Comparison of the Diagnostic Performance of PI-RADS V1 and PI-RADS V2 for the Detection of Prostate Cancer: A Meta-Analysis Ying He1, Ruochen Cong2, Jie Zhou2, Zhenyu Xu3, Jushun Yang2, Lin Wang2, Jing Xiao 4, Bosheng He 2,5* Purpose: In order to comprehensively determine the diagnostic accuracy of the Prostate Imaging Reporting and Data System version 1 (PI-RADS V1) and PI-RADS version 2 (PI-RADS V2) in prostate cancer (PCa) diagnosis. Materials and Methods: The literatures were screened from the databases, including the Pubmed, Embase, Web of science and Cochrane Library up to January 20th, 2020. The meta-analysis was conducted by Meta-DiSc and quality assessment was performed by using the QUADAS. Furthermore, the sensitivity, specificity, likelihood ratio (LR), diagnostic odds ratio (DOR), as well as receiver operating curve (ROC) related to diagnostic accuracy were pooled. Results: A total of 6 articles containing 814 participants (379 patients) were included in the study. For PI-RADS V1, the combined sensitivity, specificity, PLR, NLR and DOR were 0.82 (95% CI: 0.77-0.85), 0.81 (95% CI: 0.77- 0.85), 4.58 (95% CI: 2.55-8.22), 0.24 (95% CI: 0.18-0.34) and 24.00 (95% CI: 10.38-55.51). With regard to PI- RADS V2, the combined sensitivity, specificity, PLR, NLR and DOR were 0.88 (95% CI: 0.84-0.91), 0.81 (95% CI: 0.77-0.84), 4.34 (95% CI: 1.98-9.49), 0.16 (95% CI: 0.08-0.32) and 33.39 (95% CI: 15.05-74.05), respectively. Furthermore, except that the sensitivity of PI-RADS V2 was significantly greater than that of PI-RADS V1 (P = 0.027), there was no remarkably difference in other indicators for the diagnosis of PCa between the two versions. Conclusion: Both PI-RADS V1 and PI-RADS V2 showed good diagnostic performance for PCa diagnosis; more- over, there was no difference in the diagnostic effect between them. Keywords: PI-RADS V1; PI-RADS V2; Prostate cancer; Multiparametric MRI INTRODUCTION Prostate cancer (PCa), accounting for 20% of all can-cers diagnosed, has been the second most common cancer with more than 1.1 million new cancer cases an- nually(1,2). Usually, due to the some symptoms of PCa with similar to those of other diseases, such as prosta- titis, benign prostatic hyperplasia, cystitis and urinary tract infection, the rate of the early detection and resec- tion rate of PCa is only approximately 10–20% (3). Cur- rently, method for PCa detection includes prostate-spe- cific antigen test, digital rectal examination (DRE), and biopsies. Specifically, the elevated serum prostate-spe- cific antigen (PSA) level is the most frequently used biomarker for PCa detection (4), but it has been criticized because of the lack of specificity diagnostic accuracy (5,6). Multi-parametric magnetic resonance imaging (Mp- MRI), including anatomic T2-weighted imaging (T2 1Department of Ultrasound, the Tumor Hospital of Nantong University, Nantong 226361, China. 2Department of Radiology, Affiliated Hospital 2 of Nantong University, Nantong 226001, China. 3Department of Ultrasound, the Second People’s Hospital of Nantong, Nantong 226002, China. 4Department of Epidemiology and Medical Statistics, School of Public Health, Nantong University, Nantong 226019, China 5Clinical Medicine Research Center, Affiliated Hospital 2 of Nantong University, Nantong 226001, China *Correspondence: Department of Radiology, Affiliated Hospital 2 of Nantong University, No.6 Hai Er Xiang North Road, Nantong, Jiangsu 226001, China. Tel: +86-0513-85061220. Fax: +86-0513-85061220. E-mail: boshenghe@126.com. Received August 2019 & Accepted April 2020 W) with functional diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE), is characterized by noninvasive, multi-parameter, high soft tissue reso- lution, as well as the high subject tolerance. Hence, it has been widely used in clinical localization, qualita- tive and staging diagnosis, as well as risk and prognosis evaluation of PCa (7-9). Mp-MRI can provide the better diagnostic accuracy in the detection of PCa, and accord- ingly, the standardized reporting system for it has been published. In 2012, the first version (V1) of the Pros- tate Imaging Reporting and Data System (PI-RADS) was published by the European Society of Urogenital Radiology (ESUR)(10). Generally, PI-RADS V1 scores showed high diagnostic accuracy for PCa diagnosis (11,12); however, the clear instructions on how to integrate the overall score were lacking. Hence, the updated PI- RADS version 2 (V2) was established by the American College of Radiology, which has improved some of the UROLOGICAL ONCOLOGY Urology Journal/Vol 18 No. 1/ January-February 2021/ pp. 51-57. [DOI: 10.22037/uj.v16i7.5532 limitations of PI-RADS V1. Due to the differences be- tween PI-RADS V1 and PI-RADS v2 scoring methods, many literatures have studied the diagnostic efficacy of the two methods in PCa, but the results are not com- pletely consistent(13-15). Hence, the aim of the present study was to comprehensively analyze the diagnostic value of PI-RADS V1 and PI-RADS V2 in PCa detec- tion by using a meta-analysis, which will provide a ba- sis for PCa screening. Author Public Country Study Gold N/n* Prostate Control Age Group TP FP FNTN TN Year year standard zone characteristics (years) (PZ/TZ) Auer T 2016 Austria NA Histo- 103/32 89/14 BPH 63.0 ± 8.0 PI-RADS v1 84 1 19 31 pathological PI-RADS v2 82 4 21 28 Feng 2016 China 2013.6- Radiography 150/251 95/55 Non-PCa 64.4(34-88) PI-RADS v1 127 48 23 203 ZY 2015.7 PI-RADS v2 144 40 6 211 Kasel- 2016 Germany 2013.7- Histo- 31/51 NA Benign 65(48-81) PI-RADS v1 22 17 9 34 seibert M 2015.3 pathological PI-RADS v2 24 10 7 41 Polanec 2016 Austria 2011.6- Radio- 33/32 25/8 Benign 65.3 PI-RADS v1 31 21 2 11 S 2015.9 graphy,Histopathological (62.3-87.4) PI-RADS v2 32 12 1 20 Tewes 2016 Germany 2012.12- Histo- 31/23 26/5 Non-PCa 69.6 ± 9.6 PI-RADS v1 24 2 7 21 S 2014.12 pathological PI-RADS v2 28 4 3 19 Wang 2018 China 2015.9- Histo- 31/46 0/31 BPH 72.3±7.5 PI-RADS v1 21 2 10 44 XM 2016.7 pathological PI-RADS v2 23 3 8 41 Table 1. Characteristics of the included literatures. *: Prostate cancer/ Control; TP: true positives; TN: true negatives; FP: false positives; FN: false negatives; QUADAS: quality assessment tool of diagnostic accuracy studies; PI-RADS: prostate imaging reporting and data system; PZ, peripheral zone; TZ, transition zone; NA: not available; BPH: benign prostatic hyperplasia. Figure1. Flow diagram of the articles included in this systematic review. PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al. Urological Oncology 52 MATERIALS AND METHODS Literature search The literature searches were conducted on the basis of the databases, including Pubmed (http://www.ncbi. nlm.nih.gov/pubmed), Embase (http://www.embase. com), Web of science (http://webofknowledge.com) and Cochrane Library (http://www.cochranelibrary. com/) up to January 20th, 2020. The keywords were as follows: (“prostate cancer” OR “prostatic carcinoma” OR “carcinoma of prostate” OR (Prostatic Neoplasms)) AND (“prostate imaging reporting and data system” OR “PI-RADS V1” OR “PI-RADS V2”) AND (Diag- nostic OR diagnose OR sensitivity OR specificity). The language was restricted to English. Inclusion and exclusion criteria Inclusion criteria was as follows: 1) patients with PCa (P); 2) English literature published on PI-RADS V2; 3) PI-RADS V1 diagnostic effect in patients with PCa; 4) can provide true positive number, false positive num- ber, false negative number and true negative number of participants; 5) diagnostic test for the diagnostic value of PCa. Exclusion criteria was as following: 1) the study with incomplete data that cannot be used for statistical anal- ysis; 2) review, letters, and other non-treatises of liter- ature. In addition, for the literature with repeated pub- lication or the same population data used in multiple studies, only the latest study or the one with the most complete information was included. Data extraction and quality assessment All data from included studies was retrieved by two in- dependent researchers: first author, year of publication, study year, country, the gold standard in the diagnosis of PCa, age composition of included participants, the number of true positive (TP), false positive (FP), true negative (TN), and false negative (FN) results for either PI-RADS V1 and PI-RADS V2 analysis. Quality assessment of the included studies was per- formed by using the quality assessment of diagnostic accuracy studies tool (QUADAS)(16). 11 items were evaluated according to the three criteria of "yes" (meet- ing this standard), "no" (not meeting or not being men- tioned), and "unclear" (partly meeting or not getting enough information from the literature). Specifically, once there was a difference of opinion in the process of literature data extraction and quality evaluation, a con- sensus will be reached after a group discussion with the third researcher. PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al. Table 2. Results of diagnostic analysis. Indicators PI-RADS V1(95%CI) PI-RADS V2(95%CI) Z P Sensitivity 0.82(0.77-0.85) 0.88(0.84-0.91) 2.321 0.027 Specificity 0.81(0.77-0.85) 0.81(0.77-0.84) 0.074 0.941 PLR 4.58(2.55-18.22) 4.34(1.98-9.49) 0.100 0.920 NLR 0.24(0.18-0.34) 0.16(0.08-0.32) 1.087 0.277 DOR 24.00(10.38-55.51) 33.39(15.05-74.05) 0.496 0.620 Figure 2. The pooled sensitivity (A) and specificity (B), PLR (C), NLR (D), DOR (E), and SROC (F) estimates for PI-RADS V1 detection of PCa patients. uj.v16i7.5532/10.22037 Vol 18 No 1 January-February 2021 53 Urological Oncology 54 Statistical analysis Meta-analysis was conducted with Meta-DiSc (ver- sion 1.4), and the effect indicators, including sensitiv- ity, specificity, positive likelihood ratio (PLR), nega- tive likelihood ratio (NLR), and diagnostic odds ratio (DOR). The threshold effect was evaluated by the spear- man correlation coefficient of the sensitivity logarithm and the (1-specificity) logarithm. And the heterogeneity was determined based on the Cochran’s Q test and the I2 index(17): If significant heterogeneity was detected (P < 0.05, I2 > 50%), the combined effect value was calculat- ed by the random effect model (dersimonian-laird); oth- erwise, fixed-effect model was used (mantel-haenszel) (18). The differences between PI-RADS V1 and V2 in diagnostic indicators were determined with Z test, and the publication bias of Egger's test was conducted by using Stata software. RESULTS Characteristics of the included literatures According to the flow diagram for literature selection (Figure 1), a total of 652 studies were preliminari- ly screened from Pubmed (n = 323), Embase (n=307) and Cochrane Library (n=22), including 234 duplicated articles. After title and abstract screen, 345 unrelated researches were excluded. Next, through the full text reading, finally 6 articles with 814 participants (379 patients) were included in the study (13,14,19-22). As illustrated in Table 1, the characteristics of 6 studies were summarized. The data showed that all included articles were published between 2016 and 2018, and the location included the Australia, China and Germa- ny. Additionally, the basic characteristics of demogra- phy revealed average age of all participants was 63-72, among which the elderly were the majority. Further- more, PI-RADS V1 ≥ 10 or ≥ 4 and PI-RADS V2 ≥ 4 or ≥ 3 had been regarded as the cut-off values in the diagnosis of PCa. As shown in Supplementary table 1, the quality of in- cluded articles was evaluated according to 11 items of QUADAS. The results showed that the bias of the in- cluded studies was small, indicating that the methodo- logical quality was high. The combination of quantitative data . Spearman correlation coefficient for V1 and V2 were 0.429 (P = 0.397) and 0.600 (P = 0.428), respective- ly, indicating there was no threshold effect and other statistics should be combined. The results with random effects model (DerSimonian-Laird) revealed that the combined sensitivity was 0.82 (95% CI: 0.773-0.853), specificity was 0.81 (95% CI: 0.77-0.85), PLR was 4.58 (95% CI: 2.55-8.22), NLR was 0.24 (95% CI: 0.18-0.34) and DOR was 24.00 (95%CI: 10.38-55.51) for PI-RADS V1 (Figure 2A-2E). Similarly, based on the random effects model, the combined sensitivity, specificity, PLR, NLR and DOR for PI-RADS V2 were 0.88 (95% CI: 0.84-0.91), 0.81(95% CI: 0.77-0.84), 4.34 (95% CI: 1.98-9.49), 0.16 (95% CI: 0.08-0.32) and 33.39 (95%CI: 15.05-74.05) (Figure 3A-3E), respec- tively. The SROC curve was symmetric and random effect model (DerSimonian-Laird) was adopted. The area un- der curves of SROC for V1 and V2 were 0.8938 (Q = 0.8244) and 0.9154 (Q = 0.8482) (Figure 2F and Fig- ure 3F). After the Z test, the data revealed that there was no statistical difference between PI-RADS V1 and V2 in the diagnosis of PCa (AUC, Z = 0.557, P = 0.577).Q, Z = 0.568, P = 0.570). Taken together, sen- sitivity of PI-RADS V2 for the detection of PCa was obviously higher than that of PI-RADS V1 (Z = 2.213, P = 0.027; Table 2), suggesting the diagnostic effect of PI-RADS V2 was superior to V1. However, the dif- Figure 3. The pooled sensitivity (A) and specificity (B), PLR (C), NLR (D), DOR (E), and SROC (F) estimates for PI-RADS V2 detec- tion of PCa patients. PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al. ference of specificity, PLR, NLR or DOR between PI- RADS V1 and V2 was not significant, respectively (all, P > 0.05). Publication bias The Egger's test indicated that there was no publication bias in the diagnosis of PCa in PI-RADS V1 and V2 (t = 0.22, P = 0.823; t = 0.85, P = 0.428), which proves that our results are reliable. DISCUSSION In the present study, we for the first time compared the diagnostic performance of mpMRI with PI-RADS V1 and V2 in the PCa detection. The results of meta-anal- ysis demonstrated that both PI-RADS V1 and V2 both presented high diagnostic value. Furthermore, in addi- tion to the fact that PI-RADS V2 was more sensitive than V1, there was no difference in the other indica- tors between the two versions. Therefore, in general, there was no difference in the diagnostic effect between the two versions. Till now, although several studies have been reported to conduct the meta-analyses of PI-RADS, the difference of diagnostic effect between PI-RADS V1 and V2 has not been reported till now. For example, Maggi(23) and Zhai(24) et al only separate- ly investigate the diagnostic performance of PI-RADS 3; moreover, Barkovich et al(25) only quantitatively and qualitatively assesses the methodologic heterogeneity of the PI-RADSv2 literature and estimate the propor- tions of Gleason scores (GSs) diagnosed across PI- RADSv2 categories. Because of the high diagnostic accuracy for PCa de- tection and reproducible interpretation, mpMRI has been widely used by urologists. Hence, comprehensible and clearly defined criteria for standardized analysis of MRI for PCa should be urgent. As the initial ver- sion, PI-RADS V1 has been reported to have a good inter-observer agreement and high diagnostic accuracy (26,27). Compared with the sequence of T2-WI, DWI, and DCE in PI-RADS V1 was considered to have equal discriminatory power, PI-RADS V2 introduces the con- cept of "dominant sequence", which believes that DWI is the key sequence of PZ and T2-WI is the dominant sequence in TZ (28,29). And if there is no evidence of invasive behavior, the main difference between a find- ing with a score of 4 and that with a score of 5 on T2-W and DWI is a diameter less than 1.5 cm or equal /great- er than 1.5 cm(30,31). In recent years, numerous studies have validated the value of PI-RADS V2 but, as ex- pected, have also identified a number of ambiguities and limitations, some of which have been document- ed in the literature with potential solutions offered(32). It has been reported PI-RADS V2 in clinical practice retains higher accuracy over systematic TRUS biopsies for PCa diagnosis(33). Till now, a series of studies have reported some key differences between PI-RADS V1 and V2, but the comparisons between the two versions have been controversial. For instance, Thomas et al(13) revealed that PI-RADS V1 showed a significantly larg- er discriminative ability for the detection of PCa, due to the more false negative results in PI-RADS v2. In- versely, Moritz(14) has demonstrated PI-RADS V2 could be a reliable reporting system for PCa assessment. And Hoffmann et al(34) reports PI-RADS V2 is reproducible between radiologists but does not have improved ac- curacy for diagnosing anterior tumors of the prostate when compared to PI-RADS V1. In the present study, the results of a comprehensive comparison with me- ta-analysis suggested that there was no statistical dif- ference between PI-RADS V1 and V2 in the diagnosis of PCa. Furthermore, the sensitivity of PI-RADS V2 for PCa diagnosis in our study was significantly higher than that of PI-RADS V1, but there was no significant difference in specificity, PLR, NLR and DOR between PI-RADS V1 and PI-RADS V2. Actually, the summed PI-RADS V2 outperformed V1 in the assessments of PCa has been understood as a consequence of cancer location. Briefly, PI-RADS V2 has been reported to be the pref- erable method to evaluate the transitional zone (TZ) due to a higher sensitivity, whereas PI-RADS V1 performed better in (peripheral zone) PZ(35). Hence, in the process of mpMRI, the versions of PI-RADS should be selected based on the tumor site. Usually, the inter-reader agreement has been regard- ed as one of the most important limitations in the use of mpRMI. A previous study reports that the low in- ter-user agreement of mpMRI may reduce the overall applicability of this methodology in all centers(36). In the present study, the results of the inter-reader agree- ment analysis (Supplementary table 2) showed that in- ter-reader agreement of PI-RADS V2 and PI-RADS V1 were different in the included studies. In fact, based on the the heterogeneity of the sensitivity or specificity the included literature, we can also find that the diagnostic accuracy of PI-RADS V2 and PI-RADS V1 varies in different studies. Heterogeneity has been regarded as a critical element in meta-analysis(37). In this study, we discussed the di- agnostic value of PI-RADS V1 and V2 in PCa detec- tion, and significant heterogeneity was detected among the overall pooled analyses mainly due to the follow- ing aspects: 1) differences in race, country and region; 2) differences in living habits, cultural exchanges and living environment; 3) differences caused by age, sam- ple size and other factors. Furthermore, there were no significant publication bias between the included stud- ies, which suggested that the data of our meta-analysis are reliable. However, this study still had limitations; for example, due to the relatively small number of lit- eratures and incomplete stratification information, the study was unable to obtain the source of its heteroge- neity. CONCLUSIONS In summary, the results with meta-analysis showed the differences of diagnostic accuracy of PI-RADS V1 and V2 were not significant for detection of PCa. However, to further verify the results, a larger cohort from mul- ti-center institutions are still needed. ACKNOWLEDGEMENT This work was supported by The social development fund of Nantong(Program No. MS12018086 and No.MS22019013) and Research Topics of Teaching Re- form of Nantong University(Program No. 2016B103). CONFLICT OF INTEREST The authors declare that they have no competing inter- ests. Appendix: https://journals.sbmu.ac.ir/urolj/index.php/uj/library PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al. Vol 18 No 1 January-February 2021 55 Urological Oncology 56 Files/downloadPublic/16 REFERENCES 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet- Tieulent J, Jemal A. Global cancer statistics, 2012. Ca A Cancer J Clin. 2015;65:87-108. 2. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359-E86. 3. Helmstaedter L, Riemann JF. Pancreatic cancer-EUS and early diagnosis. Langenbecks Arch Surg. 2008;393:923-7. 4. Greene KL, Albertsen PC, Babaian RJ, et al. Prostate specific antigen best practice statement: 2009 update. J Urology. 2009;182:2232-41. 5. Adhyam M, Gupta AK. A Review on the Clinical Utility of PSA in Cancer Prostate. Indian J Surg Oncol. 2012;3:120-9. 6. Bokhorst LP, Bangma CH, Leenders GJLHV, et al. Prostate-specific Antigen–Based Prostate Cancer Screening: Reduction of Prostate Cancer Mortality After Correction for Nonattendance and Contamination in the Rotterdam Section of the European Randomized Study of Screening for Prostate Cancer. Eur Urol. 2014;65:329-36. 7. Cash H, Maxeiner A, Stephan C, et al. The detection of significant prostate cancer is correlated with the Prostate Imaging Reporting and Data System (PI-RADS) in MRI/transrectal ultrasound fusion biopsy. World J Urol. 2016;34:525-32. 8. Pummer K, Rieken M, Augustin H, Gutschi T, Shariat SF. Innovations in diagnostic imaging of localized prostate cancer. World Urol. 2014;32:881-90. 9. Axel H, Bastian PJ, Joaquim B, et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. Eur Urol. 2014;65:124- 37. 10. Iu PP. ESUR prostate MR guidelines. Eur Radiol. 2013;23:2320-1. 11. Leonardo KB, Geert L, Baris T, Emerson Leandro G, Barentsz JO. Prostate Cancer: The European Society of Urogenital Radiology Prostate Imaging Reporting and Data System Criteria for Predicting Extraprostatic Extension by Using 3-T Multiparametric MR Imaging. Radiology. 2015;276:479-89. 12. Junker D, Quentin M, Nagele U, et al. Evaluation of the PI-RADS scoring system for mpMRI of the prostate: a whole-mount step- section analysis. World J Urol. 2015;33:1023- 30. 13. Auer T, Edlinger M, Bektic J, et al. Performance of PI-RADS version 1 versus version 2 regarding the relation with histopathological results. World JUrol. 2017;35:687-93. 14. Kasel-Seibert M, Lehmann T, Aschenbach R, et al. Assessment of PI-RADS v2 for the Detection of Prostate Cancer. Eur J Radiol. 2016;85:726-31. 15. Wang X, Bao J, Ping X, et al. The diagnostic value of PI-RADS V1 and V2 using multiparametric MRI in transition zone prostate clinical cancer. Oncol Lett. 2018;16:3201-6. 16. Whiting P, Rutjes AW, Dinnes J, Reitsma J, Bossuyt PM, Kleijnen J. Development and validation of methods for assessing the quality of diagnostic accuracy studies. Health Technology Assessment. 2004;8:1-234. 17. J L, JP I, CH S. Quantitative synthesis in systematic reviews. Ann Internal Medicine. 1997;127:820-6. 18. Javier Z, Victor A, Alfonso M, Khalid K, Arri C. Meta-DiSc: a software for meta-analysis of test accuracy data. BMC Med Res methodol. 2006;6:31. 19. X W, J B, X P, et al. The diagnostic value of PI-RADS V1 and V2 using multiparametric MRI in transition zone prostate clinical cancer. Oncol Lett. 2018;16:3201-6. 20. Feng ZY, Wang L, Min XD, Wang SG, Wang GP, Cai J. Prostate Cancer Detection with Multiparametric Magnetic Resonance Imaging: Prostate Imaging Reporting and Data System Version 1 versus Version 2. Chin Med J. 2016;129:2451-9. 21. Polanec S, Helbich TH, Bickel H, et al. Head- to-head comparison of PI-RADS v2 and PI- RADS v1. EurJ Radiol. 2016;85:1125-31. 22. Tewes S, Mokov N, Hartung D, et al. Standardized Reporting of Prostate MRI: Comparison of the Prostate Imaging Reporting and Data System (PI-RADS) Version 1 and Version 2. PloS One. 2016;11:e0162879. 23. Maggi M, Panebianco V, Mosca A, et al. Prostate imaging reporting and data system 3 category cases at multiparametric magnetic resonance for prostate cancer: a systematic review and meta-analysis. Eur Urol Focus. 2019;S2405-4569:30169-5. 24. Zhai L, Fan Y, Meng Y, Feng X, Yu W, Jin J. The role of Prostate Imaging Reporting and Data System score in Gleason 3+ 3 active surveillance candidates enrollment: a diagnostic meta-analysis. Prostate Cancer P D. 2019;22:235-43. 25. Barkovich EJ, Shankar PR, Westphalen AC. A systematic review of the existing Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) literature and subset meta- analysis of PI-RADSv2 categories stratified by Gleason scores. AM J Roentgenol. 2019;212:847-54. 26. Hamoen EHJ, Maarten DR, J Alfred W, Barentsz JO, Rovers MM. Use of the Prostate Imaging Reporting and Data System (PI- RADS) for Prostate Cancer Detection with Multiparametric Magnetic Resonance Imaging: A Diagnostic Meta-analysis. Eur Urol. 2015;67:1112-21. 27. Schimmöller L, Quentin M, Arsov C, et al. Inter-reader agreement of the ESUR score for prostate MRI using in-bore MRI-guided biopsies as the reference standard. Eur Radiol. 2013;23:3185-90. PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al. Vol 18 No 1 January-February 2021 57 28. Schimmöller L, Quentin M, Arsov C, et al. Predictive power of the ESUR scoring system for prostate cancer diagnosis verified with targeted MR-guided in-bore biopsy. Eur J Radiol. 2014;83:2103-8. 29. Raphaëlle RP, Pierre M, François C, et al. Prostate Imaging Reporting and Data System and Likert Scoring System: Multiparametric MR Imaging Validation Study to Screen Patients for Initial Biopsy. Radiology. 2015;275:458-68. 30. Purysko AS, Rosenkrantz AB, Barentsz JO, Weinreb JC, Macura KJ. PI-RADS Version 2: A Pictorial Update. Radiographics. 2016;36:1354-72. 31. Muller BG, Shih JH, Sankineni S, et al. Prostate Cancer: Interobserver Agreement and Accuracy with the Revised Prostate Imaging Reporting and Data System at Multiparametric MR Imaging. Radiology. 2015;277:741-50. 32. Turkbey B, Rosenkrantz AB, Haider MA, et al. Prostate imaging reporting and data system version 2.1: 2019 update of prostate imaging reporting and data system version 2. Eur Urol. 2019;76:340-351. 33. Padhani AR, Weinreb J, Rosenkrantz AB, Villeirs G, Turkbey B, Barentsz J. Prostate imaging-reporting and data system steering committee: PI-RADS v2 status update and future directions. Eur Urol. 2019;75:385-96. 34. Hoffmann R, Logan C, O’Callaghan M, Gormly K, Chan K, Foreman D. Does the Prostate Imaging-Reporting and Data System (PI-RADS) version 2 improve accuracy in reporting anterior lesions on multiparametric magnetic resonance imaging (mpMRI)? Int Urology Nephrol. 2018;50:13-9. 35. Polanec S, Helbich TH, Bickel H, et al. Head- to-head comparison of PI-RADS v2 and PI- RADS v1. Eur J Radiol. 2016;85:1125-31. 36. Cormio L, Cindolo L, Troiano F, et al. Development and internal validation of novel nomograms based on benign prostatic obstruction-related parameters to predict the risk of prostate cancer at first prostate biopsy. Front Oncol. 2018;8:438. 37. Turner RM, Jonathan D, Clarke MJ, Thompson SG, Julian Pt H. Predicting the extent of heterogeneity in meta-analysis, using empirical data from the Cochrane Database of Systematic Reviews. Int J Epidemiol. 2012;41:818-27. PI-RADS V1 and PI-RADS V2 for prostate cancer-He et al.