Single 15 Gy High-Dose-Rate Brachytherapy Boost for Treatment of Prostate Cancer: Clinical Implications of Dosimetric Parameters

Abstract

We previously reported our observations on toxicities and clinical outcomes in patients treated with an inversely planned single fraction High Dose Rate Brachytherapy (HDR-BT) boost followed by hypofractionated External Beam Radiotherapy (EBRT). More recently, results with longer follow-up became available, revealing more information about long-term toxicities and clinical failures. This report aims to determine an optimal set of dosimetric constraints based on our experience. A Phase I-II clinical trial protocol evaluating single-fraction HDR-BT combined with EBRT was undertaken in our institution. Eligible patients had clinically localized intermediate-risk prostate cancer. The patients received HDR-BT as a single 15-Gy implant, followed by EBRT to 37.5 Gy in 15 fractions. The patients were monitored prospectively for toxicity (Common Terminology Criteria for Adverse Events, version 3.0) and health-related quality of life (Expanded Prostate Cancer Index Composite [EPIC] and post-treatment PSA obtained q3 monthly for the first year, q6 monthly for 4 years, and annually thereafter. Kaplan-Meier biochemical Disease Free Survival (bDFS) curves were generated. Univariate and multivariate logistic regression analysis were used to investigate associations between clinical/functional outcomes and baseline covariates. A ROC curve analysis was used to determine threshold values predictive for toxicities. Median follow-up was 73 months. Five years bDFS was 97.4% (95% confidence interval 94.5-100%). The Mean V100 was 96.5%(interquartile range [IQR], 95.1-98.3%). The mean D90 was 101.5%(IQR, 100.1-104.4%). There was no association between either V100, or D90 and bDFS in our cohort (p=0.1657 and 0.1659). However patients with lower V100 were more likely to have malignant biopsy results on univariate and multivariate analysis (median V100=94.8% for patients having malignant cell on post treatment biopsies vs. 96.9% for patients with no malignant cells on post treatment biopsies; p=0.0271). On univariate analysis D90≥100% was associated with achieving a nadir PSA <0.1 ng/mL with a trend toward significance (p=0.084;OR=2.140; 0.912, 5.194). From our previous Toxicity and QOL analysis, we identified V200 with a threshold of 11%, Urethral D10 with a threshold of 120% and rectal V80 with a threshold of 0.9cc as predictors of worse functional outcomes. We have been able to identify dosimetric constraints for use in HDR planning to optimize clinical outcome. Achieving a V100 ≥95% and a D90≥100% while keeping the V200 < 11%, Urethral D10 < 120% and Rectal V80 < 0.9 cc is likely to ensure a balance between therapeutic efficacy and toxicity.