Cancer is a series of heterogeneous diseases caused by mutations in oncogenes and tumor suppressor genes caused by physical, chemical, viral and other carcinogenic factors. Despite breakthroughs in high-throughput sequencing technology and targeted therapy in recent years, the high failure rate of clinical translational research has limited the innovative development of anti-tumor drugs. The patient derived xenograft model (PDX model) refers to a xenograft model that directly transplants fresh tumor tissue from a patient onto immunodeficient mice, relying on the environment provided by the mice for growth. This model preserves the stromal heterogeneity, histological characteristics, molecular diversity, and microenvironment of primary tumors, providing effective research and development resources for personalized screening and evaluation of preclinical efficacy and identification of biomarkers. The PDX model, combined with clinical data, genomic maps, and pharmacological data, can increase drug specificity and be applied to individualized treatment of cancer patients, improving the success rate of clinical treatment. This article provides a review of the research progress on PDX models, including their applications in new tumor therapies, challenges and limitations, as well as their potential applications in precision oncology medicine.