Model adds no value
A decision curve analysis was conducted to assess the clinical utility of a predictive model. Specifically, it examined whether the model provided a net benefit compared with a treat-all strategy when setting the threshold probability of having extensive disease at <20%. The analysis showed the model added no value.
Dr. Takeda suggested that the study was limited by small sample size.
“We identified 770 patients who were diagnosed with unilateral prostate cancer, met the criteria for focal therapy, and underwent radical prostatectomy, but only 98 of them had mpMRI data and tumor maps,” he said.
In addition, Dr. Takeda noted that in the years covered by the study period, biopsy was performed prior to MRI. That sequence can result in artifacts in the MRI.
“Approximately 40% of our patients with an MRI tumor score of 4 or 5 had a biopsy negative lobe. To avoid a confounding effect from biopsy, the MRI should be done first,” he said, adding that a future study may include a more contemporary group of patients from MSKCC who have undergone targeted biopsy after MRI.
Dr. Takeda also believes that a future study designed to identify variables for selecting appropriate candidates for hemi-ablative focal therapy might use less stringent criteria for defining extensive disease.
“The definition used in the present study included all Gleason pattern 4 or 5 in bilateral lobes. However, it may be appropriate to treat some men who have a very small volume of Gleason 4 disease with hemi-ablation followed by active surveillance,” Dr. Takeda told Urology Times.
More from Urology Times:
Subscribe to Urology Times to get monthly news from the leading news source for urologists.