Factors Influencing Noncompletion of Radiation Therapy Among Men With Localized Prostate Cancer

Abstract

Purpose Treatment noncompletion may occur with radiation therapy (RT), especially with protracted treatment courses such as RT for prostate cancer, and may affect the efficacy of RT. For men with localized prostate cancer managed with primary RT, we evaluated associations between rates of treatment noncompletion and RT fractionation schedules. Methods and Materials The National Cancer Database identified men diagnosed from 2004 to 2014 treated with primary RT. Patients receiving 180 cGy/fraction or 200 cGy/fraction were defined as having completed radiation therapy if they received ≥41 fractions of 180 cGy/fraction or ≥37 fractions of 200 cGy/fraction. Stereotactic body radiation therapy (SBRT) was defined as 5 to 8 fractions of 600 to 800 cGy/fraction. Odds ratios compared rates of treatment noncompletion, adjusting for sociodemographic covariates. A propensity-adjusted multivariable Cox regression assessed the association between treatment completion and overall survival. Results Of 157,657 patients, 95.7% (n = 150,847) received conventional fractionation and 4.3% (n = 6810) received SBRT. Rates of noncompletion were 12.5% (n = 18,803) among patients who received conventional fractionation and 1.9% (n = 131) among patients who received SBRT (odds ratio [OR] versus conventional, 0.21; 95% confidence interval [CI], 0.18-0.26; P < .001). The rate of noncompletion among 25,727 African American patients was 12.8%, compared with 11.8% among 126,199 white patients (OR, 1.14; 95% CI, 1.09-1.19; P < .001). In a subgroup analysis, the disparity in noncompletion persisted for conventional fractionation (13.2% vs 12.3%, respectively; OR, 1.09; 95% CI, 1.05-1.13; P < .001), but not for SBRT (2.2% vs 1.8%, respectively; OR, 1.26; 95% CI, 0.79-2.00; P = .34). Noncompletion was associated with worse survival in a propensity-adjusted multivariable analysis (hazard ratio, 1.25; 95% CI, 1.22-1.29; P < .001). Conclusions SBRT was associated with lower rates of RT noncompletion among men with localized prostate cancer. African American race was associated with greater rates of treatment noncompletion, although the disparity may be decreased among men receiving SBRT. Treatment noncompletion may occur with radiation therapy (RT), especially with protracted treatment courses such as RT for prostate cancer, and may affect the efficacy of RT. For men with localized prostate cancer managed with primary RT, we evaluated associations between rates of treatment noncompletion and RT fractionation schedules. The National Cancer Database identified men diagnosed from 2004 to 2014 treated with primary RT. Patients receiving 180 cGy/fraction or 200 cGy/fraction were defined as having completed radiation therapy if they received ≥41 fractions of 180 cGy/fraction or ≥37 fractions of 200 cGy/fraction. Stereotactic body radiation therapy (SBRT) was defined as 5 to 8 fractions of 600 to 800 cGy/fraction. Odds ratios compared rates of treatment noncompletion, adjusting for sociodemographic covariates. A propensity-adjusted multivariable Cox regression assessed the association between treatment completion and overall survival. Of 157,657 patients, 95.7% (n = 150,847) received conventional fractionation and 4.3% (n = 6810) received SBRT. Rates of noncompletion were 12.5% (n = 18,803) among patients who received conventional fractionation and 1.9% (n = 131) among patients who received SBRT (odds ratio [OR] versus conventional, 0.21; 95% confidence interval [CI], 0.18-0.26; P < .001). The rate of noncompletion among 25,727 African American patients was 12.8%, compared with 11.8% among 126,199 white patients (OR, 1.14; 95% CI, 1.09-1.19; P < .001). In a subgroup analysis, the disparity in noncompletion persisted for conventional fractionation (13.2% vs 12.3%, respectively; OR, 1.09; 95% CI, 1.05-1.13; P < .001), but not for SBRT (2.2% vs 1.8%, respectively; OR, 1.26; 95% CI, 0.79-2.00; P = .34). Noncompletion was associated with worse survival in a propensity-adjusted multivariable analysis (hazard ratio, 1.25; 95% CI, 1.22-1.29; P < .001). SBRT was associated with lower rates of RT noncompletion among men with localized prostate cancer. African American race was associated with greater rates of treatment noncompletion, although the disparity may be decreased among men receiving SBRT.