A Competing Risk Analysis of Cancer-Specific Mortality of Initial Treatment with Radical Prostatectomy versus Radiation Therapy in Clinically Localized High-Risk Prostate Cancer

Joo Yong Lee, Kang Su Cho, Jong Kyou Kwon, Seong Uk Jeh, Ho Won Kang, Richilda Red Diaz, Won Sik Ham, Woong Sub Koom, Ki Chang Keum, Young Deuk Choi

Research output: Contribution to journalArticle

24 Citations (Scopus)


Background: There is no consensus on the optimal treatment for localized high-risk prostate cancer (PC), and much debate exists regarding the ideal treatment approach. For these reasons, we evaluated the competing risks of PC-specific mortality after initial therapy with radical prostatectomy (RP) versus radiotherapy (RT) in men with clinically localized high-risk PC.

Methods: We reviewed patients treated with RP and RT combined with androgen-deprivation therapy between 1990 and 2009. High-risk PC is defined as clinical stage ≥T3a, serum prostate-specific antigen (PSA) >20 ng/mL, or a biopsy Gleason sum of 8–10 according to National Comprehensive Cancer Network guidelines. Competing risk analysis was conducted to assess the association of RP (n = 251) or RT (n = 125) with cancer-specific mortality (CSM). Thereafter, secondary analysis with propensity score matching was conducted to further elucidate patient characteristics, with optimal matching of 0.25 times the standard deviation of propensity scores.

Results: With an overall median follow-up of 76 months, 35 (9.3 %) men with high-risk PC died due to PC (23 in the RT group and 12 in the RP group). The 5-year estimates of cancer-specific survival rate for men treated with RP and RT were 96.5 % (95 % confidence interval [CI] 94.2–98.9) and 88.3 % (95 % CI 82.8–94.3), respectively. Cumulative incidence estimates for CSM were statistically increased amongst men treated with RT (p = 0.002). Propensity score matching extracted 168 men with high-risk PC from the total patient cohort. Cumulative incidence estimates for CSM were statistically different amongst men treated with RT (p < 0.001).

Conclusions: Initial treatment with RP versus RT was associated with a decreased risk of CSM in men with clinically localized high-risk PC.

Original languageEnglish
Pages (from-to)4026-4033
Number of pages8
JournalAnnals of surgical oncology
Issue number12
Publication statusPublished - 2014 Oct 8


All Science Journal Classification (ASJC) codes

  • Surgery
  • Oncology

Cite this