![2019 04 17 20 48 2514 Prostate 3 D 400](https://img.auntminnie.com/files/base/smg/all/image/2023/04/am.2019_04_17_20_48_2514_prostate_3D_400.png?auto=format%2Ccompress&q=70&w=400)
A new microstructure imaging technique, hybrid multidimensional MRI (HM-MRI), could enable better targeting of areas in the prostate that need to be biopsied, according to a presentation at the American Roentgen Ray Society annual meeting in Honolulu.
"This study validates [that an] HM-MRI based tool can identify clinically significant prostate cancer -- equal to or more than Gleason 3 + 4 -- more reliably than random biopsy and/or areas detected by a radiologist in patients undergoing MR-ultrasound fusion biopsy," presenter Aritrick Chatterjee, PhD, of the University of Chicago told session attendees.
HM-MRI depicts volumes of tissue components such as the stroma, epithelium, and lumen by "fitting the MRI data into a three-compartment signal model," Chatterjee explained.
He and colleagues conducted a study that included 85 patients with a mean prostate-specific antigen (PSA) score of 8.1 ng/ml; median time between MR imaging and biopsy was seven days. Patients underwent both a standard prostate MRI and an HM-MRI exam; all had a transrectal ultrasound (TRUS)-guided biopsy. The median time between MR imaging and biopsy was seven days.
The group found that HM-MRI had higher sensitivity per tumor of the prostate but not per patient. It had a lower negative predictive value on a per-patient basis.
Comparison of mpMRI and HM-MRI for identifying clinically significant prostate cancer | ||
Measure | mpMRI | HM-MRI |
Per-patient | ||
Sensitivity | 100% | 90% |
Specificity | 15% | 40% |
Positive predictive value | 41% | 47% |
Negative predictive value | 100% | 88% |
Accuracy | 46% | 58% |
Area under ROC curve | 0.58 | 0.65 |
Per-tumor | ||
Sensitivity | 70% | 78% |
Positive predictive value | 22% | 34% |
By sextant (i.e., from ultrasound-guided biopsies of six different areas of the prostate) | ||
Sensitivity | 50% | 68% |
Specificity | 82% | 86% |
Positive predictive value | 19% | 29% |
Negative predictive value | 95% | 97% |
Accuracy | 79% | 85% |
Area under the ROC curve | 0.66 | 0.77 |
The findings highlight the fact that radiologists need to consider the HM-MRI data with care.
"[We] have to look at these results very carefully and consider all three," he said.
HM-MRI shows promise for improving MR-ultrasound fusion biopsy results by offering more accurate clinical information compared to PI-RADS-based radiologist evaluation, Chatterjee concluded.
"The advantage of this tool is its ability to provide automated, quantitative, reproducible results," he said.