|Year : 2021 | Volume
| Issue : 1 | Page : 65-71
Development of an Indian nomogram for predicting extracapsular extension in prostate cancer
Chandran Ravi, Kalavampara V Sanjeevan, Appu Thomas, Ginil Kumar Pooleri
Department of Urology, Amrita Institute of Medical Sciences, Kochi, Kerala, India
|Date of Submission||21-Apr-2020|
|Date of Decision||01-Aug-2020|
|Date of Acceptance||23-Aug-2020|
|Date of Web Publication||1-Jan-2021|
Ginil Kumar Pooleri
Department of Urology, Amrita Institute of Medical Sciences, Kochi, Kerala
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Introduction: The aim of our study was to develop a new Indian nomogram to estimate pathologic extracapsular extension (ECE) risk in prostate cancer, by including PI-RADS v1-based magnetic resonance imaging (MRI) ECE risk score to the clinical variables used in the Partin nomogram (PN).
Materials and Methods: We analyzed 273 patients who underwent MRI of prostate and radical prostatectomy (RP). Univariate and multivariate logistic regression analyses were performed to identify predictors of ECE. We calculated the area under the receiver operating characteristic curve (AUC) for three variables used in PN and MRI ECE risk score, and a new nomogram was designed using binary logistic regression. Calibration curves assessed the agreement between the actual ECE risk and the predicted probability of the new nomogram.
Results: Out of 273 patients, 123 patients (45.1) had ECE on MRI, whereas 136 patients (49.8) had ECE on final pathology. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MRI for predicting ECE were 76.6, 66.9, 70.0, 73.9, and 71.7 (confidence interval 95), respectively. Multivariate logistic regression analyses showed that clinical T-stage (cT), Gleason score (GS), and MRI ECE risk score remained significant. The highest and the lowest values of the AUC for single variables were 0.748 (MRI ECE risk score) and 0.636 (cT stage), respectively, and AUC for PN was 0.67. New nomogram designed using R statistical package has higher predictive accuracy (0.826) compared to PN (0.67) and good calibration.
Conclusions: MRI adds incremental value to PN. A new Indian nomogram can help in the decision-making process of nerve-sparing RP. This nomogram should be used with caution as validation is pending and will require further studies.
|How to cite this article:|
Ravi C, Sanjeevan KV, Thomas A, Pooleri GK. Development of an Indian nomogram for predicting extracapsular extension in prostate cancer. Indian J Urol 2021;37:65-71
|How to cite this URL:|
Ravi C, Sanjeevan KV, Thomas A, Pooleri GK. Development of an Indian nomogram for predicting extracapsular extension in prostate cancer. Indian J Urol [serial online] 2021 [cited 2021 Jan 17];37:65-71. Available from: https://www.indianjurol.com/text.asp?2021/37/1/65/306035
| Introduction|| |
Carcinoma prostate (PCa) is the second most common cancer in men in India, and a large number of patients are diagnosed at a locally advanced stage unlike in other developed countries.,, Guidelines emphasize radical prostatectomy (RP) or, more recently, nerve-sparing radical prostatectomy (NSRP) as the surgery of choice for patients with localized disease, age less than 65 years, and with a mean life expectancy of at least 10 years, for better functional and oncological outcomes., PCa with extracapsular extension (ECE) is associated with decreased overall and cancer-specific survival following RP compared to organ-confined disease., Clinical staging based on physical examination has limited accuracy with 25-30% patients, with ECE being understaged preoperatively. The key to prognosticate oncologic outcomes and to determine the eligibility of patients for NSRP largely depends on ECE for which predictive models have been developed. Although models such as those developed by Partin and several other authors are widely used, there is a difference between their potential and actual predictive performance in clinical practice.,,,
Several studies have focused on exploring the incremental value of magnetic resonance imaging (MRI) parameters to these predictive tools to improve the predictability of tumor staging. Prostatic MRI reporting is standardized at present with the introduction of a structured uniform reporting and scoring system (PI-RADS) and ECE risk scoring. In 2015, Boesen et al. verified the ECE risk score in predicting ECE with relatively high accuracy. However, the cumulative effect or benefit of MRI parameters among different population groups is questionable.
As there is a paucity of data in the Indian subcontinent, the aim of our study is to explore the possible incremental value of MRI to the accuracy of PN in predicting the likelihood of ECE in PCa and to develop a new nomogram from an Indian population by integrating the data of MRI ECE risk score with the clinical variables of PN.
| Materials and Methods|| |
This is a retrospective single-institution study of patients who underwent primary RP between 2010 and 2019. Three hundred and thirty-eight patients with biopsy-proven primary PCa were treated with RP. After excluding 65 patients who did not have complete data or had received neoadjuvant therapy, the final study population consisted of 273 patients. The collected data included prebiopsy serum prostate-specific antigen (PSA) levels, clinical T-stage (cT) determined by digital rectal examination (DRE), Gleason score (GS) from transrectal ultrasound (TRUS)-guided prostate biopsy, MRI data, and histopathologic findings from the RP specimens of all patients. Currently, the widely used PN considers clinical T stage as per AJCC guidelines. Hence, in our nomogram, clinical T stage was considered purely on DRE findings. MRI findings were added to the existing variables of PN.
Based on Harrell's guidelines, when we planned to set a nomogram for binary situations (i.e., presence or absence of ECE), the minimum value of cases needed in either group is 10 times the number of variables used for predicting. In the present study, four variables were considered therefore a minimum of 40 patients in each outcome group was needed. In this study the number of patients in the groups were, 136 with ECE and 137 without ECE, above the required number.
Magnetic resonance imaging technique
MRI information in most cases was obtained before biopsy or at least 4 weeks after biopsy to reduce the biopsy artifacts. All patients underwent multiparametric MRI (mpMRI) (202 patients) and biparametric MRI (71 patients) using a 1.5 T or 3.0 T MRI scanner without endorectal coil. The MRI characteristics of the ECE of the tumor were assessed as follows at par with the ESUR prostate MR guidelines 2012: Score 0 – no sign of ECE, Score 1 – capsular abutment; Score 3 – capsular irregularity, retraction, or thickening; Score 4 – neurovascular bundle thickening and capsular signal loss or bulging; and Score 5 – direct sign of tumor tissue in the extraprostatic tissues.
MRI images of some patients done at outside centers were re-interpreted by radiologists at our institution in the absence of adequate details in the reports; certain MRI images were retrospectively analyzed for characterizing ECE risk score. In case of any discrepancy, an intradepartmental discussion was done to arrive at a unified consensus on the final report.
Pathology analysis and staging
All biopsy and surgical specimens were evaluated by two dedicated uropathologists. The location, primary and secondary GS, and the percentage of positive cores were recorded for every core of the TRUS-guided biopsy specimens. In case of any discrepancy, an intradepartmental discussion was done to arrive at the final report. In our study, 76 patients had positive surgical margins and capsular incision, mostly focal; they were included for analysis.
In the literature, two distinct definitions were considered for EPE – pT3a: the presence of tumor beyond the confines of the prostate without invasion of the seminal vesicles and whole EPE (wEPE): the presence of tumor beyond the confines of the prostate regardless of the status of seminal vesicles. In our study, we have considered wEPE.
Baseline descriptive statistics were used to present demographics, tumor, and MRI data. The sensitivity, specificity, positive predictive value, and negative predictive value of MRI (index test) for the diagnosis of histological ECE (reference standard) were calculated. The 2013 Partin nomogram (PN) was used to define the predictive probability of ECE. Univariate and multivariate logistic regression analyses were performed to identify predictors of ECE. The area under the receiver operating characteristic curve (AUC) values were calculated for PSA, cT, GS, and MRI ECE risk score. A New nomogram was created by binary logistic regression analysis using 300 bootstrap resamples to decrease the overfitting bias. Univariate and multivariate logistic regression analyses were used to arrive at relative significance of variables, and the nomogram was built based on R statistical package version 3.4 (R Foundation for Statistical Computing, Vienna, Austria).
New model was constructed consisting of four variables, namely PSA, cT, GS, and MRI ECE risk score. Furthermore, calibration curves assessed the agreement between the actual ECE risk and the predicted probability of the new nomogram in the current study with an intercept (ideally to be 0) and a slope (ideally to be 1). The predictive accuracy was determined for Partin ECE score and the new model, which were quantified with AUC values and compared against each other using the DeLong's method to determine if a significant difference was present. P < 0.05 was considered to be statistically significant. Statistical analysis was performed using SPSS version 22.0 (SPSS, Chicago, IL, USA) and the R statistical package version 3.4 (R Foundation for Statistical Computing, Vienna, Austria).
For collection of retrospective data, the ethics committee of Amrita Institute of Medical Sciences, Kochi institution had given approval through letter number IRB-AIMS-2020-201, 30-06-2020. All procedures adhered to the ethical guidelines of the Declaration of Helsinki and its amendments. Written permission was taken before the procedure and for the use of clinical details (without disclosing identity) for academic purpose. We confirm the availability and access of all original data reported in this study.
| Results|| |
The data of 273 patients were analyzed. The demographic and preoperative data and final histopathology details are tabulated in [Table 1]. ECE was found on MRI in 123 patients (45.1%), whereas 136 patients (49%) had ECE on final pathology. MRI had a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 76.6, 66.9, 70.0, 73.9, and 71.7 (confidence interval [CI] 95), respectively, in predicting ECE. All variables except age demonstrated a statistically significant difference in detecting ECE on final pathology on univariate analyses. Based on multivariate logistic regression analyses, cT, GS, and MRI ECE risk score remained significant predictors of ECE [Table 2]. AUC values were calculated for three variables used in PN and for MRI ECE risk score to assess the accuracy of predicting ECE [Table 3].
|Table 2: Factors that predict extracapsular extension based on univariate and multivariate analysis|
Click here to view
|Table 3: Area under the curve values for individual and combined factors in predicting extracapsular extension|
Click here to view
Using R software, the new nomogram for predicting ECE was constructed based on the logistic regression analysis [Figure 1] which included cT, PSA, GS, and MRI ECE risk score. Data of 12 patients with PSA more than 60 were excluded while creating the new nomogram. The nomogram is used by first locating the patient position on each predictor variable scale, which has corresponding prognostic points (top line on vertical axis). The points for each variable were added (total points), and the corresponding probability of ECE was estimated from the bottom line. For example, if serum PSA was 15 ng/ml, biopsy Gleason sum 7 (4 + 3), clinical stage T2a, and the ECE risk score of 4, the total score would be 190, and the probability of ECE by the new model would be 85%.
|Figure 1: New nomogram predicting extracapsular extension. Prostate.specific antigen (in ng/ml); Gleason score 1 = 3 + 3, 2 = 3 + 4, 3 = 4 + 3, 4 = Gleason sum >7; cT (clinical T stage) 1 = T1c, 2 = T2a, 3 = Higher than T2a; magnetic resonance imaging extracapsular extension risk score: 0 = no sign of extracapsular extension; 1 = capsular abutment; 3 = capsular irregularity, retraction, or thickening; 4 = neurovascular bundle thickening and capsular signal loss or bulging; 5 = direct sign of tumor tissue in the extraprostatic tissues|
Click here to view
For model validation, calibration was assessed. Calibration is interpreted by visual inspection of the plots of predicted probability of ECE versus actual ECE. [Figure 2] presents the calibration (300 bootstrap resamples) of the nomogram. The calibration plot reflects the nomogram performance. The horizontal axis is the prediction calculated with the nomogram, and the vertical axis is the actual presence of ECE. The dashed line represents the performance of an ideal nomogram in which the predicted outcome perfectly corresponded to the actual outcome. The performance of the nomogram was tested by plotting the predictions based on the nomogram without overfitting correction (apparent accuracy represented by dotted line) and bootstrap-corrected nomogram (scatter estimate of future accuracy represented by solid line). The nearness of the solid line and dashed line suggests that the nomogram-based predictions corresponded closely with actual outcomes.
|Figure 2: Calibration of the nomogram (300 bootstrap resamples). Twelve patients with prostate.specific antigen more than 60 ng/ml were excluded while constructing nomogram. Hence, n = 261|
Click here to view
In [Figure 3], the 95% confidence intervals (CIs) of nomogram calibration belt did not cross the diagonal bisector line, and the P value in calibration test is 0.414. The predicted probability of the nomogram was consistent with the actual probability, which suggested that new nomogram had strong concordance performance, and the calibration of the prediction model is good.
|Figure 3: The 45° bisector represents the identity between predicted probabilities and observed responses. The 80% and 95% confidence level calibration belts are plotted in light and dark gray, respectively. The test's P value, the sample size n, and the polynomial order m of the calibration curve are reported in the top left corner. In the lower right quadrant, the times the calibration belt significantly deviates from the bisector using 80% and 95% confidence levels are reported|
Click here to view
The AUC values [Table 3] were 0.67 and 0.82, respectively, for the PN and the new Indian nomogram (PSA, cT, GS, and MRI ECE risk score); on DeLong's test, comparisons of AUC values between the two models were statistically significant (P < 0.00023).
| Discussion|| |
Distinguishing organ-confined diseases from those with ECE is often an important step in the management of PCa. ECE prediction assists in various stages of patient counseling and surgical planning, wherein if ECE is suspected, the cavernous nerves, which are responsible for erectile function, are often resected to enhance the likelihood of achieving negative margins.
Despite the fact that MRI is widely adopted as a useful diagnostic and staging tool for PCa, its sensitivity in predicting ECE appears to be low to intermediate (48.7–81), and this was substantiated in this study, wherein the sensitivity was 73.4%. A few studies have compared the accuracy of MRI with standard clinical parameters such as variables of PN and have concluded that mpMRI is better for staging PCa.,,,,,
As the causative factors of prostate cancer differ epidemiologically and biologically from more developed nations, the predictions based on the nomograms plotted for the population in these countries may be different for an Indian cohort. The adaptability of such models to other geographic areas was poor. This led us to develop a new nomogram based on the data of prostate cancer in the Indian population by adding the MRI-based ECE risk score to clinical variables.
The reported incidence of ECE in PCa has varied widely among different studies. Rocco et al. noted 28.4 (1803 of 6360) ECE in one American center, whereas Satake et al. from Japan reported 41.1 (146 of 354 patients), and Chen et al. reported more than 55 of cT = T3. Gandaglia et al. study conducted across five European centers reported ECE rates of 54. The ECE rate in our study was 49.8 (136 of 273 patients). In the literature, two distinct definitions were considered for ECE – (1) pT3a: the presence of tumor beyond the confines of the prostate without invasion of the seminal vesicles and (2) whole ECE (wECE): the presence of tumor beyond the confines of the prostate regardless of the status of seminal vesicles. In our study, we have considered wEPE; hence, ECE rates are high, and these rates are consistent with data from other centers from India and Europe.,,
Nomograms, in the form of user-friendly graphical interfaces, assist in clinical decision making by transforming statistical predictive models into a single numerical estimate tailored to the individual patient. Several authors have developed various statistical tools to predict the pathological stage, especially after the use of PN. However, majority of those nomograms lack appropriate external validation. The 2013 PN had AUC of 0.702. Similarly, nomograms created using variables similar to PN by Egawa et al. from Japan and Song et al. from Korea had AUC of 0.793 and 0.626, respectively., Memorial Sloan-Kettering cancer center (MSKCC) nomograms generated by Ohori et al. from the USA based on cT, PSA, Gleason sum, percentage of positive cores, and percentage of cancer involvement and Steuber et al. from Europe reported predictive accuracies of 0.806 and 0.840, respectively, for side-specific ECE., The nomogram created by Satake et al. from Japan in 2010 with the same clinical data as MSKCC acquired an AUC value of 0.797. The predictive accuracy of PN for ECE in our cohort was 0.679.
Advantages of mpMRI, widely used in these days, as an efficient imaging tool for prostate cancer staging were discussed by Sciara et al. and were supported by Gupta et al. who argued that mpMRI is better for staging prostate cancer than the Partin table. In 2015, Boesen et al. analyzed the diagnostic performance of preoperative mpMRI ECE risk score, and it showed an AUC of 0.86 with moderate inter-reader agreement (K = 0.45). The predictive value with MRI ECE risk score was 0.748 in the current study. On applying the new nomogram which we have developed by adding the MRI ECE risk score to clinical variables of PN (PSA, cT, and GS), the predictive accuracy was found to have enhanced from 0.67 for PN to 0.82 (P = 0.00023). mpMRI reporting by a specialized radiologist with mpMRI prostate experience can provide good accuracy.,,,,,,, In view of the limited availability of radiologists trained in mpMRI, our proposed nomogram could provide incremental value to accuracy in staging ECE.
Our study is based on results from a single-center, retrospective cohort. It is a small sample size compared to the Partin tables. There might be selection bias since the pathology and MRI revisions were not available for the entire cohort. There is a lack of uniformity in MRI timing. Besides, patients undergoing both biparametric and mpMRI were included in the study. However, since not all centers are equipped with mpMRI, and many still rely on biparametric MRI, we found this more reflective of contemporary practice in India. All GS were not based on standard sextant biopsies since some of our patients had biopsies already done elsewhere before reaching our center, and hence, there was heterogeneity in the technique, number of cores taken, and the reporting format. Future studies are needed to validate our model with other data sets. This nomogram should be used with caution as validation was not done in other Indian populations/centers.
| Conclusions|| |
MRI adds incremental value to the existing validated risk stratification tool and provides significant additional ability for predicting ECE in prostate cancer staging. We constructed a nomogram for predicting ECE based on the results of cT, PSA, GS, and MRI ECE risk score in Indian patients. The nomogram provides a good prediction of ECE.
We would like to thank Dr. Rishi Gupta (MBBS, MD, DNB) for the effort and contribution to statistical analysis.
Financial support and sponsorship: Nil.
Conflicts of interest: There are no conflicts of interest.
| References|| |
Jain S, Saxena S, Kumar A. Epidemiology of prostate cancer in India. Meta Gene 2014;2:596-605.
Kulkarni JN, Singh DP, Bansal S, Makkar M, Valsangkar R, Siddaiah AT, et al
. Retropubic radical prostatectomy: Clinicopathological observations and outcome analysis of 428 consecutive patients. Indian J Urol 2011;27:337-44.
] [Full text]
Dogra PN, Javali TD, Singh P, Kumar R, Seth A, Gupta NP, et al
. Perioperative outcome of initial 190 cases of robot-assisted laparoscopic radical prostatectomy - A single-center experience. Indian J Urol 2012;28:159-63. [Full text]
Sanda MG, Cadeddu JA, Kirkby E, Chen RC, Crispino T, Fontanarosa J, et al
. Clinically localized prostate cancer: AUA/ASTRO/SUO guideline. Part II: Recommended approaches and details of specific care options. J Urol 2018;199:990-7.
Mottet N, Bellmunt J, Bolla M, Briers E, Cumberbatch MG, De Santis M, et al
. EAU-ESTRO-SIOG guidelines on prostate cancer. Part 1: Screening, diagnosis, and local treatment with curative intent. Eur Urol 2017;71:618-29.
Mikel Hubanks J, Boorjian SA, Frank I, Gettman MT, Houston Thompson R, Rangel LJ, et al
. The presence of extracapsular extension is associated with an increased risk of death from prostate cancer after radical prostatectomy for patients with seminal vesicle invasion and negative lymph nodes. Urol Oncol 2014;32:26.e1-7.
Tollefson MK, Karnes RJ, Rangel LJ, Bergstralh EJ, Boorjian SA. The impact of clinical stage on prostate cancer survival following radical prostatectomy. J Urol 2013;189:1707-12.
Cooperberg MR, Lubeck DP, Mehta SS, Carroll PR; CaPSURE. Time trends in clinical risk stratification for prostate cancer: Implications for outcomes (data from CaPSURE). J Urol 2003;170:S21-5.
Partin AW, Yoo J, Carter HB, Pearson JD, Chan DW, Epstein JI, et al
. The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 1993;150:110-4.
Eifler JB, Feng Z, Lin BM, Partin MT, Humphreys EB, Han M, et al
. An updated prostate cancer staging nomogram (Partin tables) based on cases from 2006 to 2011. BJU Int 2013;111:22-9.
Ohori M, Kattan MW, Koh H, Maru N, Slawin KM, Shariat S, et al
. Predicting the presence and side of extracapsular extension: A nomogram for staging prostate cancer. J Urol 2004;171:1844-9.
Steuber T, Graefen M, Haese A, Erbersdobler A, Chun FK, Schlom T, et al
. Validation of a nomogram for prediction of side specific extracapsular extension at radical prostatectomy. J Urol 2006;175:939-44.
Barentsz JO, Richenberg J, Clements R, Choyke P, Verma S, Villeirs G, et al
. ESUR prostate MR guidelines 2012. Eur Radiol 2012;22:746-57.
Boesen L, Chabanova E, Løgager V, Balslev I, Mikines K, Thomsen HS. Prostate cancer staging with extracapsular extension risk scoring using multiparametric MRI: A correlation with histopathology. Eur Radiol 2015;25:1776-85.
Rocco B, Sighinolfi MC, Sandri M, Eissa A, Elsherbiny A, Zoeir A, et al
. Is extraprostatic extension of cancer predictable? A review of predictive tools and an external validation based on a large and a single center cohort of prostate cancer patients. Urology 2019;129:8-20.
Partin AW, Borland RN, Epstein JI, Brendler CB. Influence of wide excision of the neurovascular bundle(s) on prognosis in men with clinically localized prostate cancer with established capsular penetration. J Urol 1993;150:142-6.
Gupta RT, Faridi KF, Singh AA, Passoni NM, Garcia-Reyes K, Madden JF, et al
. Comparing 3-T multiparametric MRI and the Partin tables to predict organ-confined prostate cancer after radical prostatectomy. Urol Oncol 2014;32:1292-9.
Gupta RT, Brown AF, Silverman RK, Tay KJ, Madden JF, George DJ, et al
. Can radiologic staging with multiparametric MRI enhance the accuracy of the partin tables in predicting organ-confined prostate cancer? Am J Roentgenol 2016;207:87-95.
Raskolnikov D, George AK, Rais-Bahrami S, Turkbey B, Siddiqui MM, Shakir NA, et al
. The role of magnetic resonance image guided prostate biopsy in stratifying men for risk of extracapsular extension at radical prostatectomy. J Urol 2015;194:105-11.
Feng TS, Sharif-Afshar AR, Smith SC, Miller J, Nguyen C, Li Q, et al
. Multiparametric magnetic resonance imaging localizes established extracapsular extension of prostate cancer. Urol Oncol 2015;33:109.e15-22.
Schieda N, Quon JS, Lim C, El-Khodary M, Shabana W, Singh V, et al
. Evaluation of the European Society of Urogenital Radiology (ESUR) PI-RADS scoring system for assessment of extra-prostatic extension in prostatic carcinoma. Eur J Radiol 2015;84:1843-8.
Augustin H, Fritz GA, Ehammer T, Auprich M, Pummer K. Accuracy of 3-Tesla magnetic resonance imaging for the staging of prostate cancer in comparison to the Partin tables. Acta Radiol 2009;50:562-9.
Yadav R, Arora S, Sachdeva M, Gupta NP. Assessment of the performance of Partin's nomogram (2007) in contemporary Indian cohort. Indian J Urol 2016;32:199-203.
] [Full text]
Satake N, Ohori M, Yu C, Kattan MW, Ohno Y, Miyakawa A, et al
. Development and internal validation of a nomogram predicting extracapsular extension in radical prostatectomy specimens. Int J Urol 2010;17:267-72.
Chen R, Xie L, Xue W, Ye Z, Ma L, Gao X, et al
. Development and external multicenter validation of Chinese Prostate Cancer Consortium prostate cancer risk calculator for initial prostate biopsy. Urol Oncol 2016;34:416.e1-7.
Gandaglia G, Ploussard G, Valerio M, Mattei A, Fiori C, Roumiguié M, et al
. The key combined value of multiparametric magnetic resonance imaging, and magnetic resonance imaging-targeted and concomitant systematic biopsies for the prediction of adverse pathological features in prostate cancer patients undergoing radical prostatectomy. Eur Urol 2020;77:733-41.
Iasonos A, Schrag D, Raj GV, Panageas KS. How to build and interpret a nomogram for cancer prognosis. J Clin Oncol 2008;26:1364-70.
Egawa S, Suyama K, Arai Y, Matsumoto K, Tsukayama C, Kuwao S, et al
. A study of pretreatment nomograms to predict pathological stage and biochemical recurrence after radical prostatectomy for clinically resectable prostate cancer in Japanese men. Jpn J Clin Oncol 2001;31:74-81.
Song C, Kang T, Ro JY, Lee MS, Kim CS, Ahn H. Nomograms for the prediction of pathologic stage of clinically localized prostate cancer in Korean men. J Korean Med Sci 2005;20:262-6.
Sciarra A, Barentsz J, Bjartell A, Eastham J, Hricak H, Panebianco V, et al
. Advances in magnetic resonance imaging: How they are changing the management of prostate cancer. Eur Urol 2011;59:962-77.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]