Background: Disparities in incidence and outcomes of prostate cancer among racial groups have been reported. Black men have a higher incidence of prostate cancer compared to White men and are more likely to be diagnosed at a later stage, resulting in poorer outcomes. These differences are likely due to a combination of genetic, socioeconomic, and healthcare access factors. Furthermore, evidence has revealed racial differences in molecular profiles of prostate cancer, such as a higher incidence of ERF mutations and increased transcriptional activation of the androgen receptor signaling and inflammatory pathways in tumors derived from Black men compared to White men, suggesting that underlying biology may contribute to these disparities. Here, we aimed to identify differences in the mutational profiles of prostate cancers by race and genetic ancestry in a de-identified real-world cohort of 3,804 patients that underwent the Tempus xT 648-gene NGS tumor profiling test. Methods: We used 654 ancestry-informative markers to estimate continental ancestry likelihoods for five regions: Africa (AFR), Americas (AMR), East Asia (EAS), Europe (EUR), and South Asia (SAS). We imputed Asian (ASN, n=122), Hispanic/Latino/Native American (HLN, n=300), Non-Hispanic Black (NHB, n=619) and Non-Hispanic White (NHW, n=2,679) categories using genetic ancestry thresholds based on literature and validated with available metadata. Associations between either race/ethnicity categories or continental ancestry proportions and somatic variants (copy number alterations (CNAs), small nonsynonymous, OncoKB L1/2 & R1, and predicted driver somatic mutations) were assessed for 109 prostate cancer-related genes using logistic regression models. Likelihood ratio tests were used to compare models with and without race/ethnicity or ancestry terms, and the Benjamini-Hochberg method was used to adjust p-values and control the FDR at 5%. Results: NHB and ASN men had lower odds of having a PTEN CNA compared to NHW men (OR=0.58 and 0.40, p=0.0001 and 0.009, respectively). Confirming this finding was a positive association between EUR ancestry and PTEN CNAs (OR=1.07 per doubling of EUR ancestry proportion, p=0.0007). NHB men were also less likely to have actionable OncoKB somatic mutations in the PIK3CA gene when compared to NHW men (OR=0.48, p=0.03). Increased AFR ancestry was associated with the presence of nonsynonymous predicted driver mutations in SPOP (OR=1.04, p=0.04), while both EAS and SAS ancestries were associated with fewer such mutations in TP53 (OR=0.95 in both cases, p=0.02 and 0.003). Conclusions: Analysis of a large real-world cohort leveraging genetic ancestry inferred from NGS data confirms previously known associations between the presence of somatic alterations in prostate cancer genes and race/ethnicity, and identifies novel associations of potential precision therapeutic relevance.