Site Editor

Sandy Srinivas, MD


ASCOBT 2023: Radiomic Model May Improve the Ability to Predict Prostate Cancer Recurrence

By: Julia Fiederlein Cipriano, MS
Posted: Thursday, August 24, 2023

Multiparametric MRI–derived radiomic features have been demonstrated to detect subvisual patterns for the quantitative characterization of tumor phenotype. Michael Baine, MD, PhD, of the University of Nebraska Medical Center, Omaha, and colleagues sought to develop a multiparametric MRI–based radiomic model and compare its diagnostic performance with that of currently available nomograms. Their findings, which were presented during the 2023 American Society of Clinical Oncology (ASCO) Breakthrough meeting (Abstract 76) in Yokohama, Japan, revealed this novel tool’s capacity to predict biochemical recurrence of prostate cancer following radical prostatectomy.

After undergoing radical prostatectomy for the treatment of localized prostate cancer, 76 neoadjuvant and adjuvant therapy–naive patients received multiparametric MRI. Biochemical recurrence was defined as two consecutive serum prostate-specific antigen values of at least 0.2 ng/mL.

The investigators determined the following radiomic features to be “important and nonredundant” in predicting prostate cancer recurrence: least material condition; gray-level nonuniformity; shape elongation; shape sphericity; and first-order skewness. They were aggregated to construct the novel model; repeated five-fold cross-validation yielded an area under the curve (AUC) of 0.95 (± 0.06), a sensitivity of 33%, and a specificity of 100%. The University of California, San Francisco (UCSF) Cancer of the Prostate Risk Assessment (CAPRA) and Memorial Sloan Kettering Cancer Center (MSK) Pre–Radical Prostatectomy nomograms demonstrated AUC values of 0.72 (± 0.07) and 0.82 (± 0.07), respectively.

“The multiparametric MRI–derived radiomic model performed well when compared with the UCSF-CAPRA score and MSK Pre–Radical Prostatectomy nomogram,” the investigators concluded. “Future projects will incorporate patient demographics and disease characteristics available at the time of initial prostate cancer diagnosis to improve the radiomic model’s accuracy.”

Disclosure: The study authors reported no conflicts of interest.

By continuing to browse this site you permit us and our partners to place identification cookies on your browser and agree to our use of cookies to identify you for marketing. Read our Privacy Policy to learn more.