关键词: 铁死亡, 基因表达, 乳腺癌, 新辅助化疗, 预测模型

Abstract: Objective To investigate the effect of Ferroptosis-related gene expression levels in breast cancer tissues on the treatment responsiveness of neoadjuvant chemotherapy(NAC),to construct a prediction model of chemotherapy responsiveness based on Ferroptosis-related genes and to evaluate its efficacy.Methods One hundred and forty breast cancer patients admitted to our hospital who received NAC from April 2021 to September 2023 were selected.All patients were collected by pathological puncture biopsy before chemotherapy,and the expression of Ferroptosis-related genes FSP1,ACSL4,GPX4,and p53 in breast cancer tissues was determined by reverse transcription-fluorescence quantitative PCR.After NAC,the chemotherapy reactivity of the patients was determined according to the Miller-Payne criteria,and they were divided into the histological significant response group (group R) and the histological non-significant response group (group NR) accordingly.The clinical data of the two groups were compared,and the related factors of low response to NAC were analyzed using the Logistic regression equation.The NAC reactivity model was constructed through the R language ggrisk software package,and the ROC curve was drawn to evaluate the model efficacy.Results Fifty-six patients out of 140 patients were evaluated for histologically significant response,and so there was 56 patients in R group,84 patients in NR group.comparison of clinical data showed no statistical difference between the two groups in terms of general clinical data and positive expression of ER,PR,HER-2 and Ki-67(P>0.05);in terms of Ferroptosis-related gene expression,FSP1,GPX4,ACSL4 and p53 gene relative expression in R group were 0.28±0.06,0.52±0.21,1.59±0.21,2.09±0.31 respectively;FSP1,GPX4,ACSL4 and p53 gene relative expression in NR group were 0.42±0.09、0.64±0.23、1.28±0.17、1.81±0.23 respectively,and the differences were statistically significant(P<0.05).Logistic regression analysis showed that FSP1(OR＜0.001),GPX4 gene(OR=0.008),ACSL4(OR=24 721.186),p53 gene(OR=225.671) were independently correlated with neoadjuvant chemotherapy responsiveness in breast cancer patients(P<0.05).ROC curve analysis showed that the predictive risk model constructed from the above genes had good predictive value in early assessment of NAC responsiveness(AUC=0.979,P＜0.001),and the sensitivity was 98.21% and the specificity was 89.29%.Conclusion The expression levels of Ferroptosis-related genes FSP1,ACSL4,GPX4,and p53 were correlated with the responsiveness of breast cancer NAC treatment,and the constructed prediction model based on the expression of Ferroptosis-related genes was able to assess the responsiveness of breast cancer NAC treatment at an early stage.

Key words: ferroptosis')">">ferroptosis, gene expression, breast cancer, neoadjuvant chemotherapy, prediction model