Prostate‑specific antigen density and preoperative MRI findings as predictors of biochemical recurrence in high‑risk and very high‑risk prostate cancer
- Authors:
- Published online on: May 18, 2023 https://doi.org/10.3892/ol.2023.13870
- Article Number: 284
-
Copyright: © Yang et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Prostate cancer is the most common urological cancer among older men. In 2020, 1,414,259 patients were newly diagnosed with prostate cancer globally (1). Surgical treatment, which primarily comprises radical prostatectomy, provides reasonable disease control for prostate cancer, but ~25% of patients experience disease recurrence (2). The most frequently used test to monitor disease occurrence is the detectable prostate-specific antigen (PSA) value after surgery. Patients with high-risk prostate cancer after prostatectomy have a particularly high chance of being diagnosed with biochemical recurrence (BCR) following two consecutive PSA measurements of >0.2 ng/ml. Patients with BCR have a greater risk of disease progression and mortality (2–6), and the early occurrence of BCR is associated with a poor prognosis (6–9).
Patients defined as high risk according to the National Comprehensive Cancer Network (NCCN) guidelines have an increased risk of disease progression and cancer-specific mortality (10). Furthermore, one study reported that ~50% of high-risk patients had BCR within 1 year (6), which indicates that patients categorized as high risk require considerable attention and a strong follow-up strategy, particularly those with early BCR. The present study evaluated the risk factors associated with high-risk and very high-risk patients as classified by the NCCN guidelines after radical prostatectomy. These guidelines are designed to help urologists to identify patients requiring more aggressive therapeutic programs of clinical care.
Materials and methods
Study sample
The present retrospective study enrolled patients diagnosed with prostate cancer who received radical prostatectomy at Taichung Veterans General Hospital (Taichung, Taiwan) from January 1, 2009 to June 30, 2020. The patients were all male with a median age of 73 years (range 52–85 years). All patients had received laparoscopic radical prostatectomy or robotic-assisted prostatectomy performed by a single surgical team. Two radiologists interpreted MRI images based on Prostate Imaging-Reporting and Data System scores (11). The follow-up duration for all patients was ≥6 years. Patients without follow-up data or with pathologic stage T3bN2M0 were excluded. All patient data were extracted from medical records, including demographics, surgery type, pelvic lymph node dissection (PLND) status, PSA values, Gleason scores, and pathological and image examination data at diagnosis and within 1 year after prostatectomy. The pathological and image examination in the present study was based on the standard Tumor-Node-Metastasis staging system defined by the published American Joint Committee on Cancer 8th edition (12). The detected range of primary tumor invasion was different between the pathological and image examination. Pathological examination evaluated T2a, T2b, T3a and T3b staging of prostate cancer, but image examination evaluated T2b, T2c, T3a and T3b staging. T2a indicates that tumors were located in <50% of one side of the prostate; T2b indicates that tumors were located in >50% of one side of the prostate; T2c indicates that tumors were located in both sides of the prostate; T3a indicates that tumors had broken through the capsule of the prostate gland; T3b indicates that tumors had spread into the tubes that carry semen; N0M0 indicates that tumors had not spread to nearby lymph nodes or elsewhere in the body; any T stage indicates that tumors may or may not have grown into tissues near the prostate (12). Pathological tumor percentage was calculated as a proportion of tumor volume and prostate volume. The formula used was as follow: Vc=Vol × Pc/100, where Pc is pathological tumor percentage (%); Vc is the volume of tumor tissues (cc); and Vol is the prostate volume (cc). The unit of volume measurement detected by planimetry software [Photoshop CC 2017 (version 18; Adobe Systems, Inc.)] is indicated as cc (13,14). PSA density (ng/ml2) was calculated as total PSA value (ng/ml) divided by prostate volume (ml). Detectable PSA is defined as a PSA value >0.1 ng/ml within 3 months after prostatectomy. Disease control is defined as the last PSA value being less than the PSA value during salvage therapy. The patients were assigned to high-risk (T3a, Gleason score >8 or PSA >20 ng/ml) and very high-risk groups (T3b-T4, primary Gleason pattern 5, two or three high-risk features or >4 biopsy cores with Gleason score >8) after radical prostatectomy according to the NCCN guidelines.
Ethical considerations
The study protocol was approved by the Institutional Review Boards I&II of Taichung Veterans General Hospital (ref. no. CE21174A; approval date, June 23, 2022) and followed the approved guidelines. No informed consent was required of the participants because no identifying patient information was included and the data were analyzed anonymously.
Primary outcome
The primary outcome was the BCR of prostate cancer. BCR is defined as a PSA value that had decreased to <0.1 ng/ml but then increased to >0.2 ng/ml more than 3 months after prostatectomy, during salvage therapy.
Statistical analysis
The normal distribution of continuous data was checked with the Shapiro-Wilk test. Data that did not follow a normal distribution are presented as the median with the interquartile range (IQR) in parentheses, and were analyzed using the Wilcoxon rank-sum test. Categorical data were analyzed using the Chi-square or Fisher's exact test, as appropriate, and are presented as n (%). Cox regression models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) of the risk of BCR in patients with undetectable PSA. Univariate and multivariate logistic regression models were used to estimate the odds ratios (ORs) and 95% CIs for postoperative detectable PSA rate, post-salvage therapy disease control rate and the BCR <1-year rate. Receiver operating characteristic (ROCs) curves were plotted to assess the value of PSA density and pathological tumor percentage in the prediction of outcomes. The cut-off point was calculated according to the maximum Youden index. Each area under the curve (AUC) was calculated, with higher AUCs indicating higher predictive performance. The full model ROC curve is a linear regression model that includes the values of all explanatory variables, such as age, surgical type, PLND, PSA density, pathological tumor percentage, imaging stage and pathologic stage. A two-sided P<0.05 was considered to indicate a statistically significant difference. Data management and statistical analyses were conducted using SAS version 9.4 software (SAS Institute, Inc.).
Results
Demographic and clinical characteristics of the patients
A total of 307 patients were included in the study, with a median age of 73 years (IQR, 68–77 years). According to the NCCN guidelines, 187 patients were classified in the high-risk group and 120 patients were classified in the very high-risk group. Table I presents the baseline characteristics of the study population. In the very high-risk group, the initial PSA [29.66 (17.79-45.19) vs. 21.00 (10.10-30.99) ng/ml; P<0.001], PSA density [0.74 (0.38-1.20) vs. 0.48 (0.24-0.80) ng/ml2; P<0.001] and pathological tumor percentage [35.0 (15.0-67.5) vs. 20.0 (10.0-40.0)%; P<0.001] were significantly higher than those in the high-risk group. The very high-risk group had a higher proportion of patients undergoing extended PLND, higher initial and postoperative Gleason grades, and a lower proportion of patients with imaging stage T2bN0M0/T2cN0M0 and pathologic stage T2aN0M0/T2bN0M0 and T3aN0M0 when compared with the high-risk group (all P<0.001).
Comparisons of the clinical outcomes between the very high-risk and high-risk groups are presented in Table II. In the very high-risk group, the BCR rates in the patients with undetectable PSA (52.70 vs. 30.00%; P=0.001) and postoperative detectable PSA (38.33 vs. 19.79%; P=0.004) were higher than the respective rates in the high-risk group. Furthermore, fewer patients did not receive salvage therapy (19.61 vs. 46.31%; P<0.001) and fewer patients were receiving continence medication 4 and 12 weeks after Foley catheter removal (37.82 vs. 41.94% and 50.42 vs. 53.76%; P=0.048) in the very high-risk group compared with the high-risk group.
Risk factors of BCR
The results of univariate and multivariate Cox proportional hazards regression analysis of BCR are shown in Table III. A total of 224 patients had undetectable PSA. Univariate analysis showed that patients in the very high-risk group had a 2.13-fold risk of BCR compared with those in the high-risk group (95% CI, 1.38-3.27; P=0.001), with a significant difference between the two groups. Significant differences were also detected in other variables, including extended PLND, PSA density, pathological tumor percentage, imaging stage and pathologic stage between the two groups [all P<0.001, with the exception of PSA density (P=0.003) and imaging stage any N1 vs. T2bN0M0/T2cN0M0 (P=0.014)]. After adjusting for other covariates, the PSA density [adjusted HR (aHR), 1.46, 95% CI, 1.01-2.39; P=0.042], pathological tumor percentage (aHR, 1.02, 95% CI, 1.01-1.23; P<0.001), imaging stage (T3aN0M0 vs. T2bN0M0/T2cN0M0: aHR, 1.96, 95% CI, 1.01-3.79; P=0.047; T3bN0M0 vs. T2bN0M0/T2cN0M0: aHR, 2.59, 95% CI, 1.12-6.01; P=0.026) and pathologic stage (any N1 vs. T2aN0M0/T2bN0M0: aHR, 3.21, 95% CI, 1.25-8.23; P=0.015) remained significantly associated with BCR. However, no significant difference in BCR was found between the two NCCN risk groups (aHR, 1.25; 95%CI, 0.76-2.92; P=0.372). The BCR-free survival curve based on the multivariate model is shown in Fig. 1. The median BCR-free survival for patients in the very high-risk group was 3.62 years; however, the high-risk group had a higher survival rate than the very high-risk group. The medium BCR-free survival for patients in the high-risk group was 4.66 years.
 | Table III.Univariate and adjusted HRs for BCR in the high-risk and very high-risk groups of patients with undetectable PSA. |
Table IV presents the results of the univariate and multivariate logistic regression analysis of postoperatively detectable PSA. The analysis of the NCCN-defined very high-risk group vs. the high-risk group showed statistical significance for postoperatively detectable PSA but only prior to adjustment for other covariates [OR: 2.52, 95% CI: 1.51-4.22; P<0.001; adjusted OR (aOR): 1.31, 95% CI: 0.68-2.50, P=0.419]. Other variables were also associated with this outcome, including extended PLND, PSA density, pathological tumor percentage, imaging stage and pathologic stage. Following adjustment for other covariates, the pathological tumor percentage (aOR, 1.03, 95% CI, 1.02-1.04; P<0.001) and higher pathologic stage (T3bN0M0 vs. T2aN0M0/T2bN0M0: aOR, 3.66, 95% CI, 1.14-11.70; P=0.029; any N1 vs. T2aN0M0/T2bN0M0: aOR, 4.47, 95% CI, 1.29-15.44; P=0.018) maintained their association with a significantly higher risk of postoperatively detectable PSA. Notably, the univariate OR for postoperatively detectable PSA in the very high-risk group was significantly different compared with that of the high-risk group (P<0.001).
| Table IV.Univariate and adjusted ORs for postoperatively detectable PSA levels in the high-risk and very high-risk groups. |
Univariate and multivariate logistic regression analyses of the disease control rate after salvage therapy are shown in Table V. A total of 162 patients were included in this analysis. As the number of patients in imaging stage T3bN0M0 and any N1 was small, the two groups were combined for analysis. None of the analyzed variables showed significant associations in univariate and adjusted ORs between groups. The univariate and multivariate logistic regression analysis for time to BCR <1-year is shown in Table VI. A total of 84 patients were included in this analysis. No significant associations were found between these variables and BCR within 1 year (Table VI).
| Table V.Univariate and adjusted ORs for disease control after salvage therapy in the high-risk and very high-risk groups. |
ROC curves were constructed for BCR and postoperative detectable PSA level, and the cut-off points of the full model were calculated, along with those for PSA density and the pathological tumor percentage (Fig. 2). The AUC of the full model (AUC: 0.80, 95% CI: 0.74-0.86) was significantly higher than those of PSA density (AUC: 0.65, 95% CI: 0.57-0.72; P<0.001) and pathological tumor percentage (AUC: 0.71, 95% CI: 0.64-0.78; P=0.002) for BCR. The cut-off points for PSA density and pathological tumor percentage were 0.33 and 35.04, respectively. The AUC and 95% CI of the full model, PSA density and pathological tumor percentage for postoperative detectable PSA levels were 0.78 (0.72-0.84), 0.59 (0.52-0.66) and 0.75 (0.69-0.81), respectively. No significant difference was found in the AUC between the full model and pathological tumor percentage (P=0.12), whereas the full model had a higher AUC than the PSA density (P<0.001). The cut-off points for PSA density and pathological tumor percentage were 0.35 and 29.99, respectively.
Discussion
The present study evaluated the outcomes of patients defined as high risk and very high risk after radical prostatectomy according to the NCCN guidelines, and found that patients in the very high-risk category displayed lower BCR-free survival than those in the high-risk category, and that patients with undetectable PSA had a high risk of BCR and detectable PSA levels postoperatively. Therefore, the patients were re-stratified to identify further criteria that may help in the screening of patients who are suitable for radical prostatectomy. The results of the present study suggest that NCCN-defined high-risk patients with pathologic stage T3bN0M0 have a high risk of BCR <1 year, which signals to urologists that radical prostatectomy should be considered as a salvage therapy option for disease control.
The present study also found that in patients in the NCCN-defined high-risk and very high-risk categories, BCR was associated with PSA density, imaging stage T3aN0M0 and T3bN0M0, and pathologic stage any N1, while pathologic stage T3bN0M0 and any N1 were associated with postoperatively detectable PSA. In the literature, the independent risk factors associated with BCR among high-risk patients include pathological Gleason score ≥8 (15–18), postoperative positive surgical margins (15), initial PSA value (16), seminal vesicle invasion (17), >50% positive cores (18) and several risk factors (15,18). In addition, another study showed via multivariate analysis that high preoperative PSA levels, Gleason score ≥7, lymph node invasion and positive margins were associated with early and late BCR (19). These findings indicate that NCCN-defined high-risk patients with a Gleason score ≥7 deserve merit attention from urologists regarding follow-up strategies, regardless of their postoperative PSA values. It must be noted that patients at very high-risk according to the NCCN guidelines are not good candidates for radical prostatectomy because of the relatively short time to the onset of BCR. However, patients categorized as high-risk could potentially benefit from radical prostatectomy due to tolerable trifecta results.
The rate of BCR <1 year in the high-risk and very high-risk patients in the present study was 54% (167/307), which is comparable with the reported rate of 47.5% (6). An early BCR occurring <1 year after radical prostatectomy has been regarded as a predictor of metastatic progression and prostate cancer mortality (6–9), so is therefore worthy of particular attention. However, none of the variables included in the present study was found to be associated with BCR <1 year after surgery, and no associations were noted between surgery type, PLND status and BCR. The therapeutic value of PLND status with regard to BCR remains controversial in the literature. Cao et al (20) found no significant differences between laparoscopic radical prostatectomy, robotic-assisted radical prostatectomy and open radical prostatectomy in postoperative BCR and overall complication rates. Preisser et al (21) reported that extended PLND had no significant effect on oncological outcomes in high-risk patients (21). However, lymph node invasion is associated with BCR <1 year after surgery (6). Further prospective, randomized studies are required to clarify the extent of node dissection that is necessary for satisfactory oncological outcomes.
A previous study outlined the current state of evidence for prostate cancer screening and early detection and summarized the recent recommendations of various guidelines (22). It reviewed the global public health burden and risk factors for prostate cancer with clinical implications as a screening tool, along with screening, novel biomarkers and magnetic resonance imaging (22). By contrast, the present study outlines key practice points for primary care physicians and provides a simple model to facilitate shared decision-making conversations. A decision tree for use as a diagnostic model for high-risk and very high-risk prostate cancer is provided in Fig. 3.
The present study has several limitations. First, it is a retrospective study with inherent limitations such as only certain variables being available, restriction to a given time period that rules out long-term follow-up, and the possible existence of bias. Second, the database was derived from a single center with a relatively small sample size, which may not allow the results to be generalized to other populations or locations. The imaging stages may also be variable because the MRI images were interpreted by different radiologists.
In conclusion, after radical prostatectomy, patients defined as very high-risk according to the NCCN guidelines have low BCR-free survival and patients with undetectable PSA have a high risk of BCR. Preoperative MRI imaging stage T3aN0M0 and T3bN0M0 and pathologic stage any N1 have a higher risk of BCR after prostatectomy, and PSA density predicts short-term BCR after prostatectomy. These findings may help in the screening of patients to identify those requiring active therapeutic protocols.
Acknowledgements
The authors would like to thank Ms. Szu-Han Ho (Department of Surgery, Division of Urology at Taichung Veterans General Hospital, Taichung, Taiwan) for assisting with data curation.
Funding
Funding: No funding was received.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Authors' contributions
CKY, CRY, CLC, HCH, KYC, SSW, CCC, SCW, CYL and SCH contributed to the conception and design of the study, and drafting the manuscript. YCO and JRL contributed to the conception and design of the study, acquisition of data and drafting the manuscript. CSC contributed to the conception and design of the study, drafting the manuscript and clinical analysis. CFH contributed to the conception and design of the study, acquisition of data, drafting and critical revision of the manuscript, and clinical analysis. CKY and CYL confirm the authenticity of the raw data. All authors read and approved the final version of the manuscript.
Ethics approval and consent to participate
The study protocol was approved by Institutional Review Board I&II of Taichung Veterans General Hospital (ref. no. CE21174A) and was carried out according to the approved guidelines. No informed consent was required because the data were analyzed anonymously and no identifying information associated with the participants was included.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
References
|
Wang L, Lu B, He M, Wang Y, Wang Z and Du L: Prostate cancer incidence and mortality: Global status and temporal trends in 89 countries from 2000 to 2019. Front Public Health. 10:8110442022. View Article : Google Scholar : PubMed/NCBI | |
|
Boorjian SA, Eastham JA, Graefen M, Guillonneau B, Karnes RJ, Moul JW, Schaeffer EM, Stief C and Zorn KC: A critical analysis of the long-term impact of radical prostatectomy on cancer control and function outcomes. Eur Urol. 61:664–675. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
Van den Broeck T, van den Bergh RCN, Arfi N, Gross T, Moris L, Briers E, Cumberbatch M, De Santis M, Tilki D, Fanti S, et al: Prognostic value of biochemical recurrence following treatment with curative intent for prostate cancer: A systematic review. Eur Urol. 75:967–987. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Kasuya G, Ishikawa H, Tsuji H, Nomiya T, Makishima H, Kamada T, Akakura K, Suzuki H, Shimazaki J, Haruyama Y, et al: Significant impact of biochemical recurrence on overall mortality in patients with high–risk prostate cancer after carbon–ion radiotherapy combined with androgen deprivation therapy. Cancer. 122:3225–3231. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Freedland SJ, Humphreys EB, Mangold LA, Eisenberger M, Dorey FJ, Walsh PC and Partin AW: Risk of prostate cancer–specific mortality following biochemical recurrence after radical prostatectomy. JAMA. 294:433–439. 2005. View Article : Google Scholar : PubMed/NCBI | |
|
Venclovas Z, Stanionis M, Matjosaitis AJ and Milonas D: Significance of time until PSA recurrence on clinical progression of surgically treated high-risk prostate cancer patients. Eur Urol Suppl. 18:e206910.1016/S1569-9056(19)31500-3. 2019. View Article : Google Scholar | |
|
Pak S, You D, Jeong IG, Kim YS, Hong JH, Kim CS and Ahn H: Time to biochemical relapse after radical prostatectomy and efficacy of salvage radiotherapy in patients with prostate cancer. Int J Clin Oncol. 24:1238–1246. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Briganti A, Karnes RJ, Gandaglia G, Spahn M, Gontero P, Tosco L, Kneitz B, Chun FK, Zaffuto E, Sun M, et al: Natural history of surgically treated high-risk prostate cancer. Urol Oncol. 33:163.e7–13. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Pompe RS, Gild P, Karakiewicz PI, Bock LP, Schlomm T, Steuber T, Graefen M, Huland H, Tian Z and Tilki D: Long-term cancer control outcomes in patients with biochemical recurrence and the impact of time from radical prostatectomy to biochemical recurrence. Prostate. 78:676–681. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
Garg H, Dursun F, Alsayegh F, Wang H, Wu S, Liss MA, Kaushik D, Svatek RS and Mansour AM: Revisiting current national comprehensive cancer network (NCCN) high-risk prostate cancer stratification: A national cancer database analysis. Prostate Cancer Prostatic Dis. 14:10.1038/s41391–022-00621-7. 2023. | |
|
Turkbey B, Rosenkrantz AB, Haider MA, Padhani AR, Villeirs G, Macura KJ, Tempany CM, Choyke PL, Cornud F, Margolis DJ, et al: Prostate imaging reporting and data system version 2.1: 2019 Update of prostate imaging reporting and data system version 2. Eur Urol. 76:340–351. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Buyyounouski MK, Choyke PL, McKenney JK, Sartor O, Sandler HM, Amin MB, Kattan MW and Lin DW: Prostate cancer-major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin. 67:245–253. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Vollmer RT: Percentage of tumor in prostatectomy specimens: A study of American Veterans. Am J Clin Pathol. 131:86–91. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Ito Y, Udo K, Vertosick EA, Sjoberg DD, Vickers AJ, Al-Ahmadie HA, Chen YB, Gopalan A, Sirintrapun SJ, Tickoo SK, et al: Clinical usefulness of prostate and tumor volume related parameters following radical prostatectomy for localized prostate cancer. J Urol. 201:535–540. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Otsuka M, Kamasako T, Uemura T, Takeshita N, Shinozaki T, Kobayashi M, Komaru A and Fukasawa S: Factors predicting biochemical recurrence after radical prostatectomy among patients with clinical T3 prostate cancer. Jpn J Clin Oncol. 48:760–764. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
Murata Y, Tatsugami K, Yoshikawa M, Hamaguchi M, Yamada S, Hayakawa Y, Ueda K, Momosaki S and Sakamoto N: Predictive factor of biochemical recurrence after radical prostatectomy for high–risk prostate cancer. Int J Urol. 25:284–289. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
Nkengurutse G, Tian F, Jiang S, Wang Q, Wang Y and Sun W: Preoperative predictors of biochemical recurrence–free survival in high–risk prostate cancer following radical prostatectomy. Front Oncol. 10:17612020. View Article : Google Scholar : PubMed/NCBI | |
|
Lee KS, Koo KC, Choi IY, Lee JY, Hong JH, Kim CS, Lee HM, Hong SK, Byun SS, Rha KH and Chung BH: Re–stratification of patients with high–risk prostate cancer according to the NCCN guidelines among patients who underwent radical prostatectomy: An analysis based on the K–CaP registry. Cancer Res Treat. 50:88–94. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
García-Barreras S, Nunes I, Srougi V, Secin F, Baghdadi M, Sánchez-Salas R, Barret E, Rozet F, Galiano M and Cathelineau X: Predictors of early, intermediate and late biochemical recurrence after minimally invasive radical prostatectomy in a single–centre cohort with a mean follow–up of 8 years. Actas Urol Esp (Engl Ed). 42:516–523. 2018.(In English, Spanish). View Article : Google Scholar : PubMed/NCBI | |
|
Cao L, Yang Z, Qi L and Chen M: Robot–assisted and laparoscopic vs open radical prostatectomy in clinically localized prostate cancer: Perioperative, functional, and oncological outcomes: A Systematic review and meta–analysis. Medicine (Baltimore). 98:e157702019. View Article : Google Scholar : PubMed/NCBI | |
|
Preisser F, van den Bergh RCN, Gandaglia G, Ost P, Surcel CI, Sooriakumaran P, Montorsi F, Graefen M, van der Poel H, de la Taille A, et al: Effect of extended pelvic lymph node dissection on oncologic outcomes in patients with D'Amico intermediate and high risk prostate cancer treated with radical prostatectomy: A multi–institutional study. J Urol. 203:338–343. 2020. View Article : Google Scholar : PubMed/NCBI | |
|
Lee CU, Lee SM, Chung JH, Kang M, Sung HH, Jeon HG, Jeong BC, Seo SI, Jeon SS, Lee HM and Song W: Clinical utility of prostate health index for diagnosis of prostate cancer in patients with PI-RADS 3 lesions. Cancers (Basel). 14:41742022. View Article : Google Scholar : PubMed/NCBI |
