Over the last ten years, multiparametric MRI (mpMRI) of the prostate has gained rising importance in the diagnosis of prostate cancer (PCa). With mpMRI, high accuracy for the detection of PCa has been reported [
Different systems for MRI/US image fusion are available, which allow for transrectal (TR) or transperineal (TP) biopsy or for both [
With regard to complications, with TR prostate biopsy, the incidence rate of infection and rectal bleeding is higher [
Furthermore, there are still no clear recommendations if more than one targeted biopsy core per PCa suspect MRI lesion significantly increases the cancer detection rate. A recent study by Schimmoller et al. found only minor benefit when taking a second targeted biopsy core in a cohort of 290 patients who underwent in-bore MRI-guided prostate biopsy [
The purpose of our study was to evaluate MRI/US fusion-guided biopsy using the TP or the TR approach with the same fusion system in terms of diagnostic accuracy, PCa detection rates, and feasibility in a clinical setting.
Between July 2012 and October 2016, 154 patients with clinical suspicion of PCa underwent MRI/US fusion-guided biopsy of the prostate. Between July 2012 and January 2015, only the TR approach was performed and, until then, 62 patients underwent fusion-guided biopsy. Therefore, 40% of patients from our cohort underwent TR biopsy, irrespective of prostate size, tumor localization according to MRI, negative initial biopsies, or patients’ preferences due to unavailability of the TP approach. Since February 2015, both biopsy approaches were performed at our institution. Since then, detailed information and counseling on risks and advantages of the TR and TP approaches were given to the patients under consideration of prostate size, tumor localization according to MRI, negative initial biopsies, and patients’ general condition. Under consideration of this information and patients’ preferences, 17 patients underwent TR biopsy and 75 patients underwent TP biopsy after February 2015. This retrospective study focuses particularly on a descriptive analysis of clinical parameters in association with the different biopsy settings. In total, 79/154 patients underwent TR biopsy and 75/154 patients underwent TP biopsy. Written informed consent was obtained from each patient for clinically indicated mpMRI and MRI/US fusion-guided biopsy. The local ethical board waived the requirement for obtaining informed consent for this retrospective analysis. 125 patients had at least one negative previous biopsy (1–7 prebiopsies), 16 patients had no previous biopsy, 11 patients were on active surveillance, and in two patients it was unknown whether a previous biopsy had been performed. Patients’ characteristics are summarized in Table
Patient characteristics.
Number of patients | 154 | |
Age (years) | Mean ± SD | 66 |
PSA ( |
Mean (range) | 13 (0.4–101) |
Number of prebiopsies | Mean (range) | 1 (0–7) |
Prostate volume (ml) | Mean ± SD | 63 |
PI-RADS score | Mean ± SD | 4 |
Biopsy cores | Total | 1529 |
Mean ± SD per patient | 10 |
PI-RADS: Prostate Imaging Reporting and Data System; PSA: prostate specific antigen; SD: standard deviation.
Multiparametric MRI was acquired according to European Society of Urogenital Radiology (ESUR) guidelines [
Target lesions for MRI/US fusion-guided biopsy were chosen in a clinical setting by experienced uroradiologists. Before the publication of PI-RADS version 2 by the American College of Radiology (ACR), ESUR, and AdMeTech Foundation in December 2014, lesions were scored according to PI-RADS version 1 [
MRI/US fusion-guided biopsy was performed with the BioJet™ fusion system and software (D&K Technologies, Barum, Germany). The technical data and usage of this system have been described previously [
Patients were informed about advantages and risks of TR and TP prostate biopsy. Indication for TR or TP biopsy was adjusted to the clinical setting under consideration of technical availability, localization of lesion, patients’ profile, and patients’ preferences.
TR and TP prostate biopsies were performed by two experienced urologists in the dorsal lithotomy position under antibiotic prophylaxis. Local anesthesia (periprostatic block) was performed in patients with a TR biopsy course and general anesthesia was performed for the TP approach. TR MRI/US fusion was performed using a 3D triplane TR ultrasound system (BK Medical, Analogic Ultrasound Group, Pro Focus, Transducer 8818, 9 MHz) solely operating with the side-fire function. An endocavity biplane transducer (BK 8848, 9 MHz, BK Medical, Analogic Ultrasound Group) was used for the TP approach. Biopsy cores with a core length of 22 mm were numbered according to the radiological anatomic sector map as described in the PI-RADS version 1 [
Using the TP setting, the mean number of biopsy cores per patient was
For each cancer-positive biopsy core, a pathologist determined the Gleason grade and Gleason score (GS). Clinically significant cancer was defined as GS ≥ 7 and/or PSA ≥ 10
For statistical analysis, GraphPad Prism software version 6 (GraphPad Software, Inc., USA) and SPSS software version 24 (IBM Corporation, USA) were used. Clinical data of patients with and without biopsy-proven PCa as well as patients with TR and TP biopsy approaches were compared using unpaired
In our TR cohort, 31/79 patients were diagnosed with PCa (39%) between July 2012 and September 2016. 60/79 patients had untargeted negative prebiopsies, 6 patients were under active surveillance, 12 patients were biopsy-naïve, and in one patient the prebiopsy status was unknown. One patient was excluded from the analysis because the target lesion located in the anterior part of the prostate could not be reached by TR biopsy due to restriction of the needle guide (at that time point, a TP biopsy approach was technically not available at our institution yet). 664 biopsy cores were taken in total: 271 cores were from targeted biopsy and 393 cores were from additional systematic random biopsies.
PI-RADS scores were significantly higher in patients with PCa compared to patients without PCa (
According to the sector map, 27/79 (33%) target lesions from TR biopsy were located in the anterior half of the prostate; 52/79 (66%) target lesions were in the posterior half. Concordantly, the tumor detection rate was slightly but not significantly lower in the anterior (12/31, 39%) than in the posterior prostate (19/31, 61%; Table
Distribution of MRI/TRUS fusion-guided biopsy target lesions and PCa in TR and TP cohorts.
localization | Transrectal | Transperineal | ||
---|---|---|---|---|
All lesions ( |
Lesions with |
All lesions ( |
Lesions with |
|
AFS | 9/79 |
6/31 |
27/75 |
24/56 |
Anterior | 18/79 |
6/31 |
32/75 |
20/56 |
Posterior | 52/79 |
19/31 |
16/75 |
12/56 |
In our TR cohort, sensitivity and specificity of 81% and 69% were achieved for mpMRI with a Youden-selected cut-off value of PI-RADS ≥ 4 (area under the curve (AUC) = 0.81 (95% confidence interval (CI): 0.7–0.9)). NPV was 85% and PPV was 63%.
26 out of 31 PCa positive biopsies were obtained by targeted biopsy (Table
Frequency of detection of low-risk, intermediate, and high-risk PCa with targeted MRI/TRUS fusion-guided and systematic biopsy.
Risk group according to D’Amico criteria | Transrectal | Transperineal | ||
---|---|---|---|---|
Targeted | Systematic | Targeted | Systematic | |
Low risk | 6 | 1 | 16 | 2 |
Intermediate and high risk | 15 | 9 | 36 | 2 |
All PCa | 21 | 10 | 52 | 4 |
The number of lesions with highest GS detected with targeted and systematic biopsy is given. The overall PCa detection rate is not shown. Highest GS were more frequently detected with targeted than with systematic biopsy with both biopsy routes. PSA: prostate specific antigen.
A subgroup analysis of patients with PCa positive targeted biopsy (
In the TP cohort, 56/75 patients were diagnosed with PCa (75%) between February 2015 and September 2016. 65/75 patients had untargeted negative prebiopsies, five patients were under active surveillance, four patients were biopsy-naïve, and in one patient the prebiopsy status was unknown. 865 biopsy cores were taken: 242 cores were from targeted biopsy and 623 cores from additional systematic biopsies.
PI-RADS scores were significantly higher in patients with PCa compared to patients without PCa (
In TP cohort, 59/75 (79%) dominant lesions were located in the anterior prostate and 16/75 (21%) were located in the posterior part of the organ. Concordantly, the PCa detection rate was significantly higher in the anterior gland (44/56, 79%) when compared to the posterior gland (12/56, 21%;
For this cohort, sensitivity and specificity of 86% and 84% were achieved for mpMRI with a Youden-selected cut-off value of PI-RADS 4, with an AUC of 0.89 (95% CI: 0.8–1.0). NPV was 67% and PPV was 94%.
In 54/56 patients with PCa, tumor diagnosis was obtained from targeted biopsy (Table
A subgroup analysis of patients with PCa positive targeted biopsy (
Our TP cohort was significantly older (
Comparison of clinical parameters of transrectal (TR) and transperineal (TP) cohorts.
Clinical and MRI parameters | TR |
TP |
|
---|---|---|---|
Age (years) | 65 ± 8 | 67 ± 8 | <0.05 |
PSA ( |
10 ± 11 | 16 ± 18 | <0.05 |
Number of prebiopsies | 1.3 ± 0.9 (0–4) | 1.6 ± 1.0 (0–7) | <0.05 |
Prostate volume (ml) | 58 ± 31 | 68 ± 45 | ns |
BMI (kg/m2) | 27 ± 4 | 26 ± 4 | ns |
|
3 ± 1 | 4 ± 1 | <0.001 |
ADC (10−3 mm2/s) | 0.9 ± 0.3 | 0.7 ± 0.2 | <0.001 |
Lesion size (mm) | 13 ± 5 | 17 ± 8 | <0.001 |
We could demonstrate that mpMRI in combination with MRI/US fusion-guided biopsy yields high detection rates of PCa for both biopsy approaches. The BioJet system provides the opportunity to use both biopsy routes easily and can therefore offer different strategies for the individual patient with respect to the clinical setting. Additional untargeted, random biopsy cores as well as taking a second targeted biopsy core from the target lesion improved tumor detection rates in our study for both biopsy routes.
It is known that, in biopsy-naïve patients, detection rates for PCa with randomized untargeted biopsy are 20–30% [
In our study, we could achieve cancer detection rates of 39% for TR and 75% for TP combined targeted and systematic biopsies in a patient cohort with primarily prebiopsied patients (90% with at least one previous biopsy), while less biopsy cores were needed than with saturation biopsy:
When using radical prostatectomy tissue as reference, it has been reported that significant PCa might be missed in 8–24% of patients by mpMRI [
Furthermore, evidence suggested that, in MRI/US fusion-guided biopsy, a two-core biopsy should be performed per target lesion [
With the descriptive design of our study (not randomized, different starting points of TR and TP approach, selection bias), comparing the two cohorts is not intended. We could show that, with one software system, both biopsy routes are feasible and equally effective. Our two cohorts differ in some aspects. In 2012, we started using mpMRI and MRI/US fusion-guided biopsy and procedures (MRI protocol, workflow, and reporting) were not standardized due to a lack of recommendations about when to use MRI fusion-guided biopsy in men with clinical suspicion for PCa. Between July 2012 and January 2015, only TR fusion-guided biopsy (
In our study, one anterior lesion could not be reached with TR approach. Therefore, TP biopsy could be helpful to reach the anterior part of the prostate. In our patient cohort, anterior lesions were more often biopsied with the TP approach. Notably, when comparing the yield of PCa detection within the two cohorts, the percentage of detected PCa located in the anterior half of the prostate was related to the percentage of MRI target lesions in the anterior prostate (Table
Previous studies suggested that TR biopsy might hold a higher risk of infection, since faecal bacteria can enter blood circulation after retrieving specimen from the prostate [
Limitations of our study are the retrospective design and solely descriptive analysis of TR or TP biopsy approach in a clinical setting. Patients were not randomized into TR and TP cohort. The decision for one of the two biopsy routes depended on a doctors-patient shared decision, under consideration of patients’ preference, prostate size, lesion localization, and technical availability. And, finally, the TP approach has been available since February 2015, which introduced a bias.
In conclusion, we demonstrate that MRI/US fusion-guided biopsy has high accuracy for the detection of PCa with both TR and TP approaches. Our biopsy system provides the opportunity to offer both biopsy routes to patients at risk of PCa and to adjust diagnostic strategy to the individual clinical setting. In addition, the TP approach gives the options for fusion-guided focal therapy strategies.
American College of Radiology
Apparent diffusion coefficient
Anterior fibromuscular stroma
Area under the curve
Dynamic contrast-enhanced imaging
Diffusion-weighted imaging
European Society of Urogenital Radiology
Gleason score
Multiparametric magnetic resonance imaging
Negative predictive value
Prostate cancer
Prostate Imaging Reporting and Data System
Positive predictive value
Prostate specific antigen
Standard deviation
Transperineal
Transrectal
Turbo spin echo
Ultrasound.
The authors declare that there are no conflicts of interest regarding the publication of this article.
Susanne Tewes and Inga Peters contributed equally to this work.
Katja Hueper and Susanne Tewes received funding from the Junge Akademie Program, Hannover Medical School.