Published on 21 October 2013
BERKELEY, CA (UroToday.com) - Standard greyscale transrectal ultrasound (TRUS) guided prostate biopsies are routinely used for men suspected to have prostate cancer (PCa). However, the grey- scale TRUS is not a reliable modality in differentiating cancer and normal hyperplasia of prostate. In this protocol-driven feasibility study, we used a quantitative shear wave imaging (SWI) for prostate cancer diagnosis. SWI is quantitative and has much less operator dependence, thus potentially can improve detection and characterisation of cancers. In the study, men (n = 50) suspected to have PCa were subjected to 12-core systematic biopsies under grey-scale ultrasound, and only one additional biopsy was taken if region of interest (ROI) detected by SWI was found outside the previously biopsied areas. Among these patients, 66% had PCa. In patients with PSA < 20 μg/L, the sensitivity and specificity of SWI for PCa detection was 0.9 and 0.88, respectively. While in patients with PSA > 20 μg/L, the sensitivity and specificity was 0.93 and 0.93, respectively. This indicates potential advantage of SWI in detecting cancer foci within the prostate gland. In addition, PCa had significantly higher stiffness values compared to benign tissues (p < 0.05), with a trend toward stiffness differences in different Gleason grades. These results suggest a better diagnostic accuracy of SWI than grey-scale ultrasound imaging.
Before conducting study on patients and to address the issue of adequate training with SWI, a commercially available phantom (Model 066 Prostate Elastography Phantom/CIRS Tissue Simulation and Phantom Technology, Virginia, USA) was used. This model contains 3 iso-echoic lesions that are three times stiffer than the simulated surrounding prostate tissue. Under greyscale ultrasound they cannot be detected, but are readily visible on SWI ().
Additionally, the phantom was used for optimisation of the prostate gland sonographic examination technique and assessment of inter-observer variations.
- The abnormal nodules (representative of cancerous tissues) within the phantom were only visible with SWI (shown as yellow (medium stiffness)- in figures i & ii).
Reproducibility of observations was assessed between 2 independent operators analysing the same image pairs acquired using the phantom. The observers recorded stiffness (minimum, maximum and mean with SD) in ROI by placing a cursor over the colour-coded area. As the cursor is moved around the ROI (i.e., stiffer area of the phantom), values are instantaneously displayed on the screen in a data box. Stiffness values averaged over 3 images were used for analysis.
The abnormal foci within the prostate phantom were visible only on SWI (with a higher stiffness than surrounding areas). At least three independent experiments were performed using the phantom (). The mean (±SD) Young modulus (kPa) values of the representative stiff nodule (61.3 kPa (±1.5) within the phantom was much higher than of the surrounding normal areas (22.5kPa (±0.9). The mean stiffness on each pair of SWI images acquired by these 2 independent operators led to an intraclass correlation coefficient of 0.93 (95% CI 0.62 to 0.99).
Table - Summary of the results of three independent experiments on single Prostate Elastography Phantom – the stiffness of the the tumour nodules was nearly three times higher than surrounding normal phantom tissues in all three examinations. Furthermore, this difference was statistically significant (p < 0.05).
Transrectal SWI is much closer to a standard TRUS clinical examination as it does not require any additional compression. Still, there is a learning curve to fully master the use of SWI. The pre-clinical part of our study on phantoms suggested that around 10 independent experiments (personal experience) on phantoms are required to gain adequate experience in the identification of abnormal stiffness patterns within the prostate.
§ The numbers of participants in this preliminary study were relatively small. The study was limited to a single-centre, and the numbers of observers and operators were therefore relatively small. However, these results strongly suggest the importance of performing similar studies in multiple centres on a larger cohort of patients.
§ The impact of distance of cancer foci from the rectal probe on SWI sensitivity is not addressed in this study. This may have an important implication and need to be addressed in future studies.
§ On a per-patient basis, we do not really know how much more likely one was to find cancer in an SWI-targeted biopsy versus standard systematic biopsy. This is an important issue, however; it can only be addressed with randomised clinical trials comparing SWI and greyscale-targeted systematic biopsies.
Conclusions and future directions
§ SWI can reliably differentiate between benign and malignant prostate tissue and, in most cases, can aid in targeting abnormal foci by needle biopsy. The stiffness of cancer foci was 50 % higher than of the benign tissues (133.7 kPa (±57.6) vs 74.9 kPa (± 47.3), respectively).
§ SWI provides quantitative data and is a reproducible technique. Quantitative assessments of maximum and mean stiffness measurements were found to be highly reliable and consistent across multiple SWI acquisitions though multiple operators in the phantom part of the study. Through such characteristics, the real time quantitative SWI imaging reported in this study has the potential to change clinical practice of PCa by improving the localisation and allowing limited targeted biopsies of suspicious areas, thereby reducing both the complications and cost associated with the current standard of care.
Furthermore the accurate localisation may play a role in focal therapy for PCa, and SWI can be utilised to differentiate a “significant” PCa from an “indolent” one.
Sarfraz Ahmad, MBBS, PhD, MRCSI, MRCSEd as part of Beyond the Abstract on UroToday.com. This initiative offers a method of publishing for the professional urology community. Authors are given an opportunity to expand on the circumstances, limitations etc... of their research by referencing the published abstract.
Academic Section of Urology, Division of Cancer, Medical Research Institute, Ninewells Hospital, University of Dundee, Dundee, DD1 4HN, UK