for all biopsies. read by the experienced radiologist (Table 1). 1 3 Abdominal Radiology (2019) 44:1044–1051 1047 Table 1 Characteristics of the c Patient cohort 1 read by inexpe- Patient cohort 2 read by experi- P value included patients for the two rienced radiologists enced radiologist groups of radiologists Total number of patients 54 Total number of patients 85 Total number of lesions 66 Total number of lesions 102 a b Age (years) 70 (65–73) 67 ± 7.0 0.101 a a PSA (ng/mL) 11.6 (7.7–17.0) 10 (7.3–14.2) 0.155 b b PSA density 0.20 ± 0.10 0.19 (0.14–0.28) 0.448 a a Prostate volume (cc) 69 (51–78) 50 (40–82) 0.303 a a Number of cancer-suspi- 1 (1–1) 1 (1–1) 0.191 cious lesions per patient a a Number of cores taken per 3 (3–4) 3 (3–3) 0.824 suspicious lesion Data presented as median (interquartile range) (not normally distributed according to Shapiro–Wilk test) Data presented as mean ± standard deviation (normally distributed according to Shapiro–Wilk test) According to Mann–Whitney U test (for not normally distributed independent samples) Table 2 Multivariable logistic regression: relationship of experience Table 3 FP rate according to reader experience for zonal location and and zonal location with FP reading PIRADS category Variable Odds ratio (95% CI) P value Variable FP rate inexpe- FP rate experi- P value rienced readers enced reader Inexperience 1.927 (1.017-3.651) 0.044 Zone Zonal location (transition and 1.444 (0.820-2.539) 0.202  Transition 59% (17/29) 33% (13/40) 0.033 peripheral zone)  Peripheral 43% (16/37) 29% (17/58) 0.164 CI confidence interval PIRADS  3 100% (11/11) 71% (17/24) 0.049  4 44% (16/36) 27% (13/49) 0.105 The total numbers of PIRADS 3–5 lesions called by the  5 32% (6/19) 10% (3/29) 0.059 inexperienced radiologists and the experienced radiologist Overall (PIRADS 3–5) 50% (33/66) 32% (33/102) 0.020 were 66 (in 54 patients, cohort 1) and 102 (in 85 patients, a 2 According to χ test cohort 2), respectively. 110 patients had one lesion, 23 patients had two lesions, and 4 patients had three lesions. The inexperienced radiolo- gists called 29 lesions in the TZ (7 PIRADS 3, 16 PIRADS the inexperienced radiologists, and 32% for the experienced 4, 6 PIRADS 5) and 37 lesions in the PZ (4 PIRADS 3, radiologist (P = 0.020). The FP rate in the TZ for the inex- 20 PIRADS 4, 13 PIRADS 5). The experienced radiologist perienced radiologists (59%, 17/29) was significantly higher called 40 lesions in the TZ (13 PIRADS 3, 11 PIRADS 4, (P = 0.033) than that for the experienced radiologist (33%). 16 PIRADS 5) and 58 lesions in the PZ (7 PIRADS 3, 38 On the other hand, the FP rate in the PZ for the inexperi- PIRADS 4, 13 PIRADS 5). Only the experienced radiologist enced radiologists (43%, 16/37) was not significantly dif - called four lesions in the central zone (CZ). Therefore, the ferent (P = 0.164) from that for the experienced radiologist CZ was excluded from further analysis. In total, 102 biopsies (29%). Corresponding results are also displayed in Table 3. (61%) showed PCa (of which 60 (59%) contained clinically After a further breakdown of the 17 FP lesions in the significant cancer, defined as ≥ 3 + 4 Gleason score), while TZ for the inexperienced radiologists, all were, retrospec- 66 biopsies (39%) showed no cancer or benign pathology (of tively, perceived to be misclassifications of benign prostatic which 52 (79%) with normal prostate tissue, 7 (11%) with hyperplasia (BPH) nodules (Fig. 2). Of the 16 FP lesions prostatitis, 5 (8%) with prostatic intraepithelial neoplasia in the PZ, 10 could have retrospectively been avoided (e.g., (PIN), and 2 (3.0%) with PIN and prostatitis). anatomic variants of the CZ (Fig. 3), low-grade prostatitis According to multivariable logistic regression analysis, with very little-to-no diffusion restriction, BPH nodule in inexperience of the radiologist was significantly (P = 0.044, PZ). In six cases, a FP reading was perceived as not avoid- odds ratio 1.927, 95% confidence interval [CI] 1.017–3.651) able (3 cases of prostatitis with substantial diffusion restric- and independently associated with a FP reading, while tion mimicking PCa, and 3 cases with, also in retrospect, zonal location was not (P = 0.202, odds ratio 1.444, 95% MR imaging abnormalities but with normal histopathology CI 0.820–2.539) (Table 2). Overall, FP rate was 50% for results). These latter three cases (9%, 3/33) were interpreted 1 3 1048 Abdominal Radiology (2019) 44:1044–1051 Fig. 2 74-year-old patient, with a PSA of 28 ng/mL and a history of was retrospectively perceived as an example of a misclassification of previous negative transrectal-ultrasound-guided biopsy. PSA density a benign prostatic hyperplasia nodule in the transition zone, which of 0.13. Pitfall of benign prostatic hyperplasia nodule in the transition should have been classified as a PIRADS 2 lesion. b Apparent diffu- zone. Histopathology result revealed benign changes, no malignancy. sion coefficient map with the false positive lesion (straight arrow). c a T2-weighted image with a PIRADS 4 lesion (straight arrow) on Dynamic contrast-enhanced image, slightly asymmetric enhancement the initial read, reported by an inexperienced radiologist. This lesion (straight arrow) Fig. 3 72-year-old patient, with a PSA of 12 ng/mL and a history of arrow) on the initial read, reported by an inexperienced radiologist. previous negative transrectal-ultrasound-guided biopsy. PSA density This lesion was retrospectively perceived as an example of a misclas- of 0.19. Classic pitfall of the central zone compressed between the sification of an anatomic variant of the central zone. b Apparent dif- transition zone and peripheral zone. Histopathology result revealed fusion coefficient map with the false positive lesion (straight arrow). normal tissue, no malignancy, prostatic intraepithelial neoplasia or c Dynamic contrast-enhanced image, with no contrast enhancement prostatitis. a T2-weighted image with a PIRADS 4 lesion (straight (straight arrow) as likely due to technical inaccuracies (e.g., needle delivery, as MR imaging abnormalities but with normal histopa- inaccurate fusion of MR, and TRUS images). thology results. Of the 17 FP lesions in the PZ, five were For the experienced radiologist, of the 13 FP lesions high-grade PIN and/or prostatitis, and two were perceived in the TZ, 6 were perceived as misclassifications (of BPH as misclassifications (very little diffusion restriction). nodules and prominent anterior fibromuscular stroma). Ten were perceived, also in retrospect, as MR imaging Two were prospectively called as granulomateus prostati- abnormalities, but with normal histopathology results. tis (however, biopsy was still advised), five were perceived 1 3 Abdominal Radiology (2019) 44:1044–1051 1049 Altogether for both the TZ and PZ, 15 (50%, 15/30) FP to be missed at biopsy. The findings of this study have two lesions were likely due to technical inaccuracies. potential clinical implications. First, the relative underper- formance of inexperienced readers underlines that reading MP-MRI of the prostate should be reserved to experienced Discussion radiologists. The importance of subspecialty training is a well-acknowledged issue in breast and cardiac imaging. This To our knowledge, this study is the first to investigate the should not be different for MRI of the prostate. After internal FP findings of PIRADS (V2) 3-5 lesions in a pure clini- evaluation of the results of this study, we decided to reduce cal setting with emphasis on reader experience and zonal the number of radiologists reporting MP-MRI of the prostate location, in a patient population undergoing TRUS MRI and educate internally. Since then, only two of six special- fusion-guided targeted prostate biopsy. Our results show that ized abdominal radiologists have been reporting MP-MRI of inexperience is independently and significantly associated the prostate at our institution. The second potential implica- with a FP reading while zonal location is not. Furthermore, tion of our findings is that training sessions for inexperi- the evaluation of the TZ is mainly what makes the differ - enced radiologists should, along with a sufficient caseload ence between the inexperienced and experienced radiologist. with histopathological correlation, pay special attention to In general, the TZ is an anatomic area that is considered the TZ [24, 25], CZ, and low-grade prostatitis. difficult when interpreting MP-MRI of the prostate. The The overall detection (or true positive) rate of the expe- challenging aspect of detecting PCa in the TZ is probably rienced reader in this study (68% overall detection rate because of the heterogeneous appearance of the TZ which in PIRADS 3-5 lesions) was slightly higher than other is mainly due to the BPH it contains. Our study shows that expert centers performing MRI targeted prostate biopsy the evaluation of the TZ is even more difficult for the inex- (59%–62%) [26, 27]; however, both studies were performed perienced radiologist. We also found that, to a lesser degree, using PIRADS V1 which could also have accounted for other potential sources of a FP reading for the inexperienced the slight differences. In 2013, Bratan et al. [13] were the reader are anatomic variants of the CZ and low-grade prosta- first to investigate the FPs. However, this publication was titis. One of the main advantages of MP-MRI is its potential from the pre-PIRADS era, and they did not assess different to reduce unnecessary biopsies [17, 18]. Nonetheless, in case levels of experience. In another study published in 2017, of an inexperienced radiologist and a suspected TZ lesion, the issue of experience was addressed [5]. They reported this could lead to more unnecessary biopsies. The PIRADS that reader experience may help to reduce overcalling and 3 classification had a high number of FPs, for both the inex- avoid over targeting of lesions, which is in concordance with perienced and the experienced radiologists. Even though this our results. Nevertheless, none of these studies specifically is likely to be expected as PIRADS is a Likert scale, efforts looked at the combination of reader experience and zonal for reducing the number of unnecessary biopsies in this spe- location in FP lesions. cific group should be made. A possible solution could be Our study had several limitations. First, because of its the addition of PSA density [19]. Consequently, all lesions retrospective design, MRI protocols were heterogeneous with a PIRADS score of 3 and a cut off value of for example (i.e., different magnetic field strengths and slightly vary - below 0.15 could be refrained from biopsy. However, before ing MRI sequence settings), and possibly of suboptimal implementation this needs to be investigated further with quality (e.g., majority of scans were obtained with a 1.5T special attention to the number of missed significant cancers scanner; and no endorectal coil or the inability to control with this approach. Also noteworthy, FP rates do not only imaging parameters due to referrals from outside hospi- depend on factors accountable to the reader, but also on inac- tals). However, this is completely in line with routine clini- curacies related to the biopsy technique itself, most likely in cal practice, and this issue is also frequently encountered case of smaller lesions [20–22]. Nonetheless, when analyz- in multicenter studies [5, 18], which in fact increases the ing the FPs in the patient cohort of this study that were read generalizability of our results. Second, we were not able to by the inexperienced radiologists the majority appeared to be analyze false negatives and their relation to reader experi- classical examples of misclassification for the TZ as well as ence, which could have been done if the reference stand- for the PZ. In our study, we found that only in 9% of the FP ard would have been prostatectomy specimens or if long- cases for the inexperienced radiologists and in 50% of the FP term follow-up data would have been available. Third, this cases for the experienced radiologist technical issues could study did not investigate other factors potentially asso- have been the reason for a FP lesion. A study by Sheriden ciated with a FP reading, such as clinical variables and et al. [23] reported that 28% of the FP PIRADS 5 lesions apex-base location [23]. Fourth, this study consisted of could have been missed at biopsy. The difference with our patients who had received prior biopsy (either negative or study is that they only investigated PIRADS 5 lesions that with cancer but on active surveillance) possibly reducing are usually large and therefore technique-wise less likely the future generalizability of the results. It is expected 1 3 1050 Abdominal Radiology (2019) 44:1044–1051 OpenAccess This article is distributed under the terms of the Creative that as we move forward, more biopsy-naïve patients will Commons Attribution 4.0 International License (http://creativecom- undergo MP-MRI and targeted biopsy. Fifth, we did not mons.org/licenses/by/4.0/), which permits unrestricted use, distribu- include Gleason 3 + 3 lesions in our FP definition. In the tion, and reproduction in any medium, provided you give appropriate era of the increasing number of active surveillance candi- credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. dates, we are well aware of the importance of being able to discriminate between clinically significant (currently regarded as Gleason ≥ 3 + 4) and insignificant PCa (cur - References rently regarded as Gleason ≤ 3 + 3). However, including 3 + 3 in the FP definition on MP-MRI would, unfairly, 1. M.A. Haider, X. Yao, A. Loblaw, A. Finelli, Multiparametric Mag- imply that there are proven MRI features that are prospec- netic Resonance Imaging in the Diagnosis of Prostate Cancer: A tively able to discriminate Gleason 3 + 3 from 3 + 4. There Systematic Review, Clin. Oncol. (2016). https://doi.or g/10.1016/j. are retrospective and validation studies that have investi- clon.2016.05.003 2. B. Turkbey, A.M. Brown, S. Sankineni, B.J. Wood, P.A. Pinto, gated the relationship between ADC values and Gleason P.L. Choyke, Multiparametric Prostate Magnetic Resonance grades, and they have shown promising results [28]. Yet Imaging in the Evaluation of Prostate Cancer, CA Cancer J Clin. substantial overlaps between the different grades and ADC (2016). https ://doi.org/10.3322/caac.21333 values exist [28], making it not very useful in a clinical 3. D. Yakar, O.A. Debats, J.G.R. Bomers, M.G. Schouten, P.C. Vos, E. Van Lin, J.J. Fütterer, J.O. Barentsz, Predictive value of MRI in setting. Due to the lack of robust discriminatory imaging the localization, staging, volume estimation, assessment of aggres- features, publications addressing pitfalls on prostate MRI siveness, and guidance of radiotherapy and biopsies in prostate usually only contain examples of benign disease or FPs cancer, J. Magn. Reson. Imaging. (2012). https://doi.or g/10.1002/ related to more technical issues [10, 11]. Likewise, the jmri.22790 4. G.A. Sonn, R.E. Fan, P. Ghanouni, N.N. Wang, J.D. Brooks, A.M. clinically important decision between Gleason 3 + 3 and Loening, B.L. Daniel, K.J. To’o, A.E. Thong, J.T. Leppert, Pros- 3 + 4 can also be challenging on histopathology slides in tate Magnetic Resonance Imaging Interpretation Varies Substan- biopsy specimens. There is substantial interobserver vari- tially Across Radiologists, Eur. Urol. Focus. (2018). https ://doi. ability together with shortcomings of the Gleason grading org/10.1016/j.euf.2017.11.010 5. N.L. Hansen, B.C. Koo, F.A. Gallagher, A.Y. Warren, A. Doble, scoring system itself [29]. This is probably also the reason V. Gnanapragasam, O. Bratt, C. Kastner, T. Barrett, Compari- why some supposedly Gleason 3 + 3 tumors, even though son of initial and tertiary centre second opinion reads of mul- not very common, can metastasize or show aggressive tiparametric magnetic resonance imaging of the prostate prior to behavior (such as extraprostatic extension or even seminal repeat biopsy, Eur. Radiol. (2017). https://doi.or g/10.1007/s0033 0-016-4635-5 vesical invasion) [30, 31]. Furthermore, targeted biopsies 6. S. Müller, G. Lilleaasen, T.E. Sand, L. Løfsgaard, M. Estop- obtained from the cancer-suspicious lesions detected on Garanto, D. Helgø, P. Sund, V. Mygland, Poor reproducibility of MP-MRI might not accurately represent the true aggres- PIRADS score in two multiparametric MRIs before biopsy in men siveness found in a prostatectomy specimen [32–34], with with elevated PSA, World J. Urol. (2018). https://doi.or g/10.1007/ s0034 5-018-2252-4 some studies even mentioning percentages of upgrading 7. K. Garcia-Reyes, N.M. Passoni, M.L. Palmeri, C.R. Kauffman, from 3 + 3 to 3 + 4 in 67.4% of the cases [32]. So even K.R. Choudhury, T.J. Polascik, R.T. Gupta, Detection of pros- though there is an absolute clinical need for discriminat- tate cancer with multiparametric MRI (mpMRI): effect of dedi- ing significant from insignificant PCa, it is radiologically cated reader education on accuracy and confidence of index and anterior cancer diagnosis, Abdom. Imaging. (2015). https ://doi. not meaningful when assessing FPs. Future studies should org/10.1007/s0026 1-014-0197-7 focus on this unsolved issue. 8. B.G. Muller, J.H. Shih, S. Sankineni, J. Marko, S. Rais-Bahrami, In conclusion, inexperience of the radiologist is sig- A.K. George, J.J.M.C.H. de la Rosette, M.J. Merino, B.J. Wood, nificantly and independently associated with a FP read- P. Pinto, P.L. Choyke, B. Turkbey, Prostate Cancer: Interobserver Agreement and Accuracy with the Revised Prostate Imaging ing, while zonal location is not. Inexperienced radiologists Reporting and Data System at Multiparametric MR Imaging, have a significantly higher FP rate in the TZ. Radiology. (2015). https ://doi.org/10.1148/radio l.20151 42818 9. J. Yu, A.S. Fulcher, S.G. Winks, M.A. Turner, R.D. Clayton, M. Brooks, S. Li, Diagnosis of typical and atypical transition zone prostate cancer and its mimics at multiparametric prostate MRI, Compliance with ethical standards Br. J. Radiol. (2017). https ://doi.org/10.1259/bjr.20160 693 10. A.B. Rosenkrantz, S.S. Taneja, Radiologist, be aware: Ten pitfalls Conflict of interest All authors declare that they have no disclosures that confound the interpretation of multiparametric prostate MRI, or conflicts of interest. Am. J. Roentgenol. (2014). https ://doi.org/10.2214/ajr.13.10699 11. X. Kitzing, A. Prando, C. Varol, G.S. Karczmar, F. Maclean, A. Ethical approval This retrospective study was approved by the local Oto, Benign Conditions That Mimic Prostate Carcinoma: MR ethics committee and the need for informed consent was waived (Reg- Imaging Features with Histopathologic Correlation., Radiograph- istration Number 201700780). ics. 36 (2016) 162–175. https ://doi.org/10.1148/rg.20161 50030 12. V. Panebianco, F. Giganti, Y.X. Kitzing, F. Cornud, R. Campa, G. De Rubeis, A. Ciardi, C. Catalano, G. Villeirs, An update of 1 3 Abdominal Radiology (2019) 44:1044–1051 1051 pitfalls in prostate mpMRI: a practical approach through the lens multiparametric prostate MRI, Am. J. Roentgenol. (2018). https of PI-RADS v. 2 guidelines, Insights Imaging. (2018). https://doi. ://doi.org/10.2214/ajr.17.18680 org/10.1007/s1324 4-017-0578-x 24. A.P.S. Kirkham, P. Haslam, J.Y. Keanie, I. McCafferty, A.R. 13. F. Bratan, E. Niaf, C. Melodelima, A.L. Chesnais, R. Souchon, Padhani, S. Punwani, J. Richenberg, G. Rottenberg, A. Sohaib, F. Mège-Lechevallier, M. Colombel, O. Rouvière, Influence of P. Thompson, L.W. Turnbull, L. Kurban, A. Sahdev, R. Clem- imaging and histological factors on prostate cancer detection and ents, B.M. Carey, C. Allen, Prostate MRI: Who, when, and how? localisation on multiparametric MRI: A prospective study, Eur. Report from a UK consensus meeting, Clin. Radiol. (2013). https Radiol. (2013). https ://doi.org/10.1007/s0033 0-013-2795-0 ://doi.org/10.1016/j.crad.2013.03.030 14. J.C. Weinreb, J.O. Barentsz, P.L. Choyke, F. Cornud, M.A. 25. O. Akin, C.C. Riedl, N.M. Ishill, C.S. Moskowitz, J. Zhang, H. Haider, K.J. Macura, D. Margolis, M.D. Schnall, F. Shtern, C.M. Hricak, Interactive dedicated training curriculum improves accu- Tempany, H.C. Thoeny, S. Verma, PI-RADS Prostate Imaging - racy in the interpretation of MR imaging of prostate cancer, Eur. Reporting and Data System: 2015, Version 2, Eur. Urol. (2016). Radiol. (2010). https ://doi.org/10.1007/s0033 0-009-1625-x https ://doi.org/10.1016/j.eurur o.2015.08.052 26. T. Hambrock, D.M. Somford, C. Hoeks, S.A.W. Bouwense, H. 15. PRIAS guidelines. https ://www.prias -proje ct.org/. Accessed 21 Huisman, D. Yakar, I.M. van Oort, J.A. Witjes, J.J. Fütterer, J.O. Jan 2019 Barentsz, Magnetic Resonance Imaging Guided Prostate Biopsy in 16. J.I. Epstein, L. Egevad, M.B. Amin, B. Delahunt, J.R. Srigley, P.A. Men With Repeat Negative Biopsies and Increased Prostate Specic fi Humphrey, The 2014 international society of urological pathology Antigen, J. Urol. (2010). https://doi.or g/10.1016/j.juro.2009.10.022 (ISUP) consensus conference on gleason grading of prostatic car- 27. G. Gaziev, K. Wadhwa, T. Barrett, B.C. Koo, F.A. Gallagher, E. cinoma definition of grading patterns and proposal for a new grad- Serrao, J. Frey, J. Seidenader, L. Carmona, A. Warren, V. Gnana- ing system, Am. J. Surg. Pathol. (2016). https ://doi.org/10.1097/ pragasam, A. Doble, C. Kastner, Defining the learning curve pas.00000 00000 00053 0 for multiparametric magnetic resonance imaging (MRI) of the 17. H.U. Ahmed, A. El-Shater Bosaily, L.C. Brown, R. Gabe, R. prostate using MRI-transrectal ultrasonography (TRUS) fusion- Kaplan, M.K. Parmar, Y. Collaco-Moraes, K. Ward, R.G. Hindley, guided transperineal prostate biopsies as a validation tool, BJU A. Freeman, A.P. Kirkham, R. Oldroyd, C. Parker, M. Ember- Int. (2016). https ://doi.org/10.1111/bju.12892 ton, Diagnostic accuracy of multi-parametric MRI and TRUS 28. T. Hambrock, D.M. Somford, H.J. Huisman, I.M. van Oort, J.A. biopsy in prostate cancer (PROMIS): a paired validating con- Witjes, C.A. Hulsbergen-van de Kaa, T. Scheenen, J.O. Barentsz, firmatory study, Lancet. (2017). https ://doi.org/10.1016/s0140 Relationship between Apparent Diffusion Coefficients at 3.0-T -6736(16)32401 -1 MR Imaging and Gleason Grade in Peripheral Zone Prostate Can- 18. V. Kasivisvanathan, A.S. Rannikko, M. Borghi, V. Panebi- cer, Radiology. (2011). https ://doi.org/10.1148/radio l.11091 409 anco, L.A. Mynderse, M.H. Vaarala, A. Briganti, L. Budäus, G. 29. L. Egevad, A.S. Ahmad, F. Algaba, D.M. Berney, L. Boccon- Hellawell, R.G. Hindley, M.J. Roobol, S. Eggener, M. Ghei, A. Gibod, E. Compérat, A.J. Evans, D. Griffiths, R. Grobholz, G. Villers, F. Bladou, G.M. Villeirs, J. Virdi, S. Boxler, G. Robert, Kristiansen, C. Langner, A. Lopez-Beltran, R. Montironi, S. P.B. Singh, W. Venderink, B.A. Hadaschik, A. Ruffion, J.C. Hu, Moss, P. Oliveira, B. Vainer, M. Varma, P. Camparo, Standardi- D. Margolis, S. Crouzet, L. Klotz, S.S. Taneja, P. Pinto, I. Gill, zation of Gleason grading among 337 European pathologists, C. Allen, F. Giganti, A. Freeman, S. Morris, S. Punwani, N.R. Histopathology. (2013). https ://doi.org/10.1111/his.12008 Williams, C. Brew-Graves, J. Deeks, Y. Takwoingi, M. Ember- 30. H. Wenger, A.B. Weiner, A. Razmaria, G.P. Paner, S.E. Eggener, ton, C.M. Moore, MRI-Targeted or Standard Biopsy for Prostate- Risk of lymph node metastases in pathological gleason score≤6 Cancer Diagnosis (PRECISION study), N. Engl. J. Med. (2018). prostate adenocarcinoma: Analysis of institutional and population- https ://doi.org/10.1056/nejmo a1801 993 based databases, Urol. Oncol. Semin. Orig. Investig. (2017). https 19. F.A. Distler, J.P. Radtke, D. Bonekamp, et al., The value of PSA ://doi.org/10.1016/j.urolo nc.2016.08.004 density in combination with PI-RADS™ for the accuracy of pros- 31. O. Hassan, M. Han, A. Zhou, A. Paulk, Y. Sun, A. Al-Harbi, tate cancer prediction, J. Urol. (2017). https ://doi.org/10.1016/j. A. Alrajjal, F. Baptista dos Santos, J.I. Epstein, Incidence of juro.2017.03.130 Extraprostatic Extension at Radical Prostatectomy with Pure 20. A. Borkowetz, I. Platzek, M. Toma, T. Renner, R. Herout, M. Gleason Score 3 + 3 = 6 (Grade Group 1) Cancer: Implications Baunacke, M. Laniado, G. Baretton, M. Froehner, S. Zastrow, M. for Whether Gleason Score 6 Prostate Cancer Should be Renamed Wirth, Direct comparison of multiparametric magnetic resonance “Not Cancer” and for Selection Criteria for Active Surveillance, imaging (MRI) results with final histopathology in patients with J. Urol. (2018). https ://doi.org/10.1016/j.juro.2017.11.067 proven prostate cancer in MRI/ultrasonography-fusion biopsy, 32. T. Fukagai, T. Namiki, H. Namiki, R.G. Carlile, M. Shimada, H. BJU Int. (2016). https ://doi.org/10.1111/bju.13461 Yoshida, Discrepancies between Gleason scores of needle biopsy 21. C.P. Filson, S. Natarajan, D.J.A. Margolis, J. Huang, P. Lieu, F.J. and radical prostatectomy specimens, Pathol. Int. (2001). https :// Dorey, R.E. Reiter, L.S. Marks, Prostate cancer detection with doi.org/10.1046/j.1440-1827.2001.01207 .x magnetic resonance-ultrasound fusion biopsy: The role of system- 33. N.M. Corcoran, C.M. Hovens, M.K.H. Hong, J. Pedersen, R.G. atic and targeted biopsies, Cancer. (2016). https://doi.or g/10.1002/ Casey, S. Connolly, J. Peters, L. Harewood, M.E. Gleave, S.L. Gold- cncr.29874 enberg, A.J. Costello, Underestimation of Gleason score at prostate 22. J.P. Radtke, T.H. Kuru, S. Boxler, C.D. Alt, I. V. Popeneciu, biopsy ree fl cts sampling error in lower volume tumours, BJU Int. C. Huettenbrink, T. Klein, S. Steinemann, C. Bergstraesser, (2012). https ://doi.org/10.1111/j.1464-410x.2011.10543 .x M. Roethke, W. Roth, H.P. Schlemmer, M. Hohenfellner, B.A. 34. S. Mesko, L. Marks, O. Ragab, S. Patel, D.A. Margolis, D.J. Hadaschik, Comparative Analysis of Transperineal Template Demanes, M. Kamrava, Targeted Prostate Biopsy Gleason Score Saturation Prostate Biopsy Versus Magnetic Resonance Imaging Heterogeneity and Implications for Risk Stratification, Am. J. Targeted Biopsy with Magnetic Resonance Imaging-Ultrasound Clin. Oncol. Cancer Clin. Trials. (2018). https://doi.or g/10.1097/ Fusion Guidance, J. Urol. (2015). https ://doi.or g/10.1016/j.coc.00000 00000 00030 8 juro.2014.07.098 23. A.D. Sheridan, S.K. Nath, S. Aneja, J.S. Syed, J. Pahade, M. Publisher’s Note Springer Nature remains neutral with regard to Mathur, P. Sprenkle, J.C. Weinreb, M. Spektor, MRI-ultrasound jurisdictional claims in published maps and institutional affiliations. fusion targeted biopsy of prostate imaging reporting and data system version 2 category 5 lesions found false-positive at 1 3

Journal

Abdominal Radiology – Springer Journals

Published: Feb 8, 2019

There are no references for this article.