ABSTRACT
BACKGROUND: Other-cause mortality (OCM) can serve as a surrogate for access-to-care. The authors sought to compare prostate cancer-specific mortality (PCSM) in Black versus White men matched based on their calculated OCM risk. METHODS: The Surveillance, Epidemiology, and End Results (SEER) database was queried for Black and White men diagnosed with prostate cancer between 2004 to 2009, to collect long-term follow-up. A Cox regression was used to calculate the OCM risk using all available covariates. This calculated OCM risk was used to construct a 1:1 propensity score matched (PSM) cohort. Then, a competing-risks multivariable tested the impact of race on PCSM. RESULTS: A total of 94,363 patients were identified, with 19,398 Black men and 74,965 White men. The median (IQR) follow-up was 11.3 years (9.8-12.8). In the unmatched-cohort at 10-years, PCSM and OCM were 5.5% versus 3.5% and 13.8% versus 8.4% in non-Hispanic Black (NHB) versus non-Hispanic White (NHW) patients (all p < .0001). The standardized mean difference was <0.15 for all covariates, indicating a good match. In the matched cohort at 10-years, OCM was 13.6% and 10.0% in NHB versus NHW (p < .0001), whereas the PCSM was 5.3% versus 4.7% (p < .01). On competing-risks multivariable analysis on PCSM, Black men had a hazard ratio of 1.08 (95% confidence interval, 0.98-1.20) compared to White men with a p = .13. CONCLUSIONS: The results of this study showed similar PCSM in Black and White patients, when matched with their calculated OCM risk. This report is the first to indicate at a population-based level that race has no impact on PCSM. PLAIN LANGUAGE SUMMARY: Prostate cancer is a very common cancer among men and it is associated with health disparities that disproportionately impact Black men compared to White men. There is an on-going discussion of whether disparities between these two groups stem from genetic or environmental factors. This study sought to examine if matching based on overall health status, a proxy for the impact of social determinants of health, mitigated significant differences in outcomes. When matched using risk of death from any cause other than prostate cancer, Black and White men had no significant differences in prostate cancer death.
ABSTRACT
OBJECTIVE: The European POUT III randomized controlled trial provided level-one evidence that adjuvant platinum-based chemotherapy is the standard of care following nephroureterectomy (RNU) for locally invasive or node-positive upper tract urothelial carcinoma. We aim to assess this European randomized controlled trial's generalizability (external validity) to a North American cohort, using a nationwide database. MATERIALS AND METHODS: To compare trial patients with those seen in real-world practice, we simulated the trial inclusion criteria using data from the National Cancer Database (NCDB). We identified patients with histologically confirmed transitional cell carcinoma who underwent RNU. The available demographic characteristics of the NCDB cohort were compared with the POUT III trial cohort using Chi-squared test. RESULTS: The NCDB cohort (nâ¯=â¯3,380) had a significantly higher proportion of older patients (age ≥ 80: 23.5% vs. 5%), and more males (68% vs. 56.2%) than the POUT cohort (Table 1, both p < 0.001). Additionally, the rate of advanced nodal disease was higher in the NCDB (N1 9.6%, N2 9.3%) than in the POUT (N1 6%, N2 3%) cohort (p < 0.001). A more extensive lymph node dissection was performed in NCDB vs. POUT patients (node≥10 10.9% vs. 3%, p < 0.001). Sensitivity analysis removing all subjects with a Charlson Comorbidity Index > 0 did not change the significance of any results. CONCLUSIONS: While the primary disease stage was similar, the rate of advanced nodal disease was significantly higher in NCDB, which might be explained partially by the more extensive lymph node dissection performed in the latter. These differences warrant caution when applying the POUT III findings to North American patients.
