A new model using number of needles and androgen deprivation to predict chronic urinary toxicity for high or low dose rate prostate brachytherapy

Abstract

Purpose: Prostate brachytherapy is an established treatment modality in early stage prostate cancer. We retrospectively reviewed our experience with low dose rate (LDR) and high dose rate (HDR) brachytherapy as a single treatment modality for early prostate cancer with emphasis on chronic toxicity.

Materials and methods: From June 1996 to August 2003, 253 patients with stage II prostate cancer, prostate specific antigen less than 12 and Gleason score less than 7 were treated with brachytherapy alone at our institution. A total of 92 patients underwent HDR brachytherapy with 192Ir, while 161 underwent LDR brachytherapy with 103Pd. HDR minimum prostate dose was 38 Gy, delivered in 4 fractions with a single implant during 36 hours. For HDR we used real-time dynamic 3-dimensional ultrasound base dosimetry. For 103Pd seed implants the dose was 120 Gy using selective peripheral weighted dose distribution. Treatment was given based on patient preference after pretreatment transrectal ultrasound. Toxicity was scored using the National Cancer Institute Common Toxicity Criteria 2.0. Median followup in all 253 cases was 2.9 years.

Results: In all patients the rate of 3-year urinary toxicity grade 2 or greater and grade 3 or greater was 26% and 6.9%, which was not significantly different between HDR and LDR (p = 0.3 and 0.4, respectively). However, grade 1 urogenital toxicity was lower for HDR (p = 0.002). The 3-year grade 2 rectal toxicity rate was 0.8% with no grade 3 or greater events, which was and similar in the HDR and LDR groups (1% and 0.6%, respectively). No cancer related deaths occurred and 4-year overall survival was 99% for HDR and 96.4% for LDR (p = 0.4). The 3-year American Society for Therapeutic Radiology and Oncology biochemical control rate was 90% for LDR and 93% for HDR. Cox multivariate analysis for grade 2 or greater urinary toxicity was significant for the use of 14 or greater needles (HR 6.1, p = 0.02) and hormonal therapy (HR 2.2, p = 0.02). In the absence of risk factors the 4-year grade 2 or greater urinary toxicity rate was 7% vs 65% if the 2 risk factors were present (p <0.001). Impotence crude rates were 18.3% for HDR and 41.3% for LDR (p = 0.002).

Conclusions: HDR and LDR chronic urinary toxicity grade 2 or greater rates were equivalent. However, grade 1 was lower for HDR. The impotence rate was decrease by half with HDR. Neoadjuvant hormonal therapy and 14 or greater needles were significantly associated with increased chronic urinary toxicity on multivariate analysis.