OR WAIT null SECS
New York-Endorectal magnetic resonance imaging can be usedprior to salvage prostatectomy to identify tumor sites and showextracapsular extension and seminal vesicle invasion with"reasonable accuracy," say investigators from MemorialSloan-Kettering Cancer Center.
However, interobserver variability and limited sensitivity to predict seminal vesicle invasion, both of which were observed in the Sloan-Kettering study, are reasons to proceed with caution, according to one urologist.
In the current study, published in Radiology (2006; 238:176-83), primary radiation treatments among the 45 patients had included external beam radiation therapy, brachytherapy, and combined therapy. Seventeen of the patients also received chemotherapy or hormonal therapy before salvage prostatectomy.
Patients had to be otherwise healthy, with a life expectancy of at least 10 years, along with having a locally confined, biopsy-determined recurrence of cancer and no evidence of metastasis.
Two radiologists at Sloan-Kettering independently read MR images for tumor localization and determination of local stage. Their individual interpretations were then compared with pathologic findings from surgical specimens, and inter-rater variability was estimated with the kappa statistic.
Areas under the receiver operating characteristic curve (AUCs) were used to determine the accuracy of endorectal MRI in tumor detection and determination of extracapsular extension and seminal vesicle invasion. AUCs-all with 95% confidence interval-were:
Dr. Akin said that endorectal MRIs are now commonly used with these patients at Sloan-Kettering.
"Training in prostate cancer MR imaging and experience is necessary for accurate evaluation," he said.
Raj Pruthi, MD, director of urologic oncology at the University of North Carolina, Chapel Hill, expressed concern over the potential for significant variability among those reading MR images.
"This is a study performed by a group with tremendous experience and expertise in prostate MRI," Dr. Pruthi said. "One concern is that the application of endorectal MRI before salvage radical prostatectomy is subject to significant interobserver variability, as reported by the authors, even when performed within the same institution by such a highly skilled group.
"The concern is that such variability in interpretation is likely to be even more substantial when this technique is compared between different institutions and by different radiologists with varying experience."
Dr. Pruthi had also hoped that endorectal MRI could aid in surgical planning by, for example, predicting seminal vesicle invasion.
"Unfortunately, the sensitivity to predict invasion was limited in this study: only 38% to 62%," he said.