The role of endorectal coil MRI in patient selection and treatment planning for prostate seed implants.

PURPOSE: To assess the role of endorectal coil magnetic resonance imaging (MRI) staging for patients undergoing seed implantation (SI) with or without external beam radiotherapy (EBRT).
MATERIALS AND METHODS: Between October 1994 and December 1998, 390 patients underwent prostate SI (98% Pd-103, 2% I-125). Seventy-six percent of patients had a prostate serum antigen (PSA) < 10, 17% had PSA of 10-20, and 7% of patients had PSA of > 20. Ten percent of patients had a Gleason score (GS) of 4-5, 54% had GS 6, 29% had GS 7, and 7% had GS >/= 8. Monotherapy was employed in 46% of patients, and the remaining 54% received combined EBRT and SI. Three hundred twenty-seven were staged by high-resolution phased array pelvic coil, or in most cases, an endorectal coil MRI. The MRI findings were used to guide stage-appropriate treatment recommendations, and to assist in the preplanning and optimization of seed distributions. The criteria utilized to determine MRI-based stage were founded on the reported literature from the University of Pennsylvania. All MRI studies were reviewed by C.A., D.B., or W.H., who were unaware of clinical stage at the time of their review. The biopsy report was available to them as the only clinical correlate.
RESULTS: Of the 327 patients staged by MRI, 70% were upstaged from the digital rectal examination-based clinical stage; 26% of T(1), T(2) patients were upstaged to T(3). Perineural invasion and the percentage of positive cores predicted for T(3) MRI stage (p < 0.0001 for both variables). MRI findings changed the overall treatment recommendation in 60/327 (18%) patients. The majority of these patients were advised to receive combined therapy instead of monotherapy after the MRI documented more extensive disease. The seed distribution was modified in 183/327 (56%) patients, mostly related to preplanned extracapsular coverage of bulky or extraprostatic disease seen on MRI. With a mean follow-up of 38 months (range 3-72), PSA freedom from progression (FFP) was 94% at 5 years. Cox regression analysis showed that only the percentage of positive cores (p = 0.001) and failure to have MRI staging (p = 0.0008) predicted for failure. Pretreatment PSA level, Gleason score, perineural invasion, and external beam radiotherapy did not significantly predict for PSA failure. We compared our MRI T(3) intermediate-risk group patients treated by combined therapy with a previous study of T(3) intermediate-risk group treated by radical prostatectomy (RP) at the University of Pennsylvania. Our 36-month PSA FFP was 94% compared with 21% for the previous study's RP patients.
CONCLUSION: MRI is a valuable staging procedure for prostate cancer patients treated by SI. PSA FFP results appear to be improved by MRI staging. MRI T(3) disease can be treated more effectively by SI + EBRT than by RP.