Original papers
2012, Vol. 14, no. 3, 182-186Abstract
Aim: To determine the diagnostic quality of transrectal sonoelastography (SE) in the prediction and localisation of prostate cancer, we prospectively examined patients treated with radical prostatectomy in our urology department. Methods: From April 2010 to January 2011, 61 patients with biopsy-proven prostate cancer underwent preoperative transrectal gray-scale (b- mode) ultrasound and SE of the prostate. Cancer-suspicious areas were documented for b-mode and SE, dividing the prostate into six topographic sectors. Suspicious areas in both modalities were compared to tumour localisation in the prostatectomy specimen. Sensitivity, specificity, positive- and negative predictive values were calculated for both investigation techniques.
Results: Prostate cancer was present in 232 of 366 pathological sectors (62 %). B-mode ultrasound showed 113 suspicious sectors, while SE indicated prostate cancer in 157 areas. The precise localisation of at least one pathologically confirmed cancerous lesion was possible in 42/61 (69 %) patients by b-mode ultrasound and 56/61 (92 %) patients by SE (P<0.005).
The sensitivity for b-mode ultrasound was 33 % and specificity 74 %. For SE sensitivity was 53 %, while specificity was 74
%. Conclusions: SE offers a more precise localisation of prostate carcinoma than conventional ultrasound. To investigate the possible advantages of SE in during prostate biopsy and its value in the prediction of extracapsular cancer further studies are required.
Keywords: prostate cancer, elastography, ultrasound, prostatectomy
The utility of transrectal sonoelastography in preoperative prostate cancer assessment
Steffen Rausch
1, Wibke Alt
2, Hartmut Arps
2, Berthold Alt
1, Tilman Kälble
11 Department of Urology, 2 Department of Pathology, Klinikum Fulda, Fulda, Germany
Received 11.04.2012 Accepted 29.05.2012 Med Ultrason
2012, Vol. 14, No 3, 182-186
Corresponding author: Steffen Rausch MD,
Department of Urology, Klinikum Fulda gAG, Fulda, Germany
Pacelliallee 4, D-36043 Fulda, Phone: +49(0)661-845951 Fax: +49(0)661-845952
Email: [email protected]
Introduction
As continuous efforts are undertaken to improve prostate cancer diagnosis, imaging represents a diagnos- tic sector with a high potential to offer relevant informa- tion on cancer staging and therapy planning. Within the setting of prostate biopsy, improved imaging modalities could assist investigators in identifying and localising cancer suspicious areas for targeted samples. So far, no imaging modality has established itself as practical in a broad clinical routine to close these diagnostic gaps, and
meanwhile clinical studies are being conducted to evalu- ate various diagnostic procedures. To date, 10-12 core randomised prostate biopsy remains the gold standard for cancer diagnosis [1-3], with acceptance of its known limitations.
Sonoelastography (SE) has recently been described as an emerging imaging modality for prostate carcino- ma [4-9]. As available data on SE are merely based on Hitachi technology and heterogeneous towards patient numbers and sensitivities, we prospectively evaluated our experience using an ultrasound device from an alter- native manufacturer (Siemens).
Methods
From April 2010 to January 2011, 61 patients with biopsy-proven prostate carcinoma designated to under- go radical prostatectomy at our Department of Urology were preoperatively examined by transrectal gray-scale (b-mode) ultrasound and transrectal SE. A Siemens
(62%). Histological tumour detection was most promi- nent in the mid gland area (44%), apical carcinoma was detected in 38%, while basal tumours occurred in 18% of all cases. There was no difference regarding the side of tumour incidence. Figure 1 shows an exem- plary examination picture and the corresponding post- operative histological analysis. Patient characteristics and postoperative stage distribution are shown in table I. The correct detection of at least one pathologically confirmed tumour lesion (index tumour) was possible for 56 patients (92%) by SE, whereas conventional ul- trasound predicted cancer localisation correctly in 42 (69%) patients (P<0.005). Overall, the sensitivity for the correct localisation of prostate cancer was 33% us- ing gray-scale ultrasound. The specificity accounted for 74%, the respective positive predictive value and nega- tive predictive values were 70% and 38%. SE allowed Acuson S2000 ultrasound device with an EC9-4 tran-
srectal transducer and eSie Touch™ elasticity imag- ing was used to conduct the studies. Written informed consent was obtained by all participating patients. All patients were investigated by the same examiner who was blinded to all preoperatively available data, such as level of prostate specific antigen (PSA), digital rectal examination, and the pathological reports of prior pros- tate biopsies. Patients were examined in the left lateral position. In order to validate the potential of localising prostate carcinoma in both imaging modalities, suspi- cious areas were documented in six topographic sectors, left and right prostate apex, mid gland and base. Areas were classified as suspicious, when a demarked, hypo- echogenic pattern was present in gray-scale imaging.
Elastographic images were generated by soft and rhyth- mic compression of the prostate using the ultrasound transducer.
Tumour suspicion was defined as a demarked section of increased tissue stiffness, suggested by colour scaling and a particular quality factor of the elastography soft- ware (>70), indicating quality of compression and repro- ducibility of images.
In both imaging modalities, the investigation focused on the peripheral zone of the prostate. Furthermore, pros- tate size and morphological suspect of capsular excess or seminal vesicle invasion were recorded. The included patients all underwent open radical retro-pubic prosta- tectomy with bilateral iliac lymph node dissection. The histological analysis was performed by two pathologists.
Prostatectomy specimens were step sectioned and proc- essed according to a standard protocol. Postoperatively the localisation of suspicious areas in prostate imaging was compared to the pathology results. Sensitivity, spe- cificity, positive- (PPV) and negative predictive values (NPV) for conventional ultrasound and SE were calcu- lated. To evaluate the influence of prostate size on diag- nostic quality, we analysed sensitivity in dependence of prostate volume. A chi-square test was used to determine statistical significance.
Results
We prospectively examined 61 patients with pros- tate carcinoma prior to open radical retro-pubic prosta- tectomy. Median preoperative PSA was 6.40ng/ml. The mean age at surgery was 63.8 years. Median prostate size measured 37 cm³, with a range of 16 cm³ to 78 cm³.
In conventional gray-scale ultrasound a total of 113 sec- tors were classified as suspicious for prostate cancer, SE indicated prostate cancer in 157 areas. Pathological analysis detected prostate cancer in 232 of 366 sectors
Fig 1. Examination pictures and histopathological analysis – A:
elastographic image without pathological findings at the pros- tate apex/urethral intersection (QF:75=quality factor) – B: sus- picious area in the right mid-gland section, the marked sector (box) is shown in C (QF:75)– C: histological slide with confir- mation of prostate cancer according to B.
a more precise detection of tumour infiltration with a sensitivity of 52% and a specificity of 74%. Positive- and negative predictive values are shown in table II. In comparison to small prostate volumes, the sensitivity for glands larger than 40 cm³ was reduced.
Due to the low number of patients with extracapsular prostate cancer growth in our collective no clear conclu- sion can be drawn for the worthiness of SE in terms of predicting extracapsular extension or seminal vesicle in- filtration.
Discussion
In accordance to previous single centre reports, our data underline the potential capacity of prostate SE for prostate cancer assessment. A closer view at the cur- rently available clinical reports on SE shows that the re- sults are either based on the comparison of preoperative prostate imaging and postoperative histological cancer localisation or on clinical trials with patients undergoing elastography-guided biopsies according to specific pro- Table I. Patient characteristics.
Age at surgery, mean (IQR) 63.8 y (59.02-69.81y)
PSA. median (IQR) 6.4 ng/ml (4.97–8.75ng/ml)
Prostate volume, median (range) 37 cc (16 – 79cc)
Clinical staging T1 39 (64%)
T2 20 (33%)
T3 2 (3%)
Pathological staging pT2 40 (66%)
pT3a 15 (25%)
pT3b 6 (10%)
pT4 -
Pathological Gleason Score ≤ 6 37 (61%)
7a 8 (13%)
7b 8 (13%)
≥ 8 8 (13%)
Nodal status N0 57 (93%)
N+ 4 (7%)
IQR=inter-quartile range
Table II. Sensitivity, specificity, positive and negative predictive values for correct localisation of prostate cancer using conventional ultrasound and sonoelastography.
Suspicious area, B-mode 113/366
Suspicious area, SE 157/366
Tumour foci, histology 232/366
Sensitivity Specificity PPV NPV
B-mode (95% CI) 33% (27-40) 74% (65-81) 70% (61-78) 38% (32-44)
SE (95% CI) 52% (45-58) 73% (65-81) 78% (70-84) 46% (39-53)
Size ≤ 40cc, n=38 59% (50-56) 71% (61-80)
Size > 40cc, n=23 42% (32-52) 73% (65-81)
Detection index tumour, b-mode 42/61 (62%) Detection index tumour, SE 56/61 (92%)
Chi-square-test P<0.005
SE=sonoelastography, CI= confidence interval, PPV=positive predictive value, NPV=negative predictive value
tocols. [4-9]. Relating to the used ultrasound devices, it is conspicuous that the majority of studies are based on Hitachi systems; only few available data rest upon other ultrasound systems and techniques. Table III illustrates the heterogeneity of published data regarding the sensi- tivity of SE.
With regard to technical variability, different systems of imaging and scaling represent a relevant difference.
To obtain reproducible images during Siemens-SE the system implements a quality factor indicating the meas- ure of (passive) pressure and examination quality. In our study, we observed a sensitivity of 53% and a positive predictive value of 78% for Siemens-SE. Especially the low sensitivity represents an unsatisfying result in rela- tion to the requirements for local prostate cancer imag- ing and in comparison to previous studies and varying ultrasound devices.
In our opinion it is speculative whether these differ- ences are based on a minor diagnostic power of the used SE device or potentially a lower examiner experience.
Previous publications agree with the statement that SE is an examiner dependant diagnostic tool [4,9].
Despite a lack of available data regarding the prac- tical use of the Siemens device, we were able to obtain a significant improvement of prostate cancer detection with SE compared to conventional transrectal ultrasound.
Sedelaar et al. systematically reviewed the literature on prostate ultrasound revealing that the biopsy of a suspi- cious hypo-echogenic area detects prostate cancer in only 17-57% [10]. The results are in line with our own obser- vations. The sensitivity for gray-scale ultrasound was low, accounting for only 33%. As in the majority of conducted studies on prostate SE, cancer assessment was performed on an indirect basis, comparing preoperatively stored ul- trasound pictures to postoperative pathological analysis.
In order to increase the accuracy of analysing a suspicious electrographic region and to compare the practicability of SE-guided biopsies in contrast to diagnostic standard 10- 12 core biopsies, further clinical trials should follow.
Artefacts during SE are a common finding and repre- sent a problem concerning the interpretation of examina- tion results. In accordance to our own experience with Siemens-SE, sensitivity of elastography decreases with increasing prostate size and the presence of benign pros- tate hyperplasia (BPH) [5]. In our series, the observed sensitivity of 42% for glands larger than 40 cm³ was no- ticeably low in contrast to 59% sensitivity for prostates smaller than 40 cm³. Some reports quote that routinely appearing examination artefacts, like the “soft-rim-arte- fact”, a soft layer surrounding the prostate tissue as an indication of the adventitial connective tissue, could pre- dict capsular infiltration or extracapsular growth [9,11].
Since our own observations and the majority of available studies are based on the examination results of patients with merely organ confined disease, larger series are re- quired to further validate these observations. The impact of prostate carcinoma differentiation on SE results is controversial. As Miyagawa and co-workers report from equality of diagnostic sensitivities of SE in relation to the Gleason score, a publication by Sumura et al. and the analysis of Salomon et al. showed an increased sensitiv- ity for higher graded carcinoma [5,9,12]. In contrast, an- other Japanese single centre trial obtained best detection rates for low grade prostate cancer [8].
As most modern ultrasound devices offer a variety of examination techniques, combinations of e.g. SE, con- ventional gray-scale ultrasound, colour-Doppler or con- trast enhanced ultrasound are generally feasible in order to raise detection rates. Myagawa et al. observed an im- proved detection of prostate cancer combining SE and conventional gray-scale transrectal ultrasound [5].
Magnetic resonance imaging (MRI) has been de- scribed as a valid instrument referring to local prostate cancer staging and MRI guided biopsy has demonstrated its potential value in clinical trials. Turkbey et al observed a sensitivity of 60.7% and a specificity of 72.7% for the prediction of prostate cancer applying multi-parametric MRI [13]. Recently, an expert panel established stand- Table III. Published data regarding the sensitivity and specificity of sonoelastography for prostate cancer detection [4,7,8,9,12].
Author Journal n Design Sensitivity /specificity Device
Salomon et al. Eur Urol (2008) 109 comparison SE/ histology 75.4%/76.6% Hitachi
Eggert et al. Urologe (2008) 351 biopsy 44,5% / 89% Voluson, GE medical
Pallwein et al. Eur J Radiol (2008) 492 biopsy 87% / 72% Hitachi
Tsutsumi et al. Int J Clin Oncol
(2007) 51 comparison SE/ histology Sensitivity 57-94% Hitachi Sumura et al. Int J Urol (2007) 17 comparison SE/ histology Sensitivity 74,1% Hitachi n= pacients‘ number
ardized conduct and interpretation for the use of MRI in the diagnosis and staging of prostate cancer. Neverthe- less, a general recommendation for the routine use of multi-parametric MRI should follow further validation in prospective studies [14-15].
With regard to costs, eligibility and the duration of the examination, SE is a promising alternative diagnos- tic tool with good feasibility in an ambulatory setting, which warrants further efforts in technical development and clinical investigation.
The investigator dependency of SE and the indirect approach of comparing preoperatively stored images to postoperative histological slides represent major limita- tions of our study.
Conclusions
Transrectal sonoelastography allows a more precise prostate cancer localisation than conventional ultra- sound. SE may close diagnostic gaps in local staging and in the setting of prostate biopsy, especially in absence of extensive BPH. As the method is closely related to the experience of the examiner and the specifics of the used device and technique, studies are needed to evaluate these influences. Furthermore, it is worthwhile to deter- mine the potential of SE in optimizing cancer detection during prostate biopsy.
Acknowledgements: none.
Conflicts of interest: none to declare.
References
1. National Comprehensive Cancer Network (NCCN). Pros- tate Cancer Early Detection. Clinical Practice Guidelines in Oncology. V.2.2007. Fort Washington: NCCN; 2007.
2. Eichler K, Hempel S, Wilby J, Myers L, Bachmann LM, Kleijnen J. Diagnostic value of systematic biopsy methods in the investigation of prostate cancer: a systematic review.
J Urol 2006; 175: 1605-1612.
3. NHS Cancer Screening Programmes. Undertaking a tran- srectal ultrasound guided biopsy of the prostate. PCRMP
Guide No 1. Sheffield: NHS Cancer Screening Pro- grammes; 2006.
4. Eggert T, Khaled W, Wenske S, Ermert H, Noldus J. Impact of elastography in clinical diagnosis of prostate cancer. A comparison of cancer detection between B-mode sonogra- phy and elastography-guided 10-core biopsies. Urologe A 2008; 47: 1212-1217.
5. Miyagawa T, Tsutsumi M, Matsumura T, et al. Real-time elastography for the diagnosis of prostate cancer: evalu- ation of elastographic moving images. Jpn J Clin Oncol 2009; 39: 394-398.
6. Nelson ED, Slotoroff CB, Gomella LG, Halpern EJ. Tar- geted biopsy of the prostate: the impact of color Doppler imaging and elastography on prostate cancer detection and Gleason score. Urology 2007; 70: 1136-1140.
7. Pallwein L, Mitterberger M, Struve P, et al. Real-time elas- tography for detecting prostate cancer: preliminary experi- ence. BJU Int 2007; 100: 42-46.
8. Tsutsumi M, Miyagawa T, Matsumura T, et al. The impact of real-time tissue elasticity imaging (elastography) on the detection of prostate cancer: clinicopathological analysis.
Int J Clin Oncol 2007; 12: 250-255.
9. Salomon G, Köllerman J, Thederan I, et al. Evaluation of prostate cancer detection with ultrasound real-time elastog- raphy: a comparison with step section pathological analysis after radical prostatectomy. Eur Urol 2008; 54: 1354-1362.
10. Sedelaar JP, Vijverberg PL, De Reijke TM, et al. Transrec- tal ultrasound in the diagnosis of prostate cancer: state of the art and perspectives. Eur Urol 2001; 40: 275-284.
11. Salomon G, Graefen M, Heinzer H, et al. The value of real- time elastography in the diagnosis of prostate cancer. Urol- oge A 2009; 48: 628-636.
12. Sumura M, Shigeno K, Hyuga T, Yoneda T, Shiina H, Igawa M. Initial evaluation of prostate cancer with real-time elas- tography based on step-section pathologic analysis after radical prostatectomy: a preliminary study. Int J Urol 2007;
14: 811-816.
13. Turkbey B, Xu S, Kruecker J, et al. Documenting the lo- cation of systematic transrectal ultrasound-guided prostate biopsies: correlation with multi-parametric MRI. Cancer Imaging 2011; 11: 31-36.
14. Dickinson L, Ahmed HU, Allen C, et al. Magnetic reso- nance imaging for the detection, localisation, and charac- terisation of prostate cancer: recommendations from a Eu- ropean consensus meeting. Eur Urol 2011; 59: 477-494.
15. Heidenreich A. Consensus criteria for the use of magnetic resonance imaging in the diagnosis and staging of prostate cancer: not ready for routine use. Eur Urol 2011; 59: 495-497.
