Background: The current NCCN Clinical Practice Guidelines in Oncology for Prostate Cancer recommend long-term androgen deprivation therapy (ADT) for all men with high-risk prostate cancer treated with external-beam radiation therapy (EBRT). We determined whether the use of long-term ADT varied by the recently defined subcategories of high-risk disease (favorable, other, and very high) versus unfavorable intermediate-risk disease. Methods: We identified 5,524 patients with unfavorable-risk prostate cancer diagnosed from 2004 to 2007 and managed with EBRT using the SEER-Medicare linked database. Patients were stratified by risk group: unfavorable intermediate-risk, favorable high-risk (previously defined and validated as clinical stage T1c, Gleason score of 4 + 4 = 8, and prostate-specific antigen [PSA] level <10 ng/mL, or clinical stage T1c, Gleason score of 6, and PSA level >20 ng/mL), very-high-risk (clinical stage T3b–T4 or primary Gleason pattern 5), or other high risk (ie, neither favorable nor very high). We used multivariable competing risks regression to estimate the rates of long-term (≥2 years) ADT by group. Results: Men with favorable high-risk prostate cancer were significantly less likely to receive long-term ADT than those with other high-risk disease (15.4% vs 24.6%, adjusted hazard ratio [AHR], 0.68; 95% CI, 0.60–0.76; P<.001), and similarly likely as those with unfavorable intermediate-risk disease (AHR, 1.10; 95% CI, 0.99–1.23; P=.087). Other high-risk disease was less likely to receive long-term ADT than very high-risk cancer (24.6% vs 30.8%; AHR, 0.83; 95% CI, 0.74–0.93; P=.002). Conclusions: Despite current guidelines, patients with EBRT-managed high-risk prostate cancer received significantly different rates of long-course ADT based on subclassification. Our results suggest that oncologists view these patients as a heterogeneous group with favorable high-risk cancer warranting less aggressive therapy than other high-risk or very high-risk disease.