Contrast-Enhanced Ultrasound (CEUS) and Quantitative Perfusion Analysis in Patients with Suspicion for Prostate Cancer

Abstract

Purpose The aim of this study was to investigate contrast-enhanced ultrasound (CEUS) parameters acquired by software during magnetic resonance imaging (MRI) US fusion-guided biopsy for prostate cancer (PCa) detection and discrimination.

Materials and Methods From 2012 to 2015, 158 out of 165 men with suspicion for PCa and with at least 1 negative biopsy of the prostate were included and underwent a multi-parametric 3 Tesla MRI and an MRI/US fusion-guided biopsy, consecutively. CEUS was conducted during biopsy with intravenous bolus application of 2.4 mL of SonoVue^® (Bracco, Milan, Italy). In the latter CEUS clips were investigated using quantitative perfusion analysis software (VueBox, Bracco). The area of strongest enhancement within the MRI pre-located region was investigated and all available parameters from the quantification tool box were collected and analyzed for PCa and its further differentiation was based on the histopathological results.

Results The overall detection rate was 74 (47 %) PCa cases in 158 included patients. From these 74 PCa cases, 49 (66 %) were graded Gleason ≥ 3 + 4 = 7 (ISUP ≥ 2) PCa. The best results for cancer detection over all quantitative perfusion parameters were rise time (p = 0.026) and time to peak (p = 0.037). Within the subgroup analysis (> vs ≤ 3 + 4 = 7a (ISUP 2)), peak enhancement (p = 0.012), wash-in rate (p = 0.011), wash-out rate (p = 0.007) and wash-in perfusion index (p = 0.014) also showed statistical significance.

Conclusion The quantification of CEUS parameters was able to discriminate PCa aggressiveness during MRI/US fusion-guided prostate biopsy.

Zusammenfassung

Ziel Die Aussagekraft einzelner Parameter der Kontrastmittel gestützten Sonografie (CEUS) soll hinsichtlich der Detektion des Prostatakarzinoms und der Möglichkeit einer Aggressivitäts Unterscheidung bewertet werden.

Material und Methoden Im Untersuchungszeitraum von 2012 und 2014 wurden 158 von insgesamt 165 Patienten mit mindestens einer vorausgegangenen negativen Prostatastanzbiopsie eingeschlossen und erhielten eine multiparametrische 3 T (Magnetresonanztomografie) mpMRT und konsekutiv eine MRT/US fusionierte Biopsie. Während der MRT/US-Fusion wurde nach Gabe von 2,4 ml SonoVue^® (Bracco, Milan, Italy) die CEUS Untersuchung durchgeführt. Die gespeicherten Filme wurden doppelblind mittels quantitativer Perfusionsanalysesoftware (VueBox, Bracco, Milano) ausgewertet. Die Region in der MRT-Läsion mit der stärksten KM-Anreicherung wurde als führende ROI (region of interest) untersucht und alle gewonnen Parameter wurden auf das Vorliegen eines Prostatakarzinoms und Differenzierung in unterschiedliche Aggressivitätsgrade im Abgleich mit den histologischen Ergebnissen ausgewertet.

Ergebnisse Die Gesamtdetektionsrate von Prostatakarzinomen innerhalb der Kohorte betrug 74 (48 %) von 158 Patienten. Die besten Ergebnisse unter allen quantitativen CEUS Parametern in Bezug auf die Prostatakarzinomdetektion zeigten bei folgenden Parametern: rise time (p = 0,026) und time to peak (p = 0,037). Im Rahmen der Aggressivitäts Diskriminierung insbesondere in einer Subgruppenanalyse (> vs ≤ 3 + 4 = 7a (ISUP 2)), imponierten peak enhancement (p = 0,012), wash-in rate (p = 0,011), wash-out rate (p = 0,007) und wash-in perfusion index (p = 0,014) als statistisch signifikant.

Schlussfolgerung Die Analyse einzelner Quantifikationsparameter der CEUS zeigte sowohl das Potential einer Detektion von Prostatakarzinomen als auch der Diskriminierung zwischen klinisch signifikanter und nicht signifikanter Karzinome.

Key words

multiparametric MRI - CEUS - prostate cancer - MRI/US fusion-guided biopsy - quantitative perfusion

Introduction

Prostate cancer (PCa) is the most common cancer among elderly male patients in Europe and in the United States [1]. Prostate biopsy is indicated based on prostate-specific antigen (PSA) level and/or suspicious digital rectal examination (DRE). Ultrasound-guided biopsy is the standard of care and can be performed transrectally or perineally with comparable detection with both approaches [2]. Grayscale transrectal ultrasound (TRUS) has been most commonly used for prostate biopsies, due to its superiority to other imaging modalities with respect to accessibility, noninvasiveness and cost effectiveness so that it has become essential in the diagnosis and treatment of PCa [3]. The role of conventional B-mode TRUS in the detection and localization of suspicious PCa lesions is limited [4]. However, according to current guidelines, 10 – 12 systematic core biopsies should be performed under ultrasound imaging guidance for the initial diagnosis [2]. Biopsy sampling is performed without identification of lesions suspicious for cancer and consequently there is a relatively high risk of underdetection of PCa.

Thus, alternative biopsy methods have been explored to increase PCa detection and to reduce false-negative rates at initial biopsy. Multi-parametric magnetic resonance imaging (mpMRI) has the ability to detect PCa with a high sensitivity and specificity and seems to become an integral part of PCa diagnosis and detection [5]. The combination of mpMRI and US-guided biopsy known as MRI/US fusion-guided biopsy increases detection rates and also enables the detection of significant PCa (determined as cT2b and/or Gleason grade 7) in patients with prior negative biopsies [6]. More advanced ultrasound techniques including color and power Doppler, contrast-enhanced US (CEUS), strain elastography and tissue Doppler imaging (TDI) have been applied in order to improve the detection and localization of malignant lesions by ultrasound. Several studies have shown that CEUS can improve the ability of ultrasound to detect and characterize PCa [7] [8] and thus might improve cancer detection by biopsy as well as reduce the number of biopsy cores needed for diagnosis confirmation [9]. Recent studies have shown an improved PCa detection rate when CEUS is applied for US-guided prostate biopsy in comparison to systematic biopsies without CEUS [10] [11]. One of the first contrast agents used for prostate ultrasound was sulfur hexafluoride (Sono-Vue^®, Bracco, Milan), enabling real-time low mechanical index (MI) CEUS [12]. Since the reading of CEUS images was usually performed fairly subjectively, the first guidelines by the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) [13] were updated in 2011 also for non-hepatic lesions. The quantification of tissue perfusion by time-intensity curves (TIC) or dedicated software enabled a more robust detection of regions with abnormal perfusion [14]. However, the lack of both sensitivity and specificity of enhancement interpretation is still too limited and not sufficiently confirmed in clinical practice especially in PCa lesions, so that the role of CEUS in prostate cancer needs further investigation [15]. A perfusion analysis based on quantitative parameters acquired during CEUS and the identification of cut-off values for PCa without a reading bias are desirable. The purpose of this study was to evaluate quantitative CEUS parameters as predictors for PCa detection and its aggressiveness using the results of MRI/US fusion-guided biopsy as a reference standard.

Materials and Methods

Study design and population

During the study period spanning from December 2012 to January 2015, 158 out of 165 men with persistent or initial suspicion for PCa (elevated PSA levels or suspicious digital rectal examination) and with at least one negative biopsy of the prostate were included. All patients gave written informed consent according to the ethical approval of the institutional review board (Ethikkommission der Charité – Universitätsmedizin Berlin) (EA1/283/14). Patients with contraindications for MRI including unsafe implants (3 patients) or incomplete CEUS data and not able to be processed by the Vue Box software (1 patient) or refusal of biopsy including targeted and systematic biopsy (3 patients) were excluded.

MRI-US fusion-guided biopsy

Prior to biopsy, all patients underwent 3 T mpMRI including T2-weighted imaging, diffusion-weighted imaging (DWI) in all patients, and dynamic contrast-enhanced (DCE) MRI in 62.9 % (100/159) of the patients. Imaging was performed after a minimum time interval of 6 weeks after the last TRUS-guided biopsy to avoid post-biopsy hemorrhages potentially obscuring lesions on the MR images but also in the subsequent CEUS examination. All mpMRI scans were interpreted by a single experienced uro-radiologist according to the PI-RADS guidelines Version 1 [16]. MpMR images were fused to a commercial ultrasonography workstation and probe: Aplio 500 (Toshiba, Otawara, Japan; with an endfire 11C3 probe) for real-time MRI/US fusion-guided biopsy. During MRI/US fusion, the lesion with the highest PI-RADS score was marked using a colored region of interest (ROI 1). In addition, three other regions not suspicious for cancer were marked for comparative reasons during software analysis (ROI 2 – 4) but not for biopsy. Afterwards, 2D-CEUS of the predefined PI-RADS lesion was performed using contrast software with a low MI of 0.12 and the application of 2.4 mL of SonoVue^® (Bracco, Milan, Italy) intravenous contrast medium and a further 5 mL of sodium chloride 0.9 %. A low mechanical index was also used to improve the survival of the microbubbles in the circulation, and lengthen the time of parenchymal enhancement. CEUS was recorded in anonymized video clips and stored as DICOM files. Consecutively a targeted biopsy was performed within the marked ROI with the highest PI-RADS score with a median of 2 cores taken (range: 1 – 4 cores) followed by a 10-core systematic biopsy. The biopsy was performed under local anesthesia and oral fluorochinolone treatment initialized 24 h prior to biopsy. All biopsy cores were investigated by an experienced pathologist and Gleason scores were determined. If surgery was indicated and performed based on the histopathological results, postoperative Gleason scores were collected additionally.

Quantitative perfusion analysis

Data acquired during CEUS was investigated on a double blind basis using quantitative perfusion analysis software (VueBox, Bracco, Milano). The CEUS video clips were dynamically observed. The ROI was placed in the area of the most rapid and strongest enhancement with the quality of fit (QOF) > 75 %. For all ROIs parameters from the quantification tool box were collected as shown in [Fig. 1]: peak enhancement (PE) [a.u]; wash-in area under the curve (WiAUC) [a.u]; rise time (RT) [s]; mean transit time local (mTTl) [s]; time to peak (TTP) [s], wash-in rate (WiR) [a.u]; (wash-in perfusion index (WiPI) (WiAUC/RT) [a.u], wash-out AUC (WoAUC) (TTP/TO) [a.u]; wash-in and wash-out AUC (WiWoAUC) (WiAUC+WoAUC) [a.u]; fall time (FT) (TO-TTP) [s], (wash-out rate) (WoR) [a.u]. The mean value of each parameter was recorded. ROI 1 always represented the target region while ROIs 2 – 4 have been marked as reference regions for normal or benign tissue ([Fig. 2]) but not for further analysis. The differentiation between cancer (Gleason-Score ≥ 3 + 3) was investigated within ROI 1 and its associated histopathological report. Since CEUS and subsequent quantitative analysis were applied on mpMRI predefined index lesions, we hypothesized all other regions without a readout score to be potentially benign based on imaging qualities.

Fig. 1 Quantification tool box (VueBox^®©, Bracco Suisse SA) Abbreviations: PE [a.u]: peak enhancement, WiAUC [a.u]: wash-in area under the curve (TI/TTP), RT [s]: rise time (TTP-TI), mTTl [s]: mean transit time local (mTT-TI), TTP [s]: time to peak, WiR [a.u]: wash-in-rate, WiPI [a.u]: wash-in perfusion index (WiAUC/RT), WoAUC [a.u]: wash-out AUC (TTP/TO), WiWoAUC [a.u]: wash-in and wash-out AUC (WiAUC+WoAUC), FT [s]: fall time (TO-TTP), WoR [a.u]: wash-out rate, TI: start of increase, TO: end of decrease.

Fig. 2 Patient example: Contrast-enhanced ultrasound (CEUS) with quantification by time intensity curve analysis (TIC) after bolus injection of 2.4 ml contrast agent. ROI 1: arbitrary unit [a. u.] 49 375.31 (100 %); ROI 2: 1969.55 (3.99 %); ROI 3: 4201.14 (8.51 %); ROI 4: 18 843.78 (38.16 %) of a prostate cancer patient: age: 72 years, PSA: 7.7 ng/ml, digital rectal examination: not suspicious, negative pre-biopsy: 1, volume in transrectal ultrasound: 45.9 cc, PI-RAD Score: 5, pathology result of fusion-guided biopsy: Gleason 4 + 3 = 7b (ISUP 3), pathology result of systematic biopsy: 3 + 3 = 6 (overall 8/12 biopsies positive for cancer), pathology result after prostatectomy: Gleason 4 + 3 = 7b (ISUP 3).

Statistical analysis

Data analysis was performed using SPSS version 15.0 software (IBM Corp., Armonk, NY): categorical variable comparisons were performed using the chi-square or Fisher exact test, and continuous variables were evaluated with the Student t-test or Mann-Whitney U-test, as appropriate. A p-value < 0.05 was used to define significant results. Analysis of variance (ANOVA) was used to compare means of all parameters within both tested groups: malignant or normal tissue, respectively. ROC analysis was performed for the prediction of significant disease based on an estimated cut-off value.

Results

The median age of all included patients was 69 years (IQR: 61.0 – 72.0 years), the median PSA was 9.52 ng/ml (IQR: 7.00 – 14.9 ng/ml), and the median prostate TRUS volume was 54.2 cc (range: 36.9 – 74.9 cc). All patients underwent a median number of two sets of biopsies (range: 1 – 4) prior to mpMRI and fusion-guided biopsy, respectively. The median PI-RADS score was 4 (range: 2 – 5). Out of 158 patients, 74 (47 %) were diagnosed with PCa. From these 74 PCa cases, 49 (66 %) were graded Gleason ≥ 3 + 4 = 7a (ISUP 2) PCa. Cases with Gleason ≥ 4 + 3 = 7b (ISUP 3) PCa in comparison to cases with Gleason ≤ 3 + 4 = 7a (ISUP 2) PCa differed significantly in PSA (p < 0.001) and prostate volume (p = 0.010) ([Table 1]). The median PI-RADS score was 5 within patients with Gleason 7b and 4 within patients with Gleason 7a. Out of 74 detected PCa cases, 46 patients underwent radical prostatectomy subsequent to biopsy. The postoperative Gleason scores were as follows: 3 + 3 = 6 in 8 patients, 3 + 4 = 7 in 20 patients, 4 + 3 = 7 in 13 patients and 4 + 4 = 8 in 5 patients.

Table 1 Patient characteristics grouped for malignant vs. normal prostate tissue, Gleason score ≥ 7b (ISUP 3) vs. 7a (ISUP 2) and Gleason ≥ 7a (ISUP 2) vs. 6 (ISUP 1).

		median	IQR (25 %-75 %)	p-value[1]
age [y]	malignant	70.0	11.0 (63.0 – 74.0)
	normal	66.0	11.5 (59.5 – 71)	0.23
PSA [ng/ml]	malignant	9.40	19.4 (7.20 – 16.6)
	normal	9.96	6.15 (6.86 – 13.1)	0.62
vol. [ml]	malignant	44.0	27.4 (30.0 – 57.4)
	normal	59.0	35.5 (45.5 – 81.0)	< 0.01
age [y]	GL ≥ 4 + 3	70.0	11.5 (63.5 – 75.0)
	GL ≤ 3 + 4	69.0	10.0 (63.0 – 73.0)	0.39
PSA [ng/ml]	GL ≥ 4 + 3	11.0	22.76 (7.44 – 30.2)
	GL ≤ 3 + 4	8.60	5.67 (7.03 – 12.7)	0.02
vol. [ml]	GL ≥ 4 + 3	39.0	20.3 (29.0 – 49.3)
	GL ≤ 3 + 4	51.1	31.9 (33.1 – 65.0)	0.06
age [y]	GL = 3 + 3	68.5	9.00 (63.0 – 72.0)
	GL ≥ 3 + 4	70.0	12.00 (63.0 – 75)	0.32
PSA [ng/ml]	GL = 3 + 3	8.42	2.67 (7.03 – 9.7)
	GL ≥ 3 + 4	11.6	17.66 (7.34 – 25.0)	0.16
vol. [ml]	GL = 3 + 3	48.0	35.6 (29.3 – 64.9)
	GL ≥ 3 + 4	41.4	26.0 (30.0 – 56.0)	0.49

[y] – years; PSA – prostate-specific antigen; vol. – volume; IQR – interquartile range.

¹ Mann-Whitney U-test; Gl – Gleason (3 + 3 = 6 (ISUP1); 3 + 4 = 7a (ISUP2); 4 + 3 = 7b(ISUP3)).

The best results for cancer detection over all quantitative perfusion parameters based on the analysis of variance (ANOVA) were rise time (p = 0.026) and time to peak (p = 0.037) ([Table 2]). Similar results were seen in the analysis (≥ 3 + 4 = 7a (ISUP2) vs. 3 + 3 = 6 (ISUP1)) ([Table 3]): rise time (p = 0.018) and time to peak (p = 0.030). Within the subgroup analysis (< vs. ≥ 4 + 3 = 7b (ISUP 3)), peak enhancement (p = 0.012), wash-in rate (p = 0.011), wash-out rate (p = 0.007) and wash-in perfusion index (p = 0.014) showed statistical significance ([Table 4]). The correlating box plots of wash-in-parameters and the best performing wash-out-parameters can be found in [Fig. 3a – d], [4], respectively.

Table 2 Vue Box^® parameters. Malignant versus normal tissue.

		mean	95 %-CI	p-value
PE [a. u.]	malignant	21 707	16 158 – 27 256	0.568
	benign	17 159	14 841 – 24 393
WiAUC [a. u.]	malignant	113 363	87 319 – 139 407	0.807
	benign	109 271	88 017 – 130 525
RT [s]	malignant	9564	8274 – 10 854	0.026
	benign	11 800	10 336 – 13 265
mTTl [s]	malignant	66 651	49 233 – 84 069	0.605
	benign	72 784	56 884 – 88 684
TTP [s]	malignant	15 651	13 720 – 17 583	0.037
	benign	18 726	16 585 – 20 868
WiR [a. u.]	malignant	4509.0	3187.1 – 5830.9	0.602
	benign	3993.7	2567.2 – 5420.3
WiPI [a. u.]	malignant	13 936	10 410 – 17 462.0	0.502
	benign	12 367	9319.0 – 15 416.8
WoAUC [a. u.]	malignant	225 010	174 338 – 275 683	0.632
	benign	241 693	194 340 – 289 046
WiWoAUC [a. u.]	malignant	338 310	261 509 – 415 111	0.776
	benign	352 928	284 986 – 420 870
FT [s]	malignant	19 292	16 358 – 22 226	0.084
	benign	23 249	19 854 – 26 645
WoR [a. u.]	malignant	1950.3	1349.1 – 2551.4	0.395
	benign	1595.4	1025.4 – 2165.3

Abbreviations: in [Fig. 1].

Table 3 Vue Box^® parameters. Gleason ≥ 3 + 4 = 7a (≥ ISUP 2) versus Gleason 3 + 3 = 6 (ISUP 1).

		mean	95 %-CI	p-value
PE [a. u.]	GL ≥ 3 + 4	24 889	16 044 – 33 734	0.570
	GL = 3 + 3	19 113	12 683 – 25 543
WiAUC [a. u.]	GL ≥ 3 + 4	122 291	82 497 – 162 086	0.613
	GL = 3 + 3	94 289	65 057 – 123 522
RT [s]	GL ≥ 3 + 4	9.68	7.981 – 11.38	0.018
	GL = 3 + 3	8410	6716 – 1010
mTTl [s]	GL ≥ 3 + 4	53.27	42.95 – 63.59	0.860
	GL = 3 + 3	53.09	41.74 – 64.45
TTP [s]	GL ≥ 3 + 4	16.47	13.83 – 19.11	0.030
	GL = 3 + 3	14.04	11.08 – 16.99
WiR [a. u.]	GL ≥ 3 + 4	5157.5	3080.8 – 7234.2	0.732
	GL = 3 + 3	4026.5	2448.6 – 5604.5
WiPI [a. u.]	GL ≥ 3 + 4	15 738	10 166 – 21 310	0.605
	GL = 3 + 3	12 094	8025.8 – 16 163
WoAUC [a. u.]	GL ≥ 3 + 4	235 619	164 797 – 306 441	0.418
	GL = 3 + 3	194 364	132 213 – 256 515
WiWoAUC [a. u.]	GL ≥ 3 + 4	357 911	248 094 – 467 727	0.447
	GL = 3 + 3	288 654	198 551 – 378 757
FT [s]	GL ≥ 3 + 4	19.97	15.78 – 24.16	0.102
	GL = 3 + 3	17.75	13.87 – 21.64
WoR [a. u.]	GL ≥ 3 + 4	2242.3	1303.1 – 3181.5	0.356
	GL = 3 + 4	1613.1	977.12 – 2249.1

Abbreviations: in [Fig. 1].

Table 4 Vue Box^® parameters. Gleason ≥ 4 + 3 = 7b (ISUP 3) versus Gleason ≤ 3 + 4 = 7a (ISUP 2).

		mean	95 %-CI	p-value
PE [a. u.]	GL ≥ 4 + 3	31 439	17 255 – 45 622	0.012
	GL ≤ 3 + 4	16 741	12 603 – 208 880
WiAUC [a. u.]	GL ≥ 4 + 3	146 008	85 628 – 206 388	0.071
	GL ≤ 3 + 4	96 360	71 551 – 121 169
RT [s]	GL ≥ 4 + 3	9.52	6.98 – 12.05	0.965
	GL ≤ 3 + 4	9.58	8.05 – 11.11
mTTl [s]	GL ≥ 4 + 3	50.15	35.49 – 64.82	0.179
	GL ≤ 3 + 4	75.07	49.78 – 100.36
TTP [s]	GL ≥ 4 + 3	16.92	12.99 – 20.84	353
	GL ≤ 3 + 4	15.00	12.79 – 17.21
WiR [a. u.]	GL ≥ 4 + 3	6826.7	3477.8 – 10 175	0.011
	GL ≤ 3 + 4	3301.9	2329.2 – 4274.6
WiPI [a. u.]	GL ≥ 4 + 3	19 871	10 927 – 28 814	0.014
	GL ≤ 3 + 4	10 844	8221.0 – 13 468
WoAUC [a. u.]	GL ≥ 4 + 3	271 538	161 266 – 381 810	0.179
	GL ≤ 3 + 4	199 724	147 123 – 252 324
WiWoAUC [a. u.]	GL ≥ 4 + 3	417 546	247 662 – 587 431	0.130
	GL ≤ 3 + 4	295 247	217 781 – 372 713
FT [s]	GL ≥ 4 + 3	18.78	12.71 – 24.84	0.798
	GL ≤ 3 + 4	19.57	16.27 – 22.87
WoR [a. u.]	GL ≥ 4 + 3	3035.9	1535.64 – 4536.3	0.007
	GL ≤ 3 + 4	1360.2	942.59 – 1777.8

Abbreviations: in [Fig. 1].

Fig. 3 Box plot of the “wash-in-parameters”: a (RT) – rise time [s]. b (TTP) – time to peak [s]. c (PE) – peak enhancement [a. u.]. d (WiR) – wash-in rate [a. u.].

Fig. 4 Box plot of the best performing “wash-out parameter”: (WoR1) – wash-out rate [a. u.].

The receiver operating characteristic analysis showed a substantial advantage of the transit time with an AUC of 0.643 ([Fig. 5]), even though statistically insignificant. The optimal transit time cut-off (Youden’s index) of 40.75 s was based on the highest sensitivity (0.67) and specificity (0.36) for cancer discrimination.

Fig. 5 ROC (receiver operating characteristic) of (mTT) – mean transit time [s].

A correlation of quantitative ultrasound perfusion parameters and the established PI-RADS score was not observed.

Discussion

The detection of PCa based on TRUS alone is difficult due to the heterogeneity of lesion echogenicity [17]. The percentage of detected PCa lesions based on a hypoechoic appearance is reported as 18 – 57 % with sensitivities between 15 – 96 % and specificities between 46 – 93 % [18] [19]. In combination with CEUS, the diagnostic accuracy of PCa was improved effectively [11]. To our knowledge, CEUS is mainly used in expert centers and only a few studies investigated CEUS quantification for determination of PCa location and aggressiveness [20].

However, our study showed the potential of CEUS and its quantification parameters to discriminate between PCa and normal prostate tissue. The parameters rise time and time to peak, which are both wash-in parameters, performed significantly, reflecting the widely hypothesized hypervascularity due to angiogenesis in neoplastic growth [21].

Furthermore, it has been reported that angiogenesis is essential for prostate tumors to progress from small indolent lesions less than 2 mm in size to a clinically significant disease [22]. The identification of clinically significant cancer is one of the major goals and advantages in MRI/US fusion-guided biopsy, since overdiagnosis of insignificant tumors and consequently unnecessary radical treatment may be avoided [8]. Based on our data, the differentiation between clinically significant and insignificant PCa was also possible based on selected wash-in parameters (peak enhancement, wash-in rate, and wash-in perfusion index). Since the differentiation between clinically significant and insignificant cancer is crucial, our results further separated Gleason 7a (ISUP 2) and 7b (ISUP 3) significantly. Especially Gleason 7 (ISUP 2 and 3) again shows the heterogeneity of PCa and its outcome prediction in patients after surgery for clinically localized PCa, as a statistically significant recurrence-free survival advantage was reported for Gleason 7a (ISUP 2) compared with Gleason 7b (ISUP 3) [23]. Invariably a wash-out parameter also showed its capability to discriminate between Gleason 7a (ISUP 2) and 7b (ISUP 3), emulating the potential of quantification parameters for the characterization of PCa and also in the detection of higher Gleason scores due to higher microvessel density [14] [24]. Active surveillance is mainly based on a Gleason score 3 + 3 = 6. However, the CEUS parameters within the comparison of Gleason 3 + 3 = 6 to ≥ 3 + 4 = 7 showed similar results as the analysis of benign versus malignant. A potential explanation could be that the absolute values in 3 + 3 = 6 patients seem to be very close to the benign values. Hence, CEUS parameters could be potentially limited in low-risk patients. However, well selected patients, depending on, for example, age, comorbidity, extent of cancer, MRI findings, and patient choice, could also be candidates for active surveillance with Gleason score 3 + 4 = 7 if the pattern 4 is limited, and in particular if pattern 4 is < 10 % [25]. Percent pattern 4 is currently not included in most pathology reports but would be beneficial in patients being considered for active surveillance and was also recently requested by the International Society of Urological Pathology (ISUP consensus conference 2014 on Gleason Grading). Therefore, in our opinion, a differentiation between Gleason 7a (ISUP 2) and 7b (ISUP 3) is crucial for individualized treatment strategies and seems to be more reliable in our current data.

However, the microvascular blood supply to prostate tissue is more uniform in malignant tissue but with more arbitrary pathways. Due to the presence of benign prostatic hyperplasia, a reduction of the total volume and the vessels also needs to be mentioned, since in benign tissues the capillaries are restricted for the most part to the periglandular stroma immediately adjacent to the epithelium, unlike the random distribution in carcinoma [26]. Hence, CEUS seem to perform well in aggressive PCa lesions and emerging alternative treatment strategies for localized PCa including high intensity focused ultrasound (HIFU), cryoablation and irreversible electroporation (IRE) in patients not eligible for prostatectomy or radiation therapy could be monitored in the future by CEUS. 

One of the potential disadvantages of CEUS is the subjective interpretation by the investigator since the diagnostic criteria of a cancerous lesion, such as early enhancement, usually relies on the visual impression of the examiner. Therefore, it is crucial to have also a reliable quantitative evaluation for better reproducibility of the examination results. From the literature we know that the peak intensity of a cancerous lesion is significantly higher than that of a benign lesion, i. e., in prostatic hyperplasia. Hence, the combination of visual impression and quantitative assessment could lead to improvement of the performance of targeted biopsies. Better localization of the lesion pre-located on MRI during fusion is conceivable but also reevaluation of a lesion is potentially possible. Thus, CEUS could be used as an additional tool in patients who underwent mpMRI and to confirm the PI-RADS score potentially resulting in up-scoring or down-scoring of PI-RADS lesions, in particular PI-RADS 3 lesions. Patients who cannot undergo MRI because of metal implants may also benefit from improved mpUS modalities such as CEUS. However, it is important to note that the PI-RADS score is based on findings in the whole prostate, and 2D-CEUS particularly represents a lesion-focused approach. Thus, perfusion analysis is limited to one cross-section because of the two-dimensionality of the probe. But in the future, with the use of a 3 D probe, the accuracy of perfusion analysis may be further improved.

Our overall detection rate of 74 (47 %) and the detection of 49 out of 74 (66 %) cases with a Gleason ≥ 3 + 4 = 7a (ISUP ≥ 3) also underlines the power of MRI/US fusion-guided biopsy in this setting. Since targeted as well as systematic biopsies were performed in our study setting, cancer-suspicious lesions were biopsied twice in almost every case. Thus, 44 PCa cases out of 49 were detected in both techniques. However, a further investigation of the performance of MRI/US fusion-guided biopsy in comparison to conventional biopsy was not conducted due to our major interest in the quantification of CEUS. 

Despite a positive role for CEUS in PCa detection, we acknowledge further limitations of the present study. Primarily our study represents a single-center analysis with a relatively low number of patients, whereas multicenter studies are desirable. In addition, the positioning of ROI 2 – 4 of likely benign lesions as well as the lack of histological correlation of these regions need to be mentioned as a potential study bias. To our knowledge, there are only a few studies [27] [28] investigating the potential value of using quantification software for the interpretation of CEUS in the prostate as well as other organs and thus further long-term multi-modality imaging studies should be conducted. Furthermore, we did not include an analysis of interreader agreement which seems to be increased based on CEUS [29]. Thus, quantitative software tools for measuring time intensity curves during CEUS could improve the interpretation of CEUS raw data and should therefore be considered for future work.

In conclusion, CEUS during MRI/US fusion seems to visualize hemodynamic properties of the prostate and PCa. The quantification of CEUS parameters has shown some potential in differentiating between malignant and non-malignant prostate lesions and further in the identification of clinically significant cancer. The use of CEUS quantification might increase the role of ultrasound in the diagnostic pathway of PCa especially in addition to a previously acquired mpMRI scan.

Abbreviations

PCa: prostate cancer

DRE: digital rectal examination

TRUS: transrectal ultrasound

mpMRI: multi-parametric magnetic resonance imaging

CEUS: contrast-enhanced ultrasound

MI: mechanical index

EFSUMB: European Federation of Societies for Ultrasound in Medicine and Biology

TIC: time-intensity curves

DWI: diffusion-weighted imaging

DCE: dynamic contrast-enhanced

ROI: region of interest

PE: peak enhancement

WiAUC: wash-in area under the curve

RT: rise time

mTTl: mean transit time local

TTP: time to peak

WiR: wash-in rate

WiPI: wash-in perfusion index

WoAUC: wash-out AUC

WiWoAUC: wash-in and wash-out AUC

FT: fall time

WoR: wash-out rate

a. u.: arbitrary unit

Conflict of Interest

The authors of this manuscript declare relationships with the following companies: Alexander Baur has received payments as a speaker for Bayer Healthcare. Bernd Hamm has worked as a consultant for Toshiba Medical and received research grants for the department of radiology from Toshiba Medical, Siemens, and Bracco Imaging. Ernst Michael Jung has received payments as a speaker for Bracco Imaging. Thom Fischer has received payments as a speaker for Toshiba Medical and Bracco Imaging.

Acknowledgments

We cordially thank Gunilla Maxeiner for carefully reading the manuscript.

^# These authors contributed equally

• References

• 1 Heidenreich A, Bastian PJ, Bellmunt J. et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. European urology 2014; 65: 124-137
• 2 Mottet N, Bellmunt J, Bolla M. et al. EAU-ESTRO-SIOG Guidelines on Prostate Cancer. Part 1: Screening, Diagnosis, and Local Treatment with Curative Intent. European urology 2017; 71: 618-629
• 3 Sarkar S, Das S. A Review of Imaging Methods for Prostate Cancer Detection. Biomedical engineering and computational biology 2016; 7 (Suppl. 01) 1-15
• 4 Halpern EJ, Strup SE. Using gray-scale and color and power Doppler sonography to detect prostatic cancer. American journal of roentgenology 2000; 174: 623-627
• 5 Puech P, Rouviere O, Renard-Penna R. et al. Prostate cancer diagnosis: multiparametric MR-targeted biopsy with cognitive and transrectal US-MR fusion guidance versus systematic biopsy--prospective multicenter study. Radiology 2013; 268: 461-469
• 6 Moore CM, Robertson NL, Arsanious N. et al. Image-guided prostate biopsy using magnetic resonance imaging-derived targets: a systematic review. European urology 2013; 63: 125-140
• 7 Halpern EJ, Verkh L, Forsberg F. et al. Initial experience with contrast-enhanced sonography of the prostate. American journal of roentgenology 2000; 174: 1575-1580
• 8 Maxeiner A, Stephan C, Durmus T. et al. Added Value of Multiparametric Ultrasonography in Magnetic Resonance Imaging and Ultrasonography Fusion-guided Biopsy of the Prostate in Patients With Suspicion for Prostate Cancer. Urology 2015; 86: 108-114
• 9 Mitterberger M, Horninger W, Pelzer A. et al. A prospective randomized trial comparing contrast-enhanced targeted versus systematic ultrasound guided biopsies: impact on prostate cancer detection. The Prostate 2007; 67: 1537-1542
• 10 Koh J, Jung DC, Oh YT. et al. Additional Targeted Biopsy in Clinically Suspected Prostate Cancer: Prospective Randomized Comparison between Contrast-Enhanced Ultrasound and Sonoelastography Guidance. Ultrasound in medicine & biology 2015; 41: 2836-2841
• 11 Uemura H, Sano F, Nomiya A. et al. Usefulness of perflubutane microbubble-enhanced ultrasound in imaging and detection of prostate cancer: phase II multicenter clinical trial. World journal of urology 2013; 31: 1123-1128
• 12 Mitterberger M, Pinggera GM, Horninger W. et al. Comparison of contrast enhanced color Doppler targeted biopsy to conventional systematic biopsy: impact on Gleason score. The Journal of urology 2007; 178: 464-468 ; discussion 8
• 13 Albrecht T, Blomley M, Bolondi L. et al. Guidelines for the use of contrast agents in ultrasound. Ultraschall in der Medizin 2004; 25: 249-256
• 14 Jung EM, Wiggermann P, Greis C. et al. First results of endocavity evaluation of the microvascularization of malignant prostate tumors using contrast enhanced ultrasound (CEUS) including perfusion analysis: first results. Clinical hemorheology and microcirculation 2012; 52: 167-177
• 15 Delgado Oliva F, Arlandis Guzman S, Bonillo Garcia M. et al. Diagnostic performance of power doppler and ultrasound contrast agents in early imaging-based diagnosis of organ-confined prostate cancer: Is it possible to spare cores with contrast-guided biopsy?. European journal of radiology 2016; 85: 1778-1785
• 16 Barentsz JO, Richenberg J, Clements R. et al. ESUR prostate MR guidelines 2012. European radiology 2012; 22: 746-757
• 17 Dahnert WF, Hamper UM, Eggleston JC. et al. Prostatic evaluation by transrectal sonography with histopathologic correlation: the echopenic appearance of early carcinoma. Radiology 1986; 158: 97-102
• 18 Frauscher F, Klauser A, Halpern EJ. Advances in ultrasound for the detection of prostate cancer. Ultrasound quarterly 2002; 18: 135-142
• 19 Schlenker B, Clevert DA, Salomon G. Sonographic imaging of the prostate. Der Urologe Ausg A 2014; 53: 1052-1060
• 20 Smeenge M, Mischi M, Laguna Pes MP. et al. Novel contrast-enhanced ultrasound imaging in prostate cancer. World journal of urology 2011; 29: 581-587
• 21 Cantisani V, Bertolotto M, Weskott HP. et al. Growing indications for CEUS: The kidney, testis, lymph nodes, thyroid, prostate, and small bowel. European journal of radiology 2015; 84: 1675-1684
• 22 Russo G, Mischi M, Scheepens W. et al. Angiogenesis in prostate cancer: onset, progression and imaging. BJU international 2012; 110: E794-E808
• 23 Han M, Snow PB, Epstein JI. et al. A neural network predicts progression for men with gleason score 3+4 versus 4+3 tumors after radical prostatectomy. Urology 2000; 56: 994-999
• 24 Halpern EJ, Rosenberg M, Gomella LG. Prostate cancer: contrast-enhanced us for detection. Radiology 2001; 219: 219-225
• 25 Chen RC, Rumble RB, Loblaw DA. et al. Active Surveillance for the Management of Localized Prostate Cancer (Cancer Care Ontario Guideline): American Society of Clinical Oncology Clinical Practice Guideline Endorsement. Journal of clinical oncology: official journal of the American Society of Clinical Oncology 2016; 34: 2182-2190
• 26 Bigler SA, Deering RE, Brawer MK. Comparison of microscopic vascularity in benign and malignant prostate tissue. Human pathology 1993; 24: 220-226
• 27 Hyvelin JM, Gaud E, Costa M. et al. Characteristics and Echogenicity of Clinical Ultrasound Contrast Agents: An In Vitro and In Vivo Comparison Study. Journal of ultrasound in medicine: official journal of the American Institute of Ultrasound in Medicine 2017; 36: 941-953
• 28 Wildner D, Pfeifer L, Goertz RS. et al. Dynamic contrast-enhanced ultrasound (DCE-US) for the characterization of hepatocellular carcinoma and cholangiocellular carcinoma. Ultraschall in der Medizin 2014; 35: 522-527
• 29 Stang A, Keles H, Hentschke S. et al. Incidentally detected splenic lesions in ultrasound: does contrast-enhanced ultrasonography improve the differentiation of benign hemangioma/hamartoma from malignant lesions?. Ultraschall in der Medizin 2011; 32: 582-592