Objective TP53 mutations are common in breast cancer.There is a lack of large-scale cohort studies investigating the TP53 gene landscape in breast cancer in the Chinese population.The predictive value of TP53 mutation for anti-cancer therapy in breast cancer is controversial.Therefore,we analyze the spectrum of TP53 mutations among 804 breast cancer patients in China and analyzed the relationships between TP53 mutations and their impact on the prognosis and efficacy of first-line trastuzumabbased therapy.Methods Target-region capture deep sequencing was used to analyze the whole exome and introns of the TP53 gene in circulating tumor DNA.We analyzed the spectrum of TP53 gene mutations in different breast cancer subtypes.We evaluated the prognostic and predictive values of TP53 mutation in our cohort of patients through a multivariate Cox proportional hazards model.We also validated our results in the MSK-IMPACT cohort.Results A total of 804 patients were ultimately analyzed,and 53.6%harbored a TP53 mutation,which was higher than that observed in the 35.9%of MSK-IMPACT cohort.The majority of the mutations were missense(55.7%)and 75.1%of the mutations clustered in exons 5-8,mostly spanning the DNA-binding domain of the protein.TP53 mutations were differentially distributed in different molecular subtypes(p<0.001).The triplenegative breast cancer(TNBC)and HER2-positive/hormone receptor(HR)-negative patients had a higher TP53 gene mutation frequency than luminal-type breast cancer(p<0.01).TP53 mutation located in exons 5-8 was significantly associated with poor disease-free survival(DFS)compared to TP53-wild patients.The median DFS for patients with TP53 point mutations versus that of patients with non-TP53 mutations was 28.5 months versus 40.6 months,respectively(hazard ratio=1.32,confidence interval(CI)95%,1.09-1.61,p=0.005).TP53 was an independent marker for poor overall survival(OS)in breast cancer patients of the MSK-IMPACT cohort.The median OS for patients with TP53 mutations versus that of patients without TP53 mutations was 104.3 months versus 265.0 months,respectively,in this cohort(hazard ratio=3.12,CI 95%,2.50-3.90,p<0.0001).The multivariate Cox regression analysis also indicated that TP53 was a predictive biomarker for poor OS in breast cancer patients.TP53 mutation status was not significantly associated with progression-free survival(PFS)in patients who received first-line trastuzumab-based therapy.Interestingly,we found that in the taxane-based treatment group patients with TP53 mutations had better PFS than patients without TP53 mutations,and the median PFS was 12.1 months versus 2.7 months,respectively(hazard ratio=0.27; 95%confidence interval(CI),0.09-0.97,p<0.001).However,in the nontaxane group the result was the opposite; patients with TP53 mutation had poor PFS compared with TP53 wild-type patients(median PFS was 3.9 months versus 9.5 months,respectively; hazard ratio=5.51,95%CI,1.66-18.22,p=0.005).Conclusions TP53 somatic mutations are common in Chinese breast cancer patients and are more frequently observed in HER2-positive/HR-negative patients and in those with the TNBC subtype.TP53 mutations could be a prognostic biomarker of OS in breast cancer patients,but only TP53 mutations located in exons 5-8 could be an independent prognostic marker for worse DFS.The impact of TP53 mutations on trastuzumab+taxanes and trastuzumab+nontaxanes chemotherapy was the opposite.