LAUSR

Background Historical epidemiological data indicate that Native American patients may have worse prostate cancer (PCa) characteristics than Caucasian patients (CAP). To test for cancer-specific mortality (CSM) differences among Native American vs. CAP, the most contemporary version of the SEER database (Surveillance, Epidemiology, and End Results database [2004−2016]) was used. Methods Descriptives and time trend analyses focused on a combined cohort of 357,289 Caucasian and Native American PCa patients of all stages. After 1:4 propensity-score (PS) matching for stage, grade, and other patient characteristics, cumulative incidence plots, and competing-risks-regression-models (CRR) were used, with further stratification according to non-metastatic (T any N 0 M 0 ) vs. metastatic (T any N 1 and/or M 1 ) stages. Results Native American patients accounted for 1804 (0.5%) of the study cohort. Native American patients had higher PSA (8 ng/ml vs. 6.3 ng/ml), higher rate of D’Amico high-risk PCa (30.8 vs. 24.8%), higher rate of T3/T4-PCa (5.5 vs. 3.7%), higher rate of N1 stage (4.5 vs. 2.8%), and higher rate of M1 stage (7.5 vs. 3.9%, all p  < 0.001) than CAP. In T any N 0 M 0 patients after PS-matching, 10-year CSM was 5.7 vs. 6.2% in Native American vs. CAP. In T any N 1 and/or M 1 patients, 10-year CSM was 64.3 vs. 63.3% in Native American vs. CAP, (both p  = 0.8). In CRR, Native American race did not represent an independent predictor of CSM. Conclusions Native American patients have more unfavorable stage and grade at presentation. However, after adjustment for these characteristics, CSM in Native American patients is not higher than in CAP. In consequence, PCa prognosis does not differ between Native American and Caucasian race. Therefore, efforts should be made to diagnose PCa in Native Americans at an earlier and more favorable stage like in CAP.