MRI-TRUS fusion biopsy of the prostate: Quality of image fusion in a clinical setting

Abstract

INTRODUCTION:

Prostate cancer (PCa) is one of the most common malignancies in men. The diagnostic standard to confirm prostate cancer is the transrectal ultrasound-guided biopsy. However, this procedure is associated with the underdetection of clinically significant prostate cancer and therefore needs to be improved. In the last years MRI fusion based targeted biopsy gained importance as consequence. In this study, we evaluated the quality of MRI ultrasound image fusion and evaluated factors influencing the image fusion quality. This was done by comparing fusion quality with the histopathological findings in the defined MRI target on the one hand and the PIRADS score on the other hand.

MATERIALS AND METHODS:

Single arm study including patients with elevated prostate specific antigen (PSA) and a multiparametric MRI showing a suspicious lesion underwent a MRI fusion targeted biopsy at our institution. MRI fusion targeted biopsy and an additional 12-core transrectal ultrasound (TRUS) guided biopsy was performed using the Philips Percunav device (Philips Medical Systems, Bothell, WA). The fusion accuracy was rated by two experienced clinicians (1 radiologist, 18 years of experience, 1 urologist, 5 years of experience) using a five-point rank scale (1 = best) and comparing the result with the histological findings in the target and the PIRADS score.

RESULTS:

The detection rate of clinically significant cancer (Gleason 7a or greater) by MRI-ultrasound fusion targeted biopsy was 58.6% (17/29) compared to 50% (19/38) in the standard transrectal ultrasound-guided approach. PCa was found in 36.4% (4/11 patients) of patients with a PIRADS 3 lesion, in 57.7% (15/26 patients) of patients with a PIRADS 4 lesion. In 76.9% (10/13 patients) of patients with a PIRADS 5 lesion PCa was diagnosed. No statistical significance was found comparing the quality of registration either with the PIRADS (p = 0.7873) nor with the Gleason score (p = 0.4376). The study is limited by the small number of patients.

CONCLUSIONS:

MRI fusion based targeted biopsy improves the identification of clinical significant cancer. The Gleason score of detected PCa is not influenced by the quality of fusion.

Keywords

Transrectal ultrasound-guided biopsy image fusion prostate prostate cancer multiparametric MRI

1 Introduction

Prostate cancer (PCa) is the second most common malignancy in men with an annual incidence of 1.1 million worldwide and a cancer specific mortality of 6.6% per year [1]. Therefore, prostate cancer is the fifth common cause of death among men [1].

According to national and international guidelines men with elevated prostate specific antigen (PSA) level or an abnormal digital rectal examination (DRE) are offered a standard transrectal ultrasound-guided biopsy of the prostate during which 10 to 12 biopsy cores are taken [2 –4]. Noteworthy and in contrast to most other organs many cancerous lesions are invisible to imaging modalities. The detection of PCa by conventional B-mode ultrasonography is insufficient [5 –8]. Several studies comparing the preoperative grey-scale ultrasonography with post-prostatectomy specimen showed a sensitivity between 18% and 48% to detect the tumor within the prostate [5 –7]. The negative predictive value (NPV) is between 37–45% and the positive predictive value (PPV) is 77–93% [5 –7]. Although, transrectal ultrasound-guided biopsy of the prostate is the standard of care for men with the clinical suspicion of prostate cancer, the detection rate is not very high [9 –11] with a false negative rate between 20–30% [12, 13]. Furthermore, this procedure is associated with an underdetection of clinical significant prostate cancer and the overdetection of clinical insignificant cancer [10, 11]. Therefore, different imaging modalities to diagnose PCa before or after therapy have been evaluated in the last years [14 –17]. Multiparametric magnetic resonance imaging (mpMRI) gained the greatest importance. MpMRI prior to the biopsy improved the accuracy and reproducibility to detect, localize and assess the extent and aggressiveness of cancer foci within the prostate gland. Cancer suspicious lesions within the mpMRI which are not visible in the ultrasound can be localized in the ultrasound with software– assisted image fusion of the two image modalities. However, results of single center studies analyzing the histopathological outcome of MRI-ultrasound fusion-based targeted biopsy of the prostate are mixed. MRI-ultrasound fusion-guided targeted biopsy of the prostate is especially associated with higher detection rates of clinically significant cancer compared to the standard 12-core biopsy [18 –21]. Some studies however, could not show superiority of the MRI-ultrasound fusion based targeted biopsy in the detection of prostate cancer compared to the systematic biopsy [18, 22]. In this study, we evaluated factors influencing the quality of MRI-ultrasound image fusion and compared it with the histological findings in the target region and the PIRADS score of the utilized mpMRI.

2 Materials and methods

2.1 Patient characteristics

In this retrospective single center study 50 patients with an elevated PSA level and an available multiparametric MRI (mpMRI) of the prostate underwent MRI fusion targeted biopsy between March and May 2018 at our institution. The PSA level ranged between 1.9–34.2 ng/ml (median 9.6 ng/ml), median patient age was 65 (48–81) years and median prostate volume was 56.3 ml (26–160 ml) (Table 1). The classification of tumor lesions in the mpMRI was done according to the Prostate Imaging Data and Reporting System Version 2 (PIRADS score) [23]. All patients showed PIRADS 3, 4 or 5 lesions in the mpMRI. In all patients included in the study there was only one suspicious lesion per prostate described in the mpMRI. The study was performed according to the ethical guidelines of Clinical Hemorheology and Microcirculation [24].

Table 1
Patients and baseline characteristics

Number of patients 50

Average age (years) 65 (48–81)

Average PSA level ng/ml 9.6 (1.9–34.2)

Average prostate volume ml 56.3 (26–160)

PIRADS 3 lesion 11

PIRADS 4 lesion 26

PIRADS 5 lesion 13

Number of patients	50
Average age (years)	65 (48–81)
Average PSA level ng/ml	9.6 (1.9–34.2)
Average prostate volume ml	56.3 (26–160)
PIRADS 3 lesion	11
PIRADS 4 lesion	26
PIRADS 5 lesion	13

2.2 MRI fusion targeted biopsy

MRI fusion-based targeted biopsy was performed software– assisted using the Philips Percunav device. In all patients the target biopsy was also combined with the standard transrectal ultrasound-guided 12-core biopsy according to current guidelines [4]. For image fusion of the mpMRI and the ultrasound image a magnetic field generator is required as hardware component which is placed over the pelvis of the patient. For data registration the DICOM data set of the mpMRI was uploaded to the ultrasound device. The data volume ranged between 225–5500 pictures per patient. The image fusion was performed using the axial T2-weighted MRI sequence. The fusion with the real time grey scale ultrasound image was performed by matching similar planes in the MRI and the ultrasound using anatomic landmarks (e.g. cysts) or the greatest diameter of the prostate in both image modalities. Registration of images could be adjusted manually, if necessary. Figures 1 and 2 show the image fusion of patients with a PIRADS 5 lesion in the prostate. The number of biopsy cores taken per target ranged between two or four cores depending on the size of the target and the accuracy of the MRI and ultrasound image fusion.

Fig.1

Image fusion (side by side view) showing a PIRADS 5 lesion in the MRI in the peripheral dorsolateral part of the prostate and the corresponding hypodense lesion in the B-mode ultrasound imaging.

Fig.2

Image fusion in side by side view showing a good correlation of the MRI/ultrasound fusion in a patient with a PIRADS 5 lesion.

2.3 Data analysis

Quality of fusion of MRI and ultrasound of the prostate was evaluated by two experienced clinicians (1 radiologist, 18 years of experience, 1 urologist 5 years of experience) on the basis of video recording of the image fusion and biopsy process at a later time. A qualitative five-point scale (1 = excellent, 2 = good, 3 = fair, 4 = poor, 5 = indefinable) was used to evaluate the registration on the subjective perception as consensus decision of both clinicians. The findings were compared with the histological results of the target lesion and the PIRADS score performing the chi– squared test.

3 Results

A total of 50 men underwent MRI-ultrasound image fusion-based targeted biopsy plus the standard transrectal ultrasound-guided systematic biopsy of the prostate. Baseline characteristics for all patients included are presented in Table 1. 52% of the patients included had a PIRADS 4 lesion (26/50 patients) in the mpMRI, 22% (11/50 patients) a PIRADS 3 lesion and 26% (13/50 patients) a PIRADS 5 lesion (Table 1). In all patients a successful image fusion could be realized and performed without technical problems. There was no patient in whom the image fusion was indefinable. In 66% (33/50 patients) the registration was classified as excellent or good (Table 2). Related to the PIRADS score 21.2% (7/33 patients) of the patients with a PIRADS 3 lesion, 57.6% (19/33 patients) of the patients with a PIRADS 4 lesion and 21.2% (7/33 patients) of the patients with a PIRADS 5 lesion showed an excellent or good concordance of the fusion of the two image modalities (Table 3). However, performing the chi-squared test comparing the PIRADS score with the subjective five-point scale no significant correlation could be found (p = 0.7873). Comparing the pathological findings classified according to the Gleason score with the quality of registration no statistical significant correlation could be found (p = 0.4376) (Table 4) neither.

Table 2
Number of patients according to the subjective five-point scale

Five-point scale Number of patients

1 24% (n = 12)

2 42% (n = 21)

3 22% (n = 11)

4 12% (n = 6)

5 0% (n = 0)

Five-point scale	Number of patients
1	24% (n = 12)
2	42% (n = 21)
3	22% (n = 11)
4	12% (n = 6)
5	0% (n = 0)

Table 3

Overview of the concordance of the MRI/ultrasound image fusion classified in a subjective five-point scale compared with the PIRADS score

Five-point scale	PIRADS 3	PIRADS 4	PIRADS 5	Total number of patients
1	16.7% n = 2	66.7% n = 8	16.7% n = 2	24.0% n = 12
2	23.8% n = 5	52.4% n = 11	23,8% n = 5	42.0% n = 21
3	27.3% n = 3	45.5% n = 5	27,3% n = 3	22.0% n = 11
4	16.7% n = 1	33.3% n = 2	50.0% n = 3	12.0% n = 6
5	0.0% n = 0	0.0% n = 0	0.0% n = 0	0% n = 0
Total number of patients	22.0% n = 11	52.0% n = 26	26.0% n = 13	n = 50

Table 4

Overview of pathology compared to the subjective five-point scale describing the quality of MRI/Ultrasound image fusion

Five-point scale	No malignancy	Gleason 3+3	Gleason 3+4	Gleason 4+3	Gleason 4+4	Total number of patients
1	41.7% n = 5	25.0% n = 3	16.7% n = 2	8.3% n = 1	8.3% n = 1	24.0% n = 12
2	52.4% n = 11	14.3% n = 3	14.3% n = 3	14.3% n = 3	4.8% n = 1	42.0% n = 21
3	36.4% n = 4	45.5% n = 5	18.2% n = 2	% n = 0	% n = 0	22.0% n = 11
4	16.7% n = 1	16.7% n = 1	50.0% n = 3	16.7% n = 1	% n = 0	12.0% n = 6
5	% n = 0	% n = 0	% n = 0	% n = 0	% n = 0	% n = 0
Total number of patients	42.0% n = 21	24.0% n = 12	20.0% n = 10	6.0% n = 3	8.0% n = 4	n = 50

The detection rate of PCa in the MRI fusion targeted biopsy in our study was 58% (29/50 patients) in comparison to 76% (38/50 patients) in the arm of the standard transrectal ultrasound-guided pathway. 50% (19/38 patients) of the tumor detected in the random biopsy was clinical insignificant cancer on the contrary to 41.4% (12/29) using image fusion. Clinical significant cancer (Gleason 7a or greater) was confirmed in 58.6% (17/29 patients) of the patients (Table 5) using image fusion and in 50% (19/38 patients) of the patients in the random biopsy. In 42% (21/50 patients) there was no tumor found in the target area whereas in the systematic random biopsy in 24% (12/50 patients) no prostate cancer at all could be diagnosed. 3.4% (1/29 patients) had prostate cancer in the target region only.

Table 5

Percentages of Patients with Prostate cancer in the target lesion, identified according to the PIRADS v2 score

PIRADS Score	Gleason 6	Gleason 7	Gleason 8	Total number of patients
3	75.0% n = 3	25.0% n = 1	0.0% n = 0	13.8% n = 4
4	46.7% n = 7	46.7% n = 7	6.7% n = 1	51.7% n = 15
5	20.0% n = 2	50.0% n = 5	30.0% n = 3	34.5% n = 10
Total number of patients	41.4% n = 12	44.8% n = 13	13.8% n = 4	n = 29

PCa was diagnosed in 36.4% (4/11 patients) of the patients with a PIRADS 3 lesion, in 57.7% (15/26 patients) of the patients with a PIRADS 4 lesion and in 76.9% (10/13 patients) of the patients with a PIRADS 5 lesion in the mpMRI (Table 5).

4 Discussion

MRI fusion based targeted biopsy is an emerging alternative to standard transrectal - ultrasound guided biopsy of the prostate. Although, mixed evidence exists concerning the superiority of MRI-ultrasound fusion guided biopsy of the prostate over the systematic biopsy. Several studies exist showing higher detection rates in the arm of the MRI fusion-based targeted biopsy [9 , 25]. However, there are also single-center studies showing inferiority of the MRI/ultrasound fusion targeted pathway in the detection rate of prostate cancer compared to the systematic standard transrectal-guided biopsy of the prostate [18, 22].

Performing an mpMRI prior to the biopsy leads to the identification of more clinical significant cancer lesions and reduces the detection of clinical insignificant cancer [9 , 20]. Those findings are in compliance with the results of our study with a detection rate of clinical significant cancer of 58.6% in the target region compared to 50% in the random biopsy. Our study also reflects the association of the standard 12-core biopsy with the overdetection of low grade prostate cancer [10]. 50% of the PCa detected in the random biopsy in our study was histological identified as low-grade PCa. The diagnosis of low risk PCa often leads to overtreatment, although it is known that those patients do not benefit from radical treatment options [26, 27].

The recently published PRECISION trial [9] demonstrated a PCa detection rate of 34% in PIRADS 3 lesions. This is very well comparable to the rate of 36.4% in our study.

The relatively high overall detection rate of prostate cancer in our study of 76% in the standard 12-core biopsy might be due to the use of a high– end ultrasound device Philips EPIQ 7 (Philips Medical Systems, Bothell, WA). These devices are equipped with high-resolution monitors on the one hand enabling better visualization of anatomical structures than common ultrasound devices. In addition, the higher contrast resolution delivered by specialized transrectal probes allows better demarcation of hypodense lesions in the ultrasound.

Image fusion can be done cognitive or software– assisted. Software– assisted image fusion might have some limitations as demonstrated in our study. One factor influencing the accuracy of registration was the bladder volume which often differed between the ultrasound image and the static mpMRI scan leading to different deformations of the anterior part of the prostate. The different filling of the bladder is due to the fact that the mpMRI scan and the image fusion were performed at different times. Figure 3 shows the image fusion of ultrasound and mpMRI of a PIRADS 5 lesion demonstrating that phenomenon. Another factor is the deformation of the posterior part of the prostate with the transrectal ultrasound probe. Furthermore, the concordance of the target localization in the MRI compared to the ultrasound as well as the congruence of the width of the prostate from the left to the right side often differed in both image modalities. However, those findings seem to have less impact on the histological outcome as we could not find statistical significant correlation between the Gleason score and the quality of fusion in our study. The concordance of MRI and ultrasound fusion of PIRADS 4 lesions showed no inferiority to PIRADS 5 lesions (57.6% vs 21.2% good or excellent registration). PIRADS 5 lesions are defined greater than 1.5 cm according to the PIRADS classification Version 2 [23]. This leads to the assumption that the size of the lesion described in the mpMRI does not influence the quality of registration.

Fig.3

Image fusion of a PIRADS 5 lesion. The bladder was more filled at time of the ultrasound compared to the time of mpMRI image acquisition leading to an impression of the anterior part of the prostate.

Our trial has some limitations: First, a randomization of patients is missing. All patients included in the study underwent a MRI-ultrasound fusion targeted biopsy plus the 12-core randomized standard transrectal guided biopsy of the prostate.

Second, our study is limited by the small number of patients.

5 Conclusion

MRI fusion based targeted biopsy is feasible and improves the identifiability of clinical significant tumor lesions within the prostate. Prostate fusion targeted biopsy provides a higher detection rate of clinical significant cancer compared to the standard transrectal ultrasound-guided biopsy of the prostate. Factors influencing the quality of image fusion do not have impact on the histopathological findings.

References

Ferlay

, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359–86.

Eichler

, et al. Diagnostic value of systematic biopsy methods in the investigation of prostate cancer: A systematic review. J Urol. 2006;175(5):1605–12.

Shariat

, Roehrborn

. Using biopsy to detect prostate cancer. Rev Urol. 2008;10(4):262–80.

Heidenreich

, et al. EAU guidelines on prostate cancer. Part 1: Screening, diagnosis, and local treatment with curative intent-update 2013. Eur Urol. 2014;65(1):124–37.

Xie

, et al. The utility and limitations of contrast-enhanced transrectal ultrasound scanning for the detection of prostate cancer in different area of prostate. Clin Hemorheol Microcirc. 2018. doi: 10.3233/CH-170346.

Brock

, et al. Comparison of real-time elastography with grey-scale ultrasonography for detection of organ-confined prostate cancer and extra capsular extension: A prospective analysis using whole mount sections after radical prostatectomy. BJU Int.. 2011;108(8 Pt 2):E217–22.

Sumura

, et al. Initial evaluation of prostate cancer with real-time elastography based on step-section pathologic analysis after radical prostatectomy: A preliminary study. Int J Urol. 2007;14(9):811–6.

Smeenge

, et al. Role of transrectal ultrasonography (TRUS) in focal therapy of prostate cancer: Report from a Consensus Panel. BJU Int. 2012;110(7):942–8.

Kasivisvanathan

, et al. MRI-Targeted or Standard Biopsy for Prostate-Cancer Diagnosis. N Engl J Med. 2018;378(19):1767–77.

10.

Ahmed

, et al. Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): A paired validating confirmatory study. Lancet. 2017;389(10071):815–22.

11.

Meng

, et al. Relationship between prebiopsy multiparametric magnetic resonance imaging (MRI), biopsy indication, and MRI-ultrasound fusion-targeted prostate biopsy outcomes. Eur Urol. 2016;69(3):512–7.

12.

Djavan

, et al.

Prospective evaluation of prostate cancer detected on biopsies 1, 2, 3 and 4: When should we stop?

J Urol. 2001;166(5):1679–83.

13.

Valeri

, et al. Strategies for early diagnosis and prevention of prostate cancer. Bull Cancer. 2010;97(12):1499–515.

14.

Apfelbeck

, et al. Contrast enhanced ultrasound (CEUS) with MRI image fusion for monitoring focal therapy of prostate cancer with high intensity focused ultrasound (HIFU)1. Clin Hemorheol Microcirc. 2018;69(1-2):93–100.

15.

Maxeiner

, et al. Contrast-enhanced ultrasound (CEUS) and quantitative perfusion analysis in patients with suspicion for prostate cancer. Ultraschall Med. 2018.

16.

Bratan

, et al. Influence of imaging and histological factors on prostate cancer detection and localisation on multiparametric MRI: A prospective study. Eur Radiol. 2013;23(7):2019–29.

17.

Beyer

, et al. Effect of irreversible electroporation of prostate cancer on microcirculation: Imaging findings in contrast-enhanced T1-weighted 3D MRI. Clin Hemorheol Microcirc. 2017;67(3-4):399–405.

18.

Baco

, 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 12-core systematic biopsy. Eur Urol. 2016;69(1):149–56.

19.

Schoots

, et al. Magnetic resonance imaging-targeted biopsy may enhance the diagnostic accuracy of significant prostate cancer detection compared to standard transrectal ultrasound-guided biopsy: A systematic review and meta-analysis. Eur Urol. 2015;68(3):438–50.

20.

Siddiqui

, et al. Comparison of MR/ultrasound fusion-guided biopsy with ultrasound-guided biopsy for the diagnosis of prostate cancer. Jama. 2015;313(4):390–7.

21.

Porpiglia

, et al. Diagnostic pathway with multiparametric magnetic resonance imaging versus standard pathway: Results from a randomized prospective study in biopsy-naive patients with suspected prostate cancer. Eur Urol. 2017;72(2):282–8.

22.

Tonttila

, 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 values: Results from a randomized prospective blinded controlled trial. Eur Urol. 2016;69(3):419–25.

23.

Vargas

, et al. Updated prostate imaging reporting and data system (PIRADS v2) recommendations for the detection of clinically significant prostate cancer using multiparametric MRI: Critical evaluation using whole-mount pathology as standard of reference. Eur Radiol. 2016;26(6):1606–12.

24.

Ethical guidelines for publication in clinical hemorheology and microcirculation: Update 2016

Clin Hemorheol Microcirc. 2016;63(1):1–2.

25.

Panebianco

, et al. Multiparametric magnetic resonance imaging vs. standard care in men being evaluated for prostate cancer: A randomized study. Urol Oncol. 2015;33(1):17.e1–17.e7.

26.

Hamdy

, et al. 10-Year outcomes after monitoring, surgery, or radiotherapy for localized prostate cancer. N Engl J Med. 2016;375(15):1415–24.

27.

Wilt

, et al. Follow-up of prostatectomy versus observation for early prostate cancer. N Engl J Med. 2017;377(2):132–42.