Testing the Impact of Adjuvant Radiotherapy (aRT) After Radical Prostatectomy (RP) on Overall Mortality (OM) In Prostate Cancer Patients With Pathologically Node Positive Disease: A Nationwide Analysis
Abdollah F, Sood A, Dalela D, Keeley J, Trinh QD, Alanee S, Rogers C, Peabody J, and Menon M. Testing the impact of adjuvant radiotherapy (aRT) after radical prostatectomy (RP) on overall mortality (OM) in prostate cancer patients with pathologically node positive disease: A nationwide analysis. Eur Urol, Supplements 2018; 17(2):e1622.
Introduction & Objectives: Using institutional data, we have previously identified sub-groups of patients with pN1 prostate cancer, who can benefit from aRT plus adjuvant hormonal therapy (aHT) after RP. However, our data originated from tertiary care centers, and our endpoint was limited to cancer specific mortality. To assess the generalizability of our previous findings, we set to examine their validity in a nationwide database, utilizing OM as an endpoint. Materials & Methods: We identified a total of 4276 patients with pN1 disease, who were treated with RP followed by aHT ± aRT, between 2004 and 2013, within the National Cancer DataBase (NCDB). No patient with missing data were included in the study cohort. We stratified patients into five risk groups using pathological data, and based on previously established criteria - Group 1: men with 1-2 positive nodes and Gleason score (GS) <=6; Group 2: men with 1-2 positive nodes, GS=7, and stage <=pT3a with negative surgical margins; Group 3: men with 1-2 positive nodes, GS=7, and stage >=pT3b or positive surgical margins; Group 4: men with 3-4 positive nodes; and Group 5: men with >4 positive nodes. Univariable and multivariable analyses tested the relationship between aRT status and OM in each of the aforementioned groups. Results: Median (interquartile range) age, # of nodes removed, and # of positive nodes were 62 yrs (56-67), 9 nodes (5-15), and 1 node (1-2), respectively. Mean and median follow-up were 53.6, and 49.4 months, respectively. The percentage of men included in Groups 1 to 5 was respectively 0.6%, 14.6%, 60.1%, 14.2%, and 10.4%. On univariable analysis, aRT improved outcomes in Group 3 and Group 4 only. Specifically, in patients treated without aRT vs with aRT, the 5-year OM was 15.2% vs 11.4% (p=.01) in Group3, and 24.7% vs 14.1% (p<.001) in Group 4. In multivariable analysis adjusting to age and comorbidity, aRT decrease OM risk in Group 3 (hazard ratio [HR]: 0.78, p=.01), and Group 4 (HR: 0.37, p<.001), but not in the other groups. Conclusions: Our current findings validate our previously published results, and demonstrate that certain sub-groups of patients with pathologically proven node positive prostate cancer can benefit from maximizing local control by adding aRT to RP. Besides, our new findings generalize our results to a nationwide setting, and shows that the beneficial impact of aRT is evident also for OM, which is the most significant endpoint in cancer treatment.