Combination external beam and brachytherapy for intermediate- to high-risk prostate cancer: Comparison of high-dose rate (HDR) and seed (LDR) implant methods.

Abstract

90 Background: Maximizing the therapeutic ratio for intermediate to high risk localized prostate cancer may be accomplished by using a combination of external beam radiotherapy (EBRT) and radiation implant. Differences in implant methods could be expected to influence mortality. We compared outcomes for patients treated using combination EBRT and high dose rate (HDR) versus EBRT and low dose rate (LDR) seed implant methods. Methods: Between 12/1993 and 05/2010 there were 307 cases of combination therapy with analyzable data: 227 HDR + EBRT (HDRe) and 81 LDR + EBRT (LDRe) Intermediate risk patients had 1+ risk factors: PSA ≥ 10, Gleason score (GS) ≥ 7, or AJCC clinical stage ≥ T2b. High risk patients exhibited either 2+ of these risk factors or PSA ≥ 20, GS ≥ 8, or AJCC clinical stage ≥ T3. External beam radiation was 45 Gy in 1.8 Gy daily fractions using 3D conformal or IMRT methods. The prostate, seminal vesicles, and (in some patients) pelvic lymph nodes were treated. HDR brachytherapy was delivered one week prior to beginning EBRT. Afterloading HDR catheter needles were surgically implanted into the prostate. HDR dose was 21 Gy in three equal fractions of 7 Gy peripheral dose given over two days. LDR was performed 2-3 weeks after EBRT using I-125 or Pd-103 seeds and 109 or 90 Gy respectively. PSA failure was defined as “nadir plus two.” Results: HDRe and LDRe cases were well matched for pretreatment risk factors. There were 18 PSA failures in the HDRe group (7.9%) and 6 PSA failures in the LDRe group (7.4%). Adjusted Cox proportional hazard models showed no statistically significant difference between the treatment groups with regard to PSA failure. Hazard ratio for LDRe compared to HDRe was 0.82 (95% CI 0.28, 2.46; p=0.73). There were 26 deaths from any cause (11.45%) in the HDRe group and 11 deaths from any cause (13.58%) in the LDRe group. Adjusted Cox models showed no statistically significant difference between the treatment groups for survival. Hazard ratio for LDRe compared to HDRe was 1.49 (95% CI 0.68, 3.25; p=0.32). Conclusions: There were no statistically significant differences in PSA failure free survival or overall survival detected in LDRe versus HDRe patient groups.