Diagnostic Accuracy of Microultrasound in Patients with a Suspicion of Prostate Cancer at Magnetic Resonance Imaging: A Single-Institutional Prospective Study - Beyond the Abstract
MicroUS, a novel transrectal US modality operating at 29MHz, beyond the regular 8-12MHz frequencies, allows for an increased resolution (down to 70µm) and the identification and targeting of suspicious US lesions. Tissue analysis and risk stratification of prostate areas are possible using a risk protocol called PRI-MUS (Prostate Risk Identification using microUS), similar to PI-RADS in mpMRI evaluation.
MicroUS was introduced in our center in October 2017 and has since been used in the prostate biopsy clinic for the identification and real-time targeting of suspicious lesions not seen by regular TRUS or mpMRI. Concomitantly it was adopted in the outpatient TRUS clinic, where it was applied for screening of patients with clinical suspicion of prostate cancer.
Given the positive initial experience, a protocol was outlined in order to compare the diagnostic accuracy of microUS versus mpMRI in the identification of clinically significant prostate cancer.
A total of 320 subjects have been selected for the study due to clinical suspicion of prostate cancer and a positive prostate mpMRI (i.e. at least one lesion PI-RADS ≥ 3). The systematic and MRI-guided prostate biopsy was preceded by microUS evaluation of the prostate with real-time targeting of any suspicious area. The location and PRI-MUS score of each lesion identified was recorded on a worksheet and two biopsy cores per area were obtained. The procedure was then completed with standard systematic biopsy and MRI-US software fusion-targeted biopsy of the mpMRI identified lesions, whose location and PI-RADS score was unknown to the urologist performing the procedure until this point.
The diagnostic performance of microUS in determining the presence of clinically significant prostate cancer (csPCA; i.e. GS ≥ 7) was evaluated. The per-patient sensitivity and specificity were determined, and a multivariable logistic regression analysis was fitted to determine the predictors of csPCa.
MicroUS revealed suspicious lesions in the majority of subjects (79.7%): 74.1% of patients had at least one concordant lesion identified by both microUS and mpMRI, with a concordant risk category assigned in 61.0% of them.
Overall, 116 (36.3%) patients were diagnosed with csPCa: microUS was able to detect csPCa with high accuracy in these patients, with a specificity of 89.7% and negative predictive values (NPV) of 81.5%. A comparative analysis with the diagnostic performance of mpMRI was not feasible in this study since the population was selected starting from a suspicious mpMRI. This may also explain the low specificity and PPV of microUS, at 26.0% and 40.8% respectively, since the urologist, while blinded to mpMRI results, was aware of the presence of a PI-RADS ≥ 3 lesions.
Our study confirmed that the addition of targeted biopsies improves the csPCa diagnosis compared to randomized systematic sampling alone: 27 (23.2%) of csPCa diagnoses were obtained via targeted cores and missed by random biopsies. Of these, 24 csPCa were independently diagnosed by targeted microUS cores and by targeted mpMRI cores. Additionally, the added value of targeting microUS suspicious lesions was demonstrated by the identification of 20 (17.2%) csPCa on microUS target cores, where mpMRI targeted sampling resulted negative. Conversely, mpMRI targeting was able to diagnose 14 (12.1%) csPCa patients that were missed by microUS target cores.
Furthermore, in the subpopulation of patients where microUS and mpMRI showed topographically discordant lesions (66 subjects, 25.9%), each diagnostic modality was independently able to lead to csPCa diagnosis via target cores in 12/19 (63%) cases.
The PRI-MUS score assigned during microUS evaluation was also confirmed to be statistically significantly associated with the presence of csPCa. At multiple linear regression (MLR) analysis, the presence of a PRI-MUS 3 or a PRI-MUS 5 lesion was associated with a 3.1- or a 6.2-fold higher risk of being diagnosed with csPCa (p<0.005). Thanks to the addition of microUS risk assessment, the accuracy of the predictive model based on common clinical predictors (age, prostate-specific antigen [PSA], prostate volume, initial biopsy setting) was increased to 81.2% from 77%.
Of note, this study represents the initial experience with microUS examination at our center. The learning curve of two dedicated urologists, both experts in standard TRUS, was hence included in the study and may have lowered the accuracy of microUS. Further studies are necessary in order to accurately determine the length of the training curve and assess the inter-observer variability in microUS examination.
In conclusion, the results of our study showed that the addition of microUS to standard biopsy provided a similar improvement in csPCa detection rate to that offered by the addition of mpMRI-targeted biopsies, as recommended by current European Association of Urology (EAU) guidelines. These findings could pave the way for future studies evaluating the adoption of microUS in the initial biopsy setting. Patients could indeed benefit from the single-setting, less-expensive nature of US-based biopsies. Indeed, while mpMRI value in improving csPCa diagnosis and its cost-effectiveness has been shown, it remains a time-consuming procedure, whose implementation in standard practice outside large clinical centers still remains troublesome and is penalized by long waiting lists.
What’s more, these two imaging methodologies were shown to provide complementary information that can be used both in the initial and repeat biopsy setting. Their combination in a single procedure appears feasible and promises to maximize the csPCa detection rate and reducing the need for further prostate biopsies.
Written by: Giovanni Lughezzani, Davide Maffei, Alberto Saita, Marco Paciotti, Pietro Diana, Nicolò Maria Buffi, Piergiuseppe Colombo, Grazia Maria Elefante, Rodolfo Hurle, Massimo Lazzeri, Giorgio Guazzoni, Paolo Casale
Department of Urology, Humanitas Clinical and Research Center, Rozzano, Italy; Humanitas University, Department of Biomedical Sciences, Pieve Emanuele, Milan, Italy. Electronic address: ., Department of Urology, Humanitas Clinical and Research Center, Rozzano, Italy; Humanitas University, Department of Biomedical Sciences, Pieve Emanuele, Milan, Italy., Department of Urology, Humanitas Clinical and Research Center, Rozzano, Italy., Department of Pathology, Humanitas Clinical and Research Center, Rozzano, Italy.
Read the Abstract