Background: A recent study of 13,803 men found those treated with external beam radiation therapy (EBRT) had higher prostate cancer specific mortality compared to patients undergoing radical prostatectomy (RP) (Lee BH j.eururo.2014.09.017). We used the National Cancer Database to evaluate survival in patients with stage II prostate cancer treated with EBRT or RP. To our knowledge, our study is the largest to date evaluating these differences. Methods: We analyzed 455,106 men diagnosed with stage II prostate cancer from 2004-2013. Kaplan-Meier estimates and multivariable Cox regression analysis controlling for age, race, Charlson/Deyo (CD) score and PSA level were used to compare survival between treatments. CD score tabulates 0, 1, or 2 comorbidities. Results: We evaluated 169,185 EBRT and 285,921 RP patients. The average age of RP patients was younger than EBRT (60.56 years vs 69.42 years, p< .05) and the average PSA of RP patients was lower than EBRT (7.67 ng/mL vs 11.82 ng/mL, p< .05). White patients were more likely to receive RP than non-white patients (64.5% vs 35.5%, p< .05). 87.2% and 84.6% of EBRT and RP, respectively, had no comorbidities (CD = 0). Unadjusted mean survival was significantly longer for RP than EBRT (124.5 months vs 106.5 months, p< .05). Multivariable Cox regression indicated significantly longer survival for patients who were younger, white, had a lower Charlson/Deyo score and lower PSA. Multivariate analysis favored EBRT at 3 years (HR 2.12, p< .05), while RP provided significantly greater survival at 10 years (HR 0.11, p< .05). RP and CD 0 had the most impact on survival at 10 years. Conclusions: Men treated with RP had lower survival at three years, but significantly higher 10 year survival compared to EBRT. These results imply men treated with RP had higher early mortality but significantly better long term survival.

Note. *p<.05. HR = hazard ratio; HR below 1 indicates longer survival. ^aReference CD = 2. ^bCompared to radiation treatment specifically at 10 years.