Automatic detection of prostate cancer using quantitative perfusion parameters in contrast-enhanced ultrasound. Poster No.: C-1798 Congress: ECR 2016 Type: Scientific Exhibit Authors: M. Skendi, A. KHAIROUNE, C. Delavaud, A.-M. Tissier, T. le 1 1 2 3 4 1 5 2 1 guilchet, T. Fresneau, P. Frinking, O. Hélénon, J.-M. Correas ; 1 2 3 4 Paris/FR, Paris, Cedex 15/FR, Houilles/FR, Plan les Ouates/ 5 CH, Geneva/CH Keywords: Contrast agent-intravenous, Ultrasound, Oncology, Genital / Reproductive system male, Cancer DOI: 10.1594/ecr2016/C-1798 Any information contained in this pdf file is automatically generated from digital material submitted to EPOS by third parties in the form of scientific presentations. References to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not in any way constitute or imply ECR's endorsement, sponsorship or recommendation of the third party, information, product or service. ECR is not responsible for the content of these pages and does not make any representations regarding the content or accuracy of material in this file. As per copyright regulations, any unauthorised use of the material or parts thereof as well as commercial reproduction or multiple distribution by any traditional or electronically based reproduction/publication method ist strictly prohibited. You agree to defend, indemnify, and hold ECR harmless from and against any and all claims, damages, costs, and expenses, including attorneys' fees, arising from or related to your use of these pages. Please note: Links to movies, ppt slideshows and any other multimedia files are not available in the pdf version of presentations. www.myesr.org Page 1 of 14
Aims and objectives To define quantitative perfusion parameters of prostate cancer as well as normal prostate tissue using contrast-enhanced transrectal ultrasonography (CE-TRUS) and evaluate a prototype software that allows automatic detection of prostate cancer in the peripheral zone correlated to systematic and targeted biopsies. Methods and materials A retrospective study of 109 patients (mean age = 66.5 ± 8, mean PSA 9.25 ± 8) referred for TRUS-guided biopsies from 2011 to 2015. CE-TRUS was performed with an Aplio XG/500 (Toshiba MS, Japan) using the THI mode after 2.4 ml intravenous administration of SonoVue (BR1, Bracco). The 2D (11C3) or 3D (9CV3) transducer was kept still over the most suspicious area detected at B-mode imaging in order to acquire a 30-40 sec DICOM cine loop during microbubble transit. Significant PCas were defined as cancers with Gleason score #6 and cancer size > 2mm per core at pathology. First: Perfusion parameters obtained from manually drawn regions of interest (ROIs). Time-intensity curves were processed using General Imaging VueBox software (Bracco) to calculate twelve perfusion parameters: peak enhancement (PE), wash-in rate (WiR), wash-out rate (WoR), wash-in area under the curve (WiAUC), wash-out area under the curve (WoAUC), wash-in and wash-out area under the curve (WiWoAUC), product of wash-in and wash-out rate (WiR.WoR), rise time (RT), mean transit time local (mttl), fall time (FT), time to peak (TTP) and wash-in perfusion index (WiPI= WiAUC/RT). The operator localized a ROI onto the most suspicious area of interest and a reference ROI on the contralateral peripheral zone and time-intensity curves for both ROI were obtained. All perfusion parameters obtained from the suspicious area were normalized using the ones from the contralateral peripheral zone as reference. The operator was blinded to the biopsy results. Statistical analysis of data was performed using the wilcoxon rank test. Second: Automatic analysis. Page 2 of 14
The prototype prostate-vuebox software, an operator-independent technique, calculated the dispersion of the WiR from parametric maps. The analysis of these parameters was correlated with results from targeted and systematic prostate biopsies. Images for this section: Fig. 1: General Imaging VueBox software, Bracco. Suspicious peripheral zone (yellow); contralateral peripheral zone (magenta); total prostate (green); transitional zone(white). - Paris/FR Page 3 of 14
Fig. 2: Prototype prostate-vuebox software (Bracco) analysis using the same cine-loop as in fig.1. Suspicious left peripheral zone (red), transition zone (red), non suspicious peripheral zone (green). Pathology results: PCa, Gleason score 3+4, cancer size of 15 mm on biopsy core. - Paris/FR Page 4 of 14
Results 112 focal peripheral zone lesions (49 cancerous) were analyzed. Cancerous nodules were significantly more hypervascular than benign nodules (median peak 170 ± 49% vs 96 ± 35%). In addition, the wash-in rate was higher in the cancerous group (231±84% vs 103±68%). The most important parameters that showed a significant difference between the cancer (K) versus non cancerous area (NK) in order of significance were the WiPI (WiAUC/RT), PE, WiR.WoR, WiR, WoR, WiAUC, WiWoAUC, WoAUC, (p<0.001). The other temporal perfusion parameters such as rise time (RT), fall time (FT), time to peak (TTP) were also significantly different between the K and NK group (p<0.05). The mean transit time local (mttl) was the only parameter that did not show a significant difference between K and NK areas (p=0.74) The performance of prototype prostate-vuebox for PZ cancer detection with a Gleason score # 6 (3+3, 3+4, 4+3, 4+4, 4+5) showed a sensitivity (Se) of 80%, specificity (Spe) of 78%, positive predictive value (PPV) of 74% and negative predictive value (NPV) of 83%. The automatic analysis by prototype prostate-vuebox for PCa with a Gleason score #7 (3+4, 4+3, 4+4, 4+5) showed a Se of 90%, Spe of 70%, PPV of 45% and NPV of 95%. Images for this section: Page 5 of 14
Table 1: Population characteristics. - Paris/FR Page 6 of 14
Fig. 3: The normalized ratio between the suspicious and contralateral area (%) is represented on the y-axis. Twelve perfusion parameters in cancer (K) and non cancerous (NK) groups are represented on the x-axis. *p<0.001 **p<0.05 - Paris/FR Page 7 of 14
Fig. 4: Perfusion parameters difference between the cancer (K) versus non cancerous reference area (NK) in order of significance. - Paris/FR Page 8 of 14
Fig. 5: Example of a false positive case. The red-orange area on the left PZ was prostate tissue inflammation on pathology. The central red area is the transition zone. - Paris/FR Page 9 of 14
Fig. 6: Example of a false negative case. No suspicious area (red) is seen in the peripheral zone, the red area seen here is the left transition zone of the prostate. Pathology results: PCa was detected on right side of the PZ, Gleason score 3+3, cancer size of 6+8+10+10 mm on biopsy cores. - Paris/FR Page 10 of 14
Conclusion CE-TRUS quantitative approach enables a better understanding of the tumoral angiogenesis of PCa. The automatic detection of prostate cancer in the peripheral zone is an innovative operator-independent imaging technique that can improve PCa treatment indications. Personal information Mariela Skendi Ultrasonographer, Paris, France. Medical Resident Pierre et Marie Curie University, Paris, France. Master of Science, Radiology Department at Necker University Hospital and Center for Research and Indisciplinarity, Paris, France. marielaskendi@gmail.com Images for this section: Page 11 of 14
Page 12 of 14
Fig. 7 Mariela Skendi Page 13 of 14
References 1. 2. 3. 4. 5. 6. 7. Postema AW et al. Dynamic contrast-enhanced ultrasound parametric imaging for the detection of prostate cancer. 2015 BJU Int. Xie SW et al. Contrast-enhanced ultrasonography with contrast-tuned imaging technology for the detection of prostate cancer: comparison with conventional ultrasonography. 2012 BJU Int; 109:1620-6 Pitre-Champagnat S et al. Dynamic contrast-enhanced ultrasound parametric maps to evaluate intratumoral vascularization. 2015 Investigative Radiology. 50(4):212-217 Zhao HX et al. The value and limitations of contrast-enhanced transrectal ultrasonography for the detection of prostate cancer. 2013. European Journal of Radiology. Volume 82, Issue 11, Pages e641-e647 Cheikh AB et al. Evaluation of T2-weighted and dynamic contrast-enhanced MRI in localizing prostate cancer before repeat biopsy. 2009 European Radiology. Volume 19, Issue 3, pp 770-778. Turkbey B et al. Documenting the location of systematic transrectal ultrasound-guided prostate biopsies: correlation with multi-parametric MRI. 2011 Cancer Imaging. 11(1): 31-36. Hara N et al. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a useful modality for the precise detection and staging of early prostate cancer. 2005 The Prostate. Volume 62, Issue 2, pages 140-147 Page 14 of 14