Prostate-specific Antigen Parameters and Prostate Health Index Enhance Prostate Cancer Prediction With the In-bore 3-T Magnetic Resonance Imaging-guided Transrectal Targeted Prostate Biopsy After Negative 12-Core Biopsy

Accepted Manuscript Title: Prostate-Specific Antigen Parameters and Prostate Health Index Enhance Prostate Cancer Prediction with the in-Bore 3-T MRI-Guided Transrectal Targeted Prostate Biopsy after Negative 12-Core Biopsy Author: Alexander Friedl, Kathrin Stangl, Wilhelm Bauer, Danijel Kivaranovic, Jenifer Schneeweiss, Martin Susani, Stephan Hruby, Lukas Lusuardi, Fritz Lomoschitz, Edith Eisenhuber-Stadler, Wolfgang Schima, Clemens Brössner PII: DOI: Reference:

S0090-4295(17)30880-4 http://dx.doi.org/doi: 10.1016/j.urology.2017.08.019 URL 20623

To appear in:

Urology

Received date: Accepted date:

18-4-2017 2-8-2017

Please cite this article as: Alexander Friedl, Kathrin Stangl, Wilhelm Bauer, Danijel Kivaranovic, Jenifer Schneeweiss, Martin Susani, Stephan Hruby, Lukas Lusuardi, Fritz Lomoschitz, Edith Eisenhuber-Stadler, Wolfgang Schima, Clemens Brössner, Prostate-Specific Antigen Parameters and Prostate Health Index Enhance Prostate Cancer Prediction with the in-Bore 3-T MRI-Guided Transrectal Targeted Prostate Biopsy after Negative 12-Core Biopsy, Urology (2017), http://dx.doi.org/doi: 10.1016/j.urology.2017.08.019. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Prostate-specific Antigen Parameters and Prostate Health Index Enhance Prostate Cancer Prediction with the In-bore 3-T MRI-guided Transrectal Targeted Prostate Biopsy after Negative 12-Core Biopsy

Alexander Friedl1 Kathrin Stangl1 Wilhelm Bauer1 Danijel Kivaranovic1 Jenifer Schneeweiss1 Martin Susani2 Stephan Hruby3 Lukas Lusuardi3 Fritz Lomoschitz4 Edith Eisenhuber-Stadler4 Wolfgang Schima4 Clemens Brössner1

1 Department of Urology, Barmherzige Schwestern Krankenhaus, Vienna, Austria 2 Department of Pathology, Medical University of Vienna, Vienna General Hospital, Vienna, Austria 1 Page 1 of 25

3 Department of Urology, Landeskrankenhaus Salzburg, Paracelsus Medical University, Salzburg, Austria 4 Department of Diagnostic and Interventional Radiology, Goettlicher Heiland Krankenhaus, Barmherzige Schwestern Krankenhaus, and Sankt Josef Krankenhaus, Vienna, Austria

Correspondence to: Alexander Friedl, MD Department of Urology, Barmherzige Schwestern Krankenhaus, Stumpergasse 13 1060 Vienna, Austria Tel.: + 43 1 599 88-6456 Fax: +43 1 599 88-4075 Email: [email protected] (A. Friedl) Orcid: 0000-0003-4330-8200

Key words: Prostate cancer prediction, MRI-guided transrectal targeted prostate biopsy, PSA parameters, prostate health index

Abstract Objectives 2 Page 2 of 25

To assess prostate cancer (PCa) detection and prediction by combining the in-bore MRI-guided transrectal targeted prostate biopsy (MRGB) with prostate-specific antigen (PSA) parameters and the prostate health index (PHI) in case of negative 12core standard biopsy (SB). Methods A total of 112 males (2014-2016) underwent 3.0T multi-parametric MRI (mpMRI) and subsequent MRGB of Prostate Imaging Reporting and Data System (PI-RADS) lesions 3-5. Ancillary PSA parameters (PSA ratio [%fPSA], PSA density [PSAD]) and the PHI and PHI density (PHID) were recorded. With these parameters in combination with MRGB PCa prediction was calculated. Results The most common lesions biopsied were PI-RADS 4 (66%), in the peripheral zone (64%), in the mid (58%) and anterior (65%) section of the prostate and 13 (IQR 1015) mm large. PCa was found in 62 (55%) patients (28% Gleason Score ≥7). PSAD (0.15 vs. 0.21; p=0.0051), %fPSA (16 vs. 13; p=0.0191), PHI (45 vs. 69; p<0.0001), PHID (0.7 vs. 1.5; p<0.0001) and prostate volume (ml; 56 vs. 45; p=0.0073) were significantly different in patients without/with PCa. PHI and PHID were the strongest predictors of PCa with an AUC of 0.79 and 0.77, respectively. Using optimal thresholds of 59 and 0.79, PHI and PHID were 69%/84% sensitive and 82%/62% specific for PCa. Conclusions Following negative SB of the prostate, the MRGB achieved an overall PCa detection rate of 55% in patients with PI-RADS 3 to 5 lesions. By considering PHI and PHID 3 Page 3 of 25

82%/62% of unnecessary biopsies could have been avoided, failing to detect 31%/16% cancers. Introduction Following an initial negative transrectal standard biopsy (SB) of the prostate but with persistent clinical suspicion of prostate cancer (PCa), the multi-parametric magnetic resonance imaging (mpMRI) is a recommended diagnostic tool for detection of prostate lesions if high-quality MRI units are available. The development of mpMRI findings (T2-weighted [T2W] pulse sequences, diffusion-weighted imaging [DWI], apparent-diffusion coefficient [ADC] maps and dynamic contrast-enhanced [DCE] sequences] has improved diagnostic capabilities for distinguishing clinical significant PCa from insignificant PCa and benign diseases.1 A subsequent in-bore MRI-guided transrectal targeted prostate biopsy (MRGB) can diagnose PCa if imaging quality of the MRI is high and prior diagnostic strategies are standardized by using the Prostate Imaging - Reporting and Data System (PI-RADS) classification.1,2 Compared to the ultrasound-guided SB it has similar overall PCa detection rates with a tendency to find more clinically significant PCa and requires significantly fewer cores. 3,4 Both techniques have difficulties in detecting apical lesions and can miss PCa in the dorsolateral and anterior sections.5,6 To avoid unnecessary biopsies and reduce patient burden predictive PCa-markers can be useful for reducing the need for MRGB or extended SB after mpMRI.7

The use of ancillary prostate-specific antigen (PSA)-parameters (PSA-density [PSAD], PSA-ratio [%fPSA]) including the prostate health index (PHI) and PHIdensity (PHID) can help to predict PCa in patients undergoing initial SB, Re-SB and MRI-guided transperineal biopsy.8–11 This study aimed to assess PCa prediction by 4 Page 4 of 25

combining MRGB with PSA-parameters and the prostate health index (PHI).

5 Page 5 of 25

Material and methods Population In the present single-institution study we retrospectively analyzed data of 112 caucasian males who underwent MRGB due to suspicious prostate MRI despite having had a previous negative 12-core transrectal ultrasound (TRUS) guided SB of the peripheral prostate zone. Ethics committee approval (protocol no. 201703-EK04, Barmherzige Schwestern Krankenhaus) and patient consents were obtained. Patients underwent MRGB between June 2014 and December 2016 after initial mpMRI and exclusively PI-RADS 3-5 lesions detected. Exclusion criteria were positive digital rectal findings, general MRI-contraindications (metal-containing implants, claustrophobia), existing PCa, distinct MRI artefacts and less than 3 months after SB. Pre-MRGB patient data is summarized in Table 1a.

MRI MRGB and mpMRI were performed on 3.0-T MRI units (Siemens Skyra, Erlangen, Germany) in the same department. No endorectal coil was used. For mpMRI a 18channel receive coil was used. The examination included axial T1-weighted VIBE DIXON images (TR 5.7 msec, TE 2.5 msec, slice thickness 3 mm), axial, coronal and sagittal T2-weighted turbo spin-echo (TSE) pulse sequences (TR 7500-7640 msec, TE 105-108 msec, slice thickness 3.5 mm), axial diffusion-weighted images (DWI; TR 3970 msec, TE 74 msec, b-values 0, 500, 1500), and an axial contrast-enhanced dynamic T1-weighted VIBE pulse sequence (TE 4.1 msec, TE 1.4 msec, slice thickness 3.5 mm) for perfusion imaging (DyanCAD, Invivo, Gainesville, USA). Ultrasound gel was applied rectally prior examination to reduce susceptibility artifacts. To reduce imaging artifacts from peristalsis, 20 mg of hyoscin-N-butyl bromide were IV administered. 6 Page 6 of 25

Studies were read by 2 radiologists (W.S., E.E-S) with more than 20 years and 5 years, respectively, of urologic MRI experience according to the PI-RADS v1/v2 classification based on the criteria of the European Society of Urogenital Radiology (ESUR).12,13 The Index lesion (PI-RADS 3 to 5) was defined as the largest suspicious focus (maximum tumor diameter in ADC map in axial plane) with the highest PI-RADS score, which was then subjected to biopsy on a separate day.

MRGB All patients received oral fluoroquinolones for antibiotic prophylaxis, starting in the evening before biopsy. MRGB was performed in the MR scanner in prone position with local anesthetic gel applied and under IV analgesia (5 mg nalbuphin hydrochloride). To reduce imaging artifacts from peristalsis, 20 mg of hyoscin-N-butyl bromide were IV administered. A stereotactic biopsy device (DynaTRIM, Invivo) was used and the endorectal needle guide was inserted into the rectum. Multiplanar localization sequences (T2-w TSE, DWI) were acquired to identify the region of interest. Automated software assists in placement of the needle guide by providing adjustment parameters. An 18 G biopsy MR-compatible (Invivo) core needle was used to extract specimens and median 5 cores are taken from the lesion. MRGB data is shown in Table 1b. Each single core was embedded in paraffin and histological sections were examined under the microscope by one experienced pathologist (M.S.; >3000 cases). Gleasonscore and grade groups were determined according to the 2014 adapted guidelines of the International Society of Urological Pathology (ISUP).14 A Gleason-score of 7 or above was considered as clinically significant PCa.

7 Page 7 of 25

PSA-parameters and PHI Serum PSA-parameters were measured as per standard clinical pathway in the same laboratory. %fPSA was calculated as free PSA/total PSA and PHI was assessed by [([-2]proPSA/free PSA) x (PSA)½] formula. PHID/PSAD calculations were performed using prostate volume via mpMRI (transversal x sagittal x craniocaudal x π/6). Beside optimal cut-off values determined in the analysis, thresholds at 90% sensitivity were estimated for representing PCa prediction of PSAD, %fPSA, PHI and PHID.

Statistical analysis Continuous variables were expressed as median with interquartile range (IQR) and compared using the Wilcoxon rank-sum test. Categorical variables were expressed as absolute and relative frequencies and were compared by Fisher tests. Receiver operating characteristic (ROC) curves were used to evaluate the accuracy of several clinical parameters in predicting PCa. Optimal cut-off values were determined by the Youden index. Delong’s method was used to compare ROC curves between different measurements. Multivariate logistic regression was used to combine clinical parameters for PCa prediction. P-values<0.05 were considered statistically significant. Statistical analysis was performed using R 3.3.1.

8 Page 8 of 25

Results The most common lesions biopsied were PI-RADS 4 (66%), in the peripheral zone (64%), in the mid (58%) and anterior (65%) section of the prostate and 13 mm (IQR 10-15) large (Table 1a). In a median 4 (IQR 2-5) of 5 (IQR 4-5) biopsy cores were positive with largest PCa extension of 5 (IQR 3-10) mm. Post-MRGB infection (n=3/3%) was the only complication recorded in our cohort (Table 1b).

After 112 MRGB an overall PCa was observed in 62 (55%) patients and a significant PCa in 31 (28%) patients. Age was higher in the PCa group (65 vs. 70 yr; p=0.0004) and prostate volume larger in patients without PCa (56 vs. 45; p=0.0073). PSA was not statistically different but ancillary parameters such as %fPSA (16% vs. 13%; p=0.0191), PSAD 0.15 vs. 0.21; p=0.0051), PHI (45 vs. 69; p<0.0001) and PHID (0.7 vs. 1.5; p<0.0001) showed significant distinctions between groups which are illustrated in Table 2a.

Beside age, PHI and PHID were the strongest predictors of PCa detection via MRGB. Summarizing the ROC analysis in Table 2b, the AUC of PHI (0.79) and PHID (0.77) were higher than those of prostate volume alone (0.65), PSAD (0.65) and %fPSA (0.62). Optimal cut-off values of 59 and 0.79 for PHI and PHID showed sensitivity of 69%/84% and specificity of 82%/62%. Thus, PHI and PHID could have been used to avoid 82%/62% of unnecessary biopsies while failing to detect 31%/16% of all PCa. To achieve a sensitivity of 90%/90% in PHI/PHID we estimated thresholds of 42 and 0.67 and specificities of 44%/46%. The performance/AUC of PHI (0.81 for PI-RADS 4 vs. 0.82 for PI-RADS 5) and PHID (0.77 for PI-RADS 4 vs. 0.79 for PI-RADS 5) varied only slightly between PI-RADS 9 Page 9 of 25

cohorts. In subgroup analysis %fPSA (AUC 0.56), PSAD (AUC 0.68), PHI (0.65) and PHID (AUC 0.63) had similar predictive capabilities for significant PCa (≥Gleason score 7) detection to that for all PCa patients. Multivariate logistic regression models were used to assess if combining two or more markers lead to significantly better PCa prediction than the best single marker, which was PHI (AUC: 0.79). The best bivariate model, which included PHI and PSAD as covariates, achieved an AUC of 0.79 which was no significant improvement to PHI (p= 0.8717). Also, logistic regression models with more than two covariates did not significantly perform better than PHI. ROC curves of several bivariate logistic regression models were shown in Figure 1.

10 Page 10 of 25

Comment To our knowledge this is the first study demonstrating the predictive value of ancillary PSA-parameters and PHI in patients undergoing MRGB after negative SB. We analyzed a carefully selected patient population with detected prostate index lesions at mpMRI (PI-RADS 3-5). The disproportionate number of anterior lesions (65%) may be due to that fact that the anterior section is more difficult to reach via transrectal SB, leading to higher number of prior false negative results. Therefore, accurate detection of the index lesion via mpMRI is mandatory, before MRGB is done.

In our trial PHI (AUC 0.79) and PHID (0.77) were the strongest predictors in PCa detection combined with MRGB. Total PSA was similar in patients with/without PCa (9.8 ng/ml and 8.9 ng/ml, respectively), which confirms the indication for this intervention. In compliance with the literature the MRGB in this study had a good safety profile and showed a PCa detection of 55% in previously biopsied patients. 15 Tan et al. achieved a PCa detection of 59% (of these 68% were a significant PCa) with the transrectal in-bore 3T MRGB and Penzkofer et al. reported a rate of 57% of PCa with the transperineal in-bore 3T MRGB.2,16 Compared to cognitively/visually targeted SB after mpMRI (27% PCa detection), the MRI-ultrasound fusion biopsy showed superior results (32–59% PCa detection) depending on the MRI/ultrasound units, software application and number of cores taken.17–19 In contrast, a study group from Finland did not find any benefit in their randomized single-center trial using MRIultrasound fusion biopsy in PCa diagnosis compared to ultrasound-guided SB (64% vs. 57%, p=0.5).20

11 Page 11 of 25

In the last few months an increasing amount of studies dealing with mpMRI and MRGB has been published, demonstrating the diagnostic reliability of using mpMRI pre/post-SB. Recently Ahmed et al. (PROMIS trial) reported that using mpMRI to triage men might allow 27% of patients to avoid a primary SB and diagnosis of 5% fewer clinically insignificant cancers.21 This data might change the diagnostic pathway of PCa in the future and radiologist will get more involved in PCa detection, providing that 3.0T-MRI units are adequately available. A consensus statement of the AUA (American Urological Association) and the SAR (Society of Abdominal Radiology) recently stated that repeat biopsy with image-guided targeting is indicated in PI-RADS 3-5 lesions after negative SB.7 They suggest the use of ancillary markers (PSA, PSAD, PHI) in patients with negative or low suspicion lesions (PI-RADS <3) may be of value in identifying patients warranting repeat biopsy. 7 We found a benefit using these parameters in patients with PI-RADS 3 to 5 lesions but we have no data on patients with PI-RADS 1 and 2 lesions. Regarding the index lesion, which is considered the biological driver of the PCa 22, we found many index lesions overlapping different prostate sections. Compared to normal random biopsy and saturation biopsy which have a higher PCa detection rate if the number of cores taken out is higher 23, we do not currently know if this is also true for the MRGB.

In our experience the use of PHI and PSA derivates in terms of PCa prediction is common but still not routinely employed due to its ambiguous cost-benefit factor. Worthwhile cost-effectiveness studies are still missing on this aspect. Recently a large, real-time clinical study reported of a lower rate of unnecessary biopsies (39% vs 48%; P<0.001) if the PHI was incorporated into the daily practice, whereas the frequency of high-grade cancer detection did not change.24 Uncontested, the PHI provides complementary information if MRI is suspect, but how strongly can one trust 12 Page 12 of 25

these parameters before biopsy in the outpatient setting? The main dilemma of predictive parameters is still the low reliability and the unclear cut off values in different patient/risk groups. Even though several parameters have already been evaluated for PCa prediction and the optimal parameter which might change the management of the PCa has not yet been found.25–27 Thus, the important question whether the PHI (or other biomarkers) might be used as a triaging test to predict PCa in 3T MRGB has not yet been investigated. Regarding this, we found lower AUC values in clinically significant PCa (Gleason score ≥7) compared to the overall PCa group which therefore slightly limits the clinical performance of these biomarkers by only predicting the meaningful PCa of PI-RADS 3 to 5 lesions. Hence the impact of these mentioned predictors is of greater interest in PI-RADS 1 to 2 lesions and may indicate the need for a re-biopsy of the prostate. Depending on the study setting and biopsy approach an AUC of PHI between 0.75-0.78 has been reported in large prospective, multi-site early detection trials.25,28 The PHID was recently evaluated in two prospective studies and showed promising accuracy in the prediction of PCa in primary and secondary biopsy (AUC 0.70 to 0.77).8,29 Therefore, we assumed this parameter in the analysis.

Our study has several limitations. The cohort was small (n=112), selective (PI-RADS 3-5 scores, no PI-RADS 1-2) and heterogeneous in terms of age, number of previous biopsies and possible PCa risk factors (such as family history). The data collection was retrospective and for mpMRI evaluation PI-RADSv1 (diagnostic preference of PZ) and v2 (diagnostic preference of TZ) were used which might have led to a detection bias of the index lesion. Furthermore, deficient inter-reader reproducibility could have occurred between radiologists, as published recently by Polanec et al. 30 The number of biopsy cores obtained varied, depending on the specimen quality and 13 Page 13 of 25

only the index lesion was biopsied. Biopsy outcome was not compared to final prostatectomy pathology. Therefore we cannot exclude a multifocal PCa outside the biopsied region. Finally, these results describe a single-institution experience without any patient randomization. To avoid study bias a randomization with multi-center design will be the next step to validate these results.

14 Page 14 of 25

Conclusions Based on the present study the MRGB is an efficient diagnostic procedure (55% PCa detection rate) after negative SB and has low complication rates (3% infections). Ancillary PSA-parameters can optimize the prediction outcome and may reduce the need for repetitive biopsies. Thus, by considering PHI and PHID 82%/62% of unnecessary biopsies could have been avoided, failing to detect 31%/16% cancers.

15 Page 15 of 25

References: 1.

Weinreb JC, Barentsz JO, Choyke PL, et al. PI-RADS Prostate Imaging – Reporting and Data System: 2015, Version 2. Eur Urol. 2016;69(1):16-40. doi:10.1016/j.eururo.2015.08.052.

2.

Tan N, Lin W-C, Khoshnoodi P, et al. In-Bore 3-T MR-guided Transrectal Targeted Prostate Biopsy: Prostate Imaging Reporting and Data System Version 2–based Diagnostic Performance for Detection of Prostate Cancer. Radiology. 2016:152827. doi:10.1148/radiol.2016152827.

3.

Wegelin O, van Melick HHE, Hooft L, et al. Comparing Three Different Techniques for Magnetic Resonance Imaging-targeted Prostate Biopsies: A Systematic Review of In-bore versus Magnetic Resonance Imaging-transrectal Ultrasound fusion versus Cognitive Registration. Is There a Preferred Technique? Eur Urol. 2016. doi:10.1016/j.eururo.2016.07.041.

4.

Quentin M, Blondin D, Arsov C, et al. Prospective Evaluation of Magnetic Resonance Imaging Guided In-bore Prostate Biopsy versus Systematic Transrectal Ultrasound Guided Prostate Biopsy in Biopsy Naïve Men with Elevated Prostate Specific Antigen. J Urol. 2014;192(5):1374-1379. doi:10.1016/j.juro.2014.05.090.

5.

Schouten MG, van der Leest M, Pokorny M, et al. Why and Where do We Miss Significant Prostate Cancer with Multi-parametric Magnetic Resonance Imaging followed by Magnetic Resonance-guided and Transrectal Ultrasound-guided Biopsy in Biopsy-naïve Men? Eur Urol. 2017. doi:10.1016/j.eururo.2016.12.006.

6.

Cerantola Y, Haberer E, Torres J, et al. Accuracy of cognitive MRI-targeted biopsy in hitting prostate cancer-positive regions of interest. World J Urol. 16 Page 16 of 25

2016;34(1):75-82. doi:10.1007/s00345-015-1588-2. 7.

Rosenkrantz AB, Verma S, Choyke P, et al. Prostate Magnetic Resonance Imaging and Magnetic Resonance Imaging Targeted Biopsy in Patients with a Prior Negative Biopsy: A Consensus Statement by AUA and SAR. J Urol. 2016;196(6):1613-1618. doi:10.1016/j.juro.2016.06.079.

8.

Tosoian JJ, Druskin SC, Andreas D, et al. Prostate health index density improves detection of clinically-significant prostate cancer. BJU Int. 2017. doi:10.1111/bju.13762.

9.

Gnanapragasam VJ, Burling K, George A, et al. The Prostate Health Index adds predictive value to multi-parametric MRI in detecting significant prostate cancers in a repeat biopsy population. Sci Rep. 2016;6:35364. doi:10.1038/srep35364.

10.

Washino S, Okochi T, Saito K, et al. Combination of prostate imaging reporting and data system (PI-RADS) score and prostate-specific antigen (PSA) density predicts biopsy outcome in prostate biopsy naïve patients. BJU Int. 2017;119(2):225-233. doi:10.1111/bju.13465.

11.

Hansen NL, Kesch C, Barrett T, et al. Multicentre evaluation of targeted and systematic biopsies using magnetic resonance and ultrasound image-fusion guided transperineal prostate biopsy in patients with a previous negative biopsy. BJU Int. 2016. doi:10.1111/bju.13711.

12.

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(4):726-731. doi:10.1016/j.ejrad.2016.01.011.

13.

Barentsz JO, Richenberg J, Clements R, et al. ESUR prostate MR guidelines 2012. Eur Radiol. 2012;22(4):746-757. doi:10.1007/s00330-011-2377-y.

14.

Epstein JI, Egevad L, Amin MB, et al. The 2014 International Society of 17 Page 17 of 25

Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol. 2015;40(2):1. doi:10.1097/PAS.0000000000000530. 15.

Meier-Schroers M, Homsi R, Kukuk G, et al. In-bore transrectal MRI-guided prostate biopsies: Are there risk factors for complications? Eur J Radiol. 2016;85(12):2169-2173. doi:10.1016/j.ejrad.2016.09.029.

16.

Penzkofer T, Tuncali K, Fedorov A, et al. Transperineal In-Bore 3-T MR Imaging–guided Prostate Biopsy: A Prospective Clinical Observational Study. Radiology. 2015;274(1):170-180. doi:10.1148/radiol.14140221.

17.

Baco E, Rud E, Eri LM, et al. A Randomized Controlled Trial To Assess and Compare the Outcomes of Two-core Prostate Biopsy Guided by Fused Magnetic Resonance and Transrectal Ultrasound Images and Traditional 12core Systematic Biopsy. Eur Urol. 2016;69(1):149-156. doi:10.1016/j.eururo.2015.03.041.

18.

Wysock JS, Rosenkrantz AB, Huang WC, et al. A Prospective, Blinded Comparison of Magnetic Resonance (MR) Imaging–Ultrasound Fusion and Visual Estimation in the Performance of MR-targeted Prostate Biopsy: The PROFUS Trial. Eur Urol. 2014;66(2):343-351. doi:10.1016/j.eururo.2013.10.048.

19.

Oberlin DT, Casalino DD, Miller FH, et al. Diagnostic Value of Guided Biopsies: Fusion and Cognitive-registration Magnetic Resonance Imaging Versus Conventional Ultrasound Biopsy of the Prostate. Urology. 2016;92:75-9. doi:10.1016/j.urology.2016.02.041.

20.

Tonttila PP, Lantto J, Pääkkö E, et al. Prebiopsy Multiparametric Magnetic Resonance Imaging for Prostate Cancer Diagnosis in Biopsy-naive Men with Suspected Prostate Cancer Based on Elevated Prostate-specific Antigen 18 Page 18 of 25

Values: Results from a Randomized Prospective Blinded Controlled Trial. Eur Urol. 2016;69(3):419-425. doi:10.1016/j.eururo.2015.05.024. 21.

Ahmed HU, El-Shater Bosaily A, Brown LC, et al. Diagnostic accuracy of multiparametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet (London, England). 2017. doi:10.1016/S0140-6736(16)32401-1.

22.

Valerio M, Anele C, Freeman A, et al. Identifying the Index Lesion with Template Prostate Mapping Biopsies. J Urol. 2015;193(4):1185-1190. doi:10.1016/j.juro.2014.11.015.

23.

Radtke JP, Schwab C, Wolf MB, et al. Multiparametric Magnetic Resonance Imaging (MRI) and MRI–Transrectal Ultrasound Fusion Biopsy for Index Tumor Detection: Correlation with Radical Prostatectomy Specimen. Eur Urol. 2016;70(5):846-853. doi:10.1016/j.eururo.2015.12.052.

24.

Tosoian JJ, Druskin SC, Andreas D, et al. Use of the Prostate Health Index for detection of prostate cancer: results from a large academic practice. Prostate Cancer Prostatic Dis. 2017;20(2):228-233. doi:10.1038/pcan.2016.72.

25.

Foley RW, Maweni RM, Gorman L, et al. European Randomised Study of Screening for Prostate Cancer (ERSPC) risk calculators significantly outperform the Prostate Cancer Prevention Trial (PCPT) 2.0 in the prediction of prostate cancer: a multi-institutional study. BJU Int. 2016;118(5):706-713. doi:10.1111/bju.13437.

26.

Lin DW, Newcomb LF, Brown MD, et al. Evaluating the Four Kallikrein Panel of the 4Kscore for Prediction of High-grade Prostate Cancer in Men in the Canary Prostate Active Surveillance Study. Eur Urol. 2016. doi:10.1016/j.eururo.2016.11.017.

27.

Fenstermaker M, Mendhiratta N, Bjurlin MA, et al. Risk Stratification by Urinary 19 Page 19 of 25

Prostate Cancer Gene 3 Testing Before Magnetic Resonance ImagingUltrasound Fusion-targeted Prostate Biopsy Among Men With No History of Biopsy. Urology. 2017;99:174-179. doi:10.1016/j.urology.2016.08.022. 28.

Loeb S, Shin SS, Broyles DL, et al. Prostate Health Index improves multivariable risk prediction of aggressive prostate cancer. BJU Int. 2016. doi:10.1111/bju.13676.

29.

Mearini L, Ferri C, Lazzeri M, et al. Evaluation of Prostate-Specific Antigen Isoform p2PSA and Its Derivates, %p2PSA, Prostate Health Index and Prostate Dimension-Adjusted Related Index in the Detection of Prostate Cancer at First Biopsy: An Exploratory, Prospective Study. Urol Int. 2014;93(2):135-145. doi:10.1159/000356240.

30.

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(6):1125-1131. doi:10.1016/j.ejrad.2016.03.025.

20 Page 20 of 25

Figure 1: Area under the curve (AUC) comparison according to performance of single parameters (a) and combined parameters (b) in the detection of PCa with MRGB

PSA=Prostate specific antigen; fPSA=free PSA; %fPSA=PSA ratio; PSAD=PSA density; PHI=Prostate health index; PHID=PHI density;

21 Page 21 of 25

Table 1a. Descriptive characteristics of the study cohort prior MRGB including mpMRI data of the index lesion; Median (IQR) and n (%) Patient characteristics Age (yr)

67 (61-72)

ASA

1 (1-2)

ECOG

0 (0-0)

Prostate volume (ml)

51 (28-72)

IPSS

9 (6-11)

mpMRI – Index lesion Diameter lesion (mm)

13 (10-15)

PI-RADS 3

11 (10%)

PI-RADS 4

74 (66%)

PI-RADS 5

27 (24%)

Base – Mid - Apex

25 (22%) - 65 (58%) - 39 (35%)

Left - Right

49 (44%) - 69 (62%)

Anterior - Posterior

63 (65%) - 52 (46%)

Peripheral zone

72 (64%)

Transitional zone

39 (35%)

Zentral zone

7 (6%)

mpMRI – Pathological findings Seminal vesicels involvement

0 (0%)

Lymph nodes

9 (8%)

Extracapsular extension

5 (4%)

Neurovascular bundle involvement 2 (2%) MRGB=MRI-guided transrectal targeted prostate biopsy; mpMRI=multi-parametric magnetic resonance imaging; IQR=Interquartile range; ASA= American society of anesthesiologists score; ECOG=Eastern cooperative oncology group performance status; IPSS=International prostate symptom score; PI-RADS=Prostate imaging reporting and data system;

22 Page 22 of 25

Table 1b. Efficacy of the MRGB and histopathological findings; Median (IQR) and n (%) mpMRI to MRGB (weeks)

9.4 (6.6-14.2)

Biopsy cores Extracted

5 (5-5)

PCa affected

4 (2-5)

Highest extent with PCa (mm)

5.5 (3.2-9.9)

Post-MRGB complications Infection with fever

3 (3%)

Unexpected rectal bleeding

0 (0%)

Urinary retention

0 (0%)

Histopathology Benign

50 (45%)

-

38 (34%)

Prostatitis

PCa

62 (55%)

-

31 (28%)

GS 3+3

Clinically significant PCa (GS >7)

31 (28%)

-

GS 3+4

10 (9%)

-

GS 4+3

10 (9%)

-

GS 4+4

6 (5%)

-

GS 4+5

4 (4%)

-

GS 5+4

0 (0%)

-

GS 5+5

1 (1%)

MRGB=MRI-guided transrectal targeted prostate biopsy; IQR=Interquartile range; mpMRI=multi-parametric magnetic resonance imaging; PCa=Prostate cancer; GS=Gleason score;

23 Page 23 of 25

Table 2a. Distribution and comparison of ancillary PSA-parameters in patients without/with PCa; Median (IQR) Overall

No PCa

PCa

p-value

Age (yr)

67 (61-72)

65 (57-69)

70 (64-73)

0.0004

PVol (ml)

51 (38-72)

56 (45-75)

45 (36-65)

0.0073

PSA (ng/ml)

9 (6.5-13.3)

8.9 (6.6-12.4)

9.8 (6-14.1)

0.5006

fPSA (ng/ml)

1.1 (0.8-1.9)

1.4 (0.9-2.1)

0.9 (0.7-1.8)

0.0917

%fPSA

13 (10-18)

16 (12-21)

13 (10-16)

0.0191

PSAD

0.18 (0.120.27)

0.15 (0.1-0.21)

0.21 (0.130.31)

0.0051

PHI

54 (42-72)

45 (39-52)

69 (52-82)

<0.0001

PHID

1.2 (0.7-1.9)

0.7 (0.5-1.2)

1.5 (1.1-2.2)

<0.0001

PCa=Prostate cancer; PVol=Prostate volume; PSA=Prostate specific antigen; fPSA=free PSA; %fPSA=PSA ratio; PSAD=PSA density; PHI=Prostate health index; PHID=PHI density;

24 Page 24 of 25

Table 2b. Sensitivity, specificity and predictive values of parameters by using optimal cut-off values in combination with MRGB for PCa detection Threshold

Sensitivity

Specificity

PPV

NPV

AUC

Age (yr)

70.0

0.55

0.82

0.79

0.59

0.69

PVol (ml)

51.0

0.65

0.68

0.71

0.61

0.65

PSA (ng/ml) 9.7

0.52

0.67

0.67

0.53

0.54

fPSA (ng/ml) 0.95

0.51

0.72

0.68

0.56

0.60

%fPSA

15

0.63

0.54

0.60

0.57

0.62

PSAD

0.15

0.61

0.52

0.61

0.52

0.65

PHI

40

0.92

0.33

0.63

0.76

0.79

PHID

0.43

1.00

0.05

0.57

1.00

0.77

PVol=Prostate volume; PSA=Prostate specific antigen; fPSA=free PSA; %fPSA=PSA ratio; PSAD=PSA density; PHI=Prostate health index; PHID=PHI density; PPV=Positive predictive value; NPV=Negative predictive value; AUC=Area under the curve

25 Page 25 of 25