WCE 2018: Multiparametric MRI Features Identify Aggressive Prostate Cancer At the Phenotypic and Transcriptomic Level

Paris, France (UroToday.com) In the setting of prostate cancer, multiparametric MRI (mpMRI) is a valuable diagnostic tool. This has shown promising results in diagnosis, localization, and even staging of clinically significant prostate cancer. T2 morphologic assessment and functional assessment by diffusion imaging remains the mainstay for diagnosis of prostate cancer on mp-MRI. A downside of this imaging technique is the limited data on prognostic value, especially because the assessment is subjective Dr. Beksac spoke on how his team sought to evaluate whether mpMRI could be used to identify certain aggressive cancer features of the prostate at both the morphological and genomic level.

A retrospective analysis of 206 patients who underwent radical prostatectomy between 2013 and 2017 was done. All of these patients had RNA expression data available and the location of RNA expression analysis corresponded to a preoperative mpMRI location. The association between Prostate Imaging Reporting and Data Systems (PI-RADS) score and adverse pathology features were analyzed. Adverse pathology was defined as any grade group 3, > pT3a, or any node-positive disease. Finally, they utilized a multivariate logistic regression model to look for any predictors of adverse pathology.

Under multivariate analysis, the risk factors associated with adverse pathology were: PI-RADS score (Ref: PI-RADS 3; odds ratio [OR]: 8.1; 95% confidence interval [CI]: 1.2–57.5, p = 0.04), Grade group (Ref: Gleason 3; OR: 5.6; 95% CI: 1.5–21.1; p = 0.01), and prostate specific antigen (OR:1.103; 95% CI: 1.011–1.203). In univariate analysis, the lesion size (p = 0.03), PI-RADS score (p = 0.02) and extraprostatic extension (p = 0.01) were associated with adverse pathology. Both PI-RADS and Grade group were associated with adverse pathology in univariate analysis as well.

The research group observed differential pathway expressions between the PI-RADS 5 group compared with PI-RADS 4. Certain signaling pathways (PI3K-AKT-mTOR, WNT-beta, and E2F) were more active in PI-RADS 5. Furthermore, the heatwave map (below) showed a good correlation between these pathways and PI-RADS score. 

UroTodayWCE2018 Multiparametric MRI Features Identify Aggressive Prostate Cancer At the Phenotypic and Transcriptomic Level

The PI-RADS score was associated with adverse pathology and differential genomic pathway activation. This suggests a possible role of mpMRI in predicting certain aggressive tumor behavior.


Presented by: Alp Tuna Beksac, Fellow, Icahn School of Medicine at Mount Sinai, New York, New York

Co-Authors: Shivaram Cumarasamy, cahn School of Medicine at Mount Sinai, New York, New York, Ugo Falagario, University of Foggia, New York, New York, Paige Xu, New York, New York, Mandeep Takhar, San Diego, California, Mohamed Alshalalfa, Vancouver, British Columbia, Canada, Akriti Gupta, Sonya Prasad, Stephanie Hectors, Jennifer Jordan, Sujit Nair, Kenneth Haines, Kamlesh Yadav, Isuru Jayaratna,  New York, New York, Alberto Martini, Fellow, Icahn School of Medicine at Mount Sinai, New York, New York,  Elai Davicioni, Vancouver, British Columbia, Canada, Sara Lewis, Assistant Professor of Radiology, Icahn School of Medicine at Mount Sinai, New York, New York, Art Rastinehad, The Icahn School of Medicine at Mount Sinai, Department of Urology, New York, New York , Bachir Taouli, Professor of Radiology, Icahn School of Medicine at Mount Sinai, New York, New York, Ashutosh Tewari, Icahn School of Medicine at Mount Sinai, New York, New York

Written by: John Sung, Department of Urology, University of California-Irvine, medical writer for UroToday.com at the 36th World Congress of Endourology (WCE) and SWL - September 20-23, 2018 Paris, France