Objective:To predict the expression level of Ki-67 in breast cancer patients based on ultrasonic characteris-tics.Methods:The data of 172 patients with breast cancer admitted to the Affiliated Cancer Hospital of Nantong University were retrospectively analyzed,which 51 cases were younger than 50 years old and 121 cases older than 50 years old.Ki-67 high expression group(n=108)and low expression group(n=64)were divided into two groups according to pathological immuno-histochemical results.All patients underwent color Doppler ultrasonography before surgery.Clinical data of patients including age,family history and underlying diseases were collected to analyze the correlation between the ultrasonic characteristics of mass and Ki-67 expression level.The ultrasonic parameters with statistical significance between the two groups were included in the multivariate Logistic regression.Independent predictors of Ki-67 high expression of breast cancer were then obtained to create a clinical prediction model.The receiver operating characteristic(ROC)curve and De Long test was used to evaluate its performance.Results:There were statistically significant differences in the maximum tumor diameter,internal echo and Adler blood grading between the high expression group and the low expression group(P＜0.05),and the area under ROC curve of the combined prediction model of the three features reached 0.825.The AUC value of the combined model is statistically signifi-cant compared with that of the single feature model,suggesting that the prediction model was capable of distinguishing Ki-67 expression.Conclusion:The clinical prediction model based on ultrasound is effective in predicting the expression level of Ki-67 in breast cancer patients and has high clinical application value.
万方数据