Minyong Kang1, Jin-Woo Jung2, Jong Jin Oh3, Sangchul Lee3, Sung Kyu Hong3, Sang Eun Lee3, Seok-Soo Byun4. 1. Department of Urology, Seoul National University Hospital, Chongno-Ku, Seoul, Korea. 2. Department of Urology, International St. Mary's Hospital, Incheon, Korea. 3. Department of Urology, Seoul National University Bundang Hospital, Seongnam-si, Korea. 4. Department of Urology, Seoul National University Bundang Hospital, Seongnam-si, Korea. Electronic address: ssbyun@snubh.org.
Abstract
INTRODUCTION: The present study evaluated the incidence and prognostic value of second primary cancers in patients with prostate cancer (PCa) who had undergone radical prostatectomy (RP). MATERIALS AND METHODS: From 2003 to 2013, 1915 patients who had undergone RP were included in the present analysis. We calculated the propensity scores of various clinicopathologic factors and matched 298 patients with and without second primary cancers in a 1:1 ratio. To assess the baseline variables, we compared the descriptive statistics between the 2 groups. The postoperative biochemical recurrence (BCR)-free survival rates were calculated using the Kaplan-Meier method. Multivariate Cox regression analysis was performed to identify the independent predictors of BCR after RP. RESULTS: Overall, 159 patients with PCa (8.3%) who had undergone RP were diagnosed with second primary cancers. After adjusting the patient characteristics in the propensity score-matched analysis, no variables were significantly different between the 2 groups with 149 with and 149 without other primary cancers. Moreover, the BCR-free survival rates were not significantly associated with the incidence of a second primary malignancy or the time to diagnosis. In the multivariate Cox regression model, serum prostate-specific antigen (hazard ratio [HR], 1.04), extraprostatic extension (HR, 3.29), seminal vesicle invasion (SVI; HR, 2.85), and surgical margin positivity (HR, 4.11) remained as independent predictors for BCR. However, the presence of a second primary malignancy was not predictive for BCR. In patients with a second primary cancer, multivariate analysis identified SVI (HR, 10.38) and positive surgical margin (HR, 3.48) as significant predictors for BCR. CONCLUSIONS: Our results suggest that the presence of second primary malignancies might not affect BCR in patients with PCa who undergo RP.
INTRODUCTION: The present study evaluated the incidence and prognostic value of second primary cancers in patients with prostate cancer (PCa) who had undergone radical prostatectomy (RP). MATERIALS AND METHODS: From 2003 to 2013, 1915 patients who had undergone RP were included in the present analysis. We calculated the propensity scores of various clinicopathologic factors and matched 298 patients with and without second primary cancers in a 1:1 ratio. To assess the baseline variables, we compared the descriptive statistics between the 2 groups. The postoperative biochemical recurrence (BCR)-free survival rates were calculated using the Kaplan-Meier method. Multivariate Cox regression analysis was performed to identify the independent predictors of BCR after RP. RESULTS: Overall, 159 patients with PCa (8.3%) who had undergone RP were diagnosed with second primary cancers. After adjusting the patient characteristics in the propensity score-matched analysis, no variables were significantly different between the 2 groups with 149 with and 149 without other primary cancers. Moreover, the BCR-free survival rates were not significantly associated with the incidence of a second primary malignancy or the time to diagnosis. In the multivariate Cox regression model, serum prostate-specific antigen (hazard ratio [HR], 1.04), extraprostatic extension (HR, 3.29), seminal vesicle invasion (SVI; HR, 2.85), and surgical margin positivity (HR, 4.11) remained as independent predictors for BCR. However, the presence of a second primary malignancy was not predictive for BCR. In patients with a second primary cancer, multivariate analysis identified SVI (HR, 10.38) and positive surgical margin (HR, 3.48) as significant predictors for BCR. CONCLUSIONS: Our results suggest that the presence of second primary malignancies might not affect BCR in patients with PCa who undergo RP.