Comparative Effectiveness of the Brachytherapy Boost and Androgen Deprivation Therapy for African-American Men with Unfavorable Intermediate-Risk Prostate Cancer Treated With External Beam Radiotherapy.

Abstract

PURPOSE/OBJECTIVE(S)\nComparative effectiveness (CE) data for radiotherapy treatment options for African-American (AfA) men with unfavorable intermediate-risk prostate cancer (UIR-PCa) is limited as AfA are under-represented on clinical trials and few large-scale trials have been conducted in UIR-PCa, with most trials grouping these patients with favorable intermediate-risk or high-risk disease. CE data is needed as AfA men with UIR-PCa have survival outcomes close to those for high-risk disease. Using the National Cancer Database (NCDB), we examined whether 1) the addition of a brachytherapy boost (BT) to external beam radiotherapy (EBRT) ± androgen deprivation therapy (ADT) was associated with improved overall survival (OS) vs. EBRT ± ADT; 2) whether EBRT+BT without ADT was associated with better OS vs EBRT+ADT; and 3) whether the addition of ADT was associated with improved OS in patients treated with EBRT+BT.\n\n\nMATERIALS/METHODS\n5,817 AfA men with UIR-PCa diagnosed between 2004 and 2015 that received EBRT ± ADT or EBRT+BT ± ADT were identified. EBRT was delivered with conventional fractionation (≥ 72 Gy in 1.8-2.0 Gy/fraction) or moderate hypofractionation (2.4-3.2 Gy/fraction with a biologically equivalent dose > 120 Gy). EBRT+BT consisted of EBRT to 40-50.4 Gy plus high-dose rate or low-dose rate BT. OS was evaluated using multivariable analysis (MVA). Covariables included in the analysis included age, CDCI score, insurance status, educational attainment, income, treatment at an academic center, PSA, Gleason score, clinical T stage, and year of diagnosis. Inverse probability of treatment weighting was used to balance measured confounders. Unweighted- and weighted- MVA using Cox regression was used to compare OS hazard ratios (HR).\n\n\nRESULTS\nAfA men were stratified into four treatment groups: (i) EBRT without ADT (n\u202f=\u202f2505), (ii) EBRT\u202f+\u202fADT (n\u202f=\u202f2238), (iii) EBRT+BT without ADT (n\u202f=\u202f631), and (iv) EBRT+BT\u202f+\u202fADT (n\u202f=\u202f443). Median age 66 (range 40-88). Median follow-up was 4.4 years (range, 0-14 years). Relative to EBRT without ADT, EBRT+ADT (HR: 0.92, [95% Confidence Interval: 0.80-1.05], P\u202f=\u202f.23) was not associated with improved OS but EBRT+BT without ADT (HR: 0.67 [0.52-0.86], P\u202f=\u202f0.002) and EBRT+BT+ADT (HR: 0.63 [0.49-0.82], P\u202f=\u202f0.0006) were associated with improved OS on MVA. EBRT+BT without ADT was associated with improved OS vs. EBRT+ADT on MVA (HR: 0.73 [0.57-0.94], P\u202f=\u202f0.01). There was no difference between EBRT+BT without ADT vs. EBRT+BT+ADT (HR: 1.00 [0.71-1.42], P\u202f=\u202f0.96). Weight-adjusted MVA demonstrated that brachytherapy boost (HR: 0.68 [0.57-0.82], P\u202f=\u202f0.00006) correlated with improved OS, while ADT did not (HR: 0.94 [0.79-1.12], P\u202f=\u202f.48).\n\n\nCONCLUSION\nIn an NCDB cohort of AfA men with UIR-PCa, brachytherapy boost was associated with improved OS. EBRT+BT without ADT was associated with improved OS vs. EBRT+ADT. The addition of ADT to EBRT+BT was not associated with improved OS. These findings are concordant with data that inform NCCN guidelines recommending EBRT+BT as an option for men with UIR-PCa.