The Declining Use of Brachytherapy Boost for Prostate Cancer Despite Associated Survival Advantage

Abstract

Purpose: Dose-escalated external beam radiotherapy (DE-EBRT) has been demonstrated to improve outcomes in men with intermediate-high risk prostate cancer. Brachytherapy represents the optimal modality for conformal dose escalation. A recent randomized trial demonstrated an improved biochemical-failure free survival in men who received EBRT plus a brachytherapy boost (BB) compared to DE-EBRT. Materials and Methods: We queried the National Cancer Data Base from 2004-2013 and identified 113,719 men with non-metastatic, non-operative, intermediate-high risk prostate cancer who were treated with EBRT+BB (44-54 Gy of EBRT followed by a brachytherapy boost) or DE-EBRT (75.6-81 Gy of EBRT at 1.8-2.0 Gy per fraction). We performed univariate analysis of all available factors potentially predictive of receipt of treatment selection for intermediate risk and high risk patients. Significant factors were then entered into a multivariate logistic regression model and factors significant on multivariable analysis were used to generate a propensity score corresponding to likelihood of treatment selection. Likewise, parsimonious survival analysis was conducted both with and without inclusion of the propensity score. Cox- Proportional Hazards modelling was used for the multivariate component of survival analysis and log-rank statistics were used for the univariate portion. Results: Median follow-up was 54 months (IQR533-78). Utilization of BB declined significantly over time both in an absolute sense and in relation to use of DE-EBRT. For intermediate-risk patients utilization of BB decreased from 33.1% (n=1,742) in 2004 to 12.5% (n=766) in 2013, with a similar decrease for Gleason 3+4 and 4+3. For high-risk patients utilization of BB dropped from 27.6% (n=879) in 2004 to 10.8% (n=479) in 2013. On multivariate analysis, factors predictive for preferential use of EBRT+BB boost for both intermediate and high risk patients were: earlier year of diagnosis, younger age, Charlson-Deyo comorbidity score< 2, race other than white or black, private insurance, higher residential area median income, treatment at a non-academic facility, geographical location (South for both intermediate and high risk and West for high risk), higher facility volume, higher T stage (except 2C for high risk), PSA ≤ 10, Gleason score (7 for intermediate-risk and 7-8 for high-risk), and primary Gleason pattern of 4; for intermediate-risk higher alone: residential area graduation rate; and for high-risk alone: metropolitan residential setting, increased distance from facility to residence, and no receipt of hormonal therapy. For intermediate-risk patients, unadjusted 5 and 10-yr OS for DE-EBRT vs. EBRT+BB were 87.3% (95% CI = 86.9-87.7) vs. 91.6% (95% CI = 91.0-92.2) and 58.0% (95% CI=56.4-59.6) vs. 71.6% (95% CI = 69.4-73.8) respectively (p<0.0005). For high-risk patients, unadjusted 5 and 10-yr OS for DE-EBRT vs. EBRT+BB were 82.0% (95% CI = 81.4-82.6) vs. 88.7% (95% CI = 87.9-89.5) and 50.4% (95% CI = 48.6-52.2) vs. 63.0% (60.0-66.0) respectively (p<0.0005). On multivariate analysis of factors predictive for survival, this correlated to a HR of 0.71 (95% CI = 0.67-0.75, p<0.0005) for intermediate-risk patients and 0.73 (95% CI = 0.68-0.78, p<0.0005) for high-risk patients. With incorporation of propensity score into the model, HR50.73 (95% CI = 0.68-0.77, p<0.0005) for intermediate-risk patients and HR50.74 (95% CI = 0.69-0.79, p<0.0005) for high-risk patients. Conclusions: Using a hospital-based national data set, we have demonstrated a concerning decline in the utilization of brachytherapy boost for intermediate and high risk prostate cancer patients. This is especially worrisome as brachytherapy boost is associated with a survival advantage as compared to DE-EBRT on both multivariate analysis and propensity-adjusted multivariate analysis. Furthermore, we have described numerous factors associated with preferential use of brachytherapy boost.