The number of deaths caused by tumors in China is increasing year by year and has ranked the top of the list of causes of death. Local data show that lung, liver and intestinal cancers are still the three most common causes of tumor-related deaths. There is no doubt that the recent new research on “personalized” tumor animal models, i.e., xenograft (PDX) models derived from patients’ tumor tissues, can be a boon to tumor patients. Early detection, diagnosis and treatment Early detection, early diagnosis and early treatment are indeed the key points of cancer treatment, but the unsatisfactory treatment effect has been plaguing the clinic. Of course, in recent years, the technology to treat tumors has increased. However, studies have shown that 50% of colon cancer patients recur and metastasize after treatment, resulting in a high mortality rate that is truly heartbreaking! Despite surgical resection and systemic chemotherapy, some patients continue to deteriorate and are in unsatisfactory condition. New targeted therapies such as Cetuxim and Pani, which are widely used to treat metastatic colorectal cancer, sometimes do not work at all due to the different conditions of patients. Different tumors, different patients, when heterogeneity and polymorphism become obstacles in the treatment of tumors, we start to call for more individualized treatment plans, and treatment plans need to be “customized”. So how can we achieve individualized tumor treatment? PDX is a patient-derived xenograft tumor model, which is a patient tumor transplanted in immunodeficient mice, to a certain extent, the hematologic characteristics and necrosis around the transplanted tumor are less differentiated than the original tumor cells compared to the transplanted tumor in the patient. However, it is these differences between tumor tissue and the original tumor that reflect the heterogeneity and adaptability of the tumor after natural selection, and therefore provide important and reliable indicators of tumor growth in vivo, creating an experimental “in vivo laboratory” that is highly consistent with the patient’s body condition. As the tumor grows, the number of mice used for transplantation can be increased, and these mice can be used as a platform to test drug response and simulate the effect of different drugs on the tumor in vivo. Screening drugs using the PDX model allows the results to be used as an assessment of therapeutic suitability, resulting in more appropriate dosing regimens that are more specific to the patient and can also increase the success rate of treatment. It also reduces the harm caused to the patient’s body by using multiple drugs. In fact, individualized tumor therapy with tumor cell line transplantation has been investigated before. So what is the difference between PDX technology and the previous tumor cell line transplantation? Although both methods involve implanting tumors into experimental mice to study tumor development and other conditions. But tumor cell line xenotransplantation is a model established by screening human tumor cells in vitro (in a culture dish), establishing a stable cell line through passaged culture, and then injecting it into immunodeficient mice subcutaneously, under the kidney envelope, or in situ. However, it is clear that this now in vitro culture environment has caused some alterations in the tumor microenvironment with the absence of specific substances such as extracellular matrix and non-tumor cells, making the tumor homogeneous after transplantation and unable to show heterogeneity, histopathology or genetic characteristics of the tumor. There are many mutations in the cells of actual tumor patients, which are more favorable to tumor growth. While PDX model is established as direct transplantation, although the success rate is relatively lower, PDX model can better reflect the genetic diversity of the tumor, more realistically simulate the patient’s situation, and better predict the tumor’s response to different drugs. So what can we apply the PDX model to do specifically? The PDX model has actually been used by several institutions to study different tumors. In a recent article published in Nature, a joint study by Samsung Medical Center for Surgery in Korea and several universities, they developed PDX models for 241 colon cancer patients with a success rate of 62.2%. They studied these models for 3 years, revealing the relationship between clinical treatment outcomes and tumorigenesis, and found the relationship between the presence or absence of KRAS gene mutations in patients and sensitivity to EGFR-targeted drugs. In terms of drug treatment, the study found that when the PDX model was administered, relatively few stage III patients were able to be completely cured, and the cure rate could be independently predicted from tumorigenesis (p=0.034), and that mutations in TP53 were frequently detected in stage III patients. The Clinical Institute for Surgery at the Gothenburg Academy of Sciences in Sweden published an article reporting that they identified treatment options for patients with metastatic malignant melanoma through in vitro drug screening and treatment with the MEK inhibitor trametinib in the PDX model, and using the PDX model, they found that detection of BRAF gene mutations predicted the effect of MAPK-targeted therapy. In Cancer, Justin Stebbing, Keren Paz et al. also reported that clinical treatment outcomes did correlate with tumor transplantation outcomes in 16 patients observed, and they concluded that the PDX model could guide the treatment of rare tumors such as sarcomas. The PDX model can be used to predict the therapeutic effect of drugs on tumors, which provides important guidance and basis for the development of clinical individualized treatment plans, and will promote more research on the relationship between tumor drugs and genetic mutations, which can further realize individualized tumor treatment. We sincerely hope that such “personalization” can improve the quality of survival of patients!