UCSF

Purpose

Only retrospective data are available for low-dose-rate (LDR) salvage prostate brachytherapy for local recurrence after external beam radiation therapy (EBRT). The primary objective of this prospective phase 2 trial (NCT00450411) was to evaluate late gastrointestinal and genitourinary adverse events (AEs) after salvage LDR brachytherapy.

Methods and materials

Eligible patients had low- or intermediate-risk prostate cancer before EBRT and biopsy-proven recurrence >30 months after EBRT, with prostate-specific antigen levels <10 ng/mL and no regional/distant disease. The primary endpoint was grade 3 or higher late treatment-related gastrointestinal or genitourinary AEs occurring 9 to 24 months after brachytherapy. These AEs were projected to be ≤10%, with ≥20% considered unacceptable. All events were graded with National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0. Multivariate analyses investigated associations of pretreatment or treatment variables with AEs.

Results

One hundred patients from 20 centers were registered from May 2007 to January 2014. The 92 analyzable patients had a median follow-up of 54 months (range, 4-97) and a median age of 70 years (interquartile range [IQR], 65-74). The initial Gleason score was 7 in 48% of patients. The median dose of EBRT was 74 Gy (IQR, 70-76) at a median interval of 85 months previously (IQR, 60-119). Only 16% had androgen deprivation at study entry. Twelve patients (14%) had late grade 3 gastrointestinal/genitourinary AEs, with no treatment-related grade 4 or 5 AEs. No pretreatment variable predicted late AEs, including prior EBRT dose and elapsed interval. Higher V100 (percentage of prostate enclosed by prescription isodose) predicted both occurrence of late AEs (odds ratio, 1.24; 95% confidence interval, 1.02-1.52; P = .03) and earlier time to first occurrence (hazard ratio, 1.18; 95% CI, 1.03-1.34; P = .02).

Conclusions

This prospective multicenter trial reports outcomes of salvage LDR brachytherapy for post-EBRT recurrence. The rate of late grade 3 AEs did not exceed the unacceptable threshold. The only factor predictive of late AEs was implant dosimetry reflected by V100. Efficacy outcomes will be reported at a minimum of 5-year follow-up.