Background Prostate cancer (PCa) shows racial disparity in clinical and genomic characteristics, and Asian patients with PCa often present with more aggressive phenotypes at diagnosis. The ability of TP53 to serve as a prognostic biomarker of PCa has been well studied in Western populations. However, no studies to date have examined the role of TP53 in the disparities of primary hormone-naïve prostate cancer (HNPC) between Chinese and Western populations. Methods We collected prostate tumors and matched normal tissues or blood samples to perform targeted next-generation sequencing of 94 Chinese primary localized HNPC samples, and correlated these genomic profiles with clinical outcomes. The OncoKB knowledge database was used to identify and classify actionable alterations. Results The aberrations of PTEN, CDK12, and SPOP in Chinese HNPC samples were similar to those in the Western samples. However, we demonstrated an association of a high frequency of TP53 alterations (21/94) with a relatively higher percentage of alterations in the Wnt signaling pathway (15/94) in Chinese HNPC. Additionally, we highlighted alterations of LRP1B as accounting for a high proportion of PCa and found more frequent alterations in CDH1 in Chinese PCa. Of these, only CDH1 alteration was associated with rapid biochemical recurrence (BCR). However, we verified that TP53 status was at the core of the genomic alteration landscape in Chinese HNPC with putative driver mutations because of the strong connections with other signaling pathways. The mutually exclusive relationship between alterations in TP53 and Wnt/ CTNNB1 further molecularly characterizes subsets of prostate cancers. Moreover, the alteration of KMT2C was more likely to co-occur with TP53 alteration, indicating a more aggressive phenotype of PCa, which was associated with sensitivity to treatment with poly ADT-ribose polymerase (PARP) inhibitors. Conclusions Detection of TP53 alterations has clinical utility for guiding precision cancer therapy for HNPC, especially in the Chinese population.