Based on breast CEUS parameters combined with serum CA153, a nomogram model was constructed to predict the molecular classification of breastcancer

ObjectiveTo analyze the role of contrast-enhanced ultrasound (CEUS) parameters combined with serum tumor marker CA153 in the prediction of Breast Cancer (BC) molecular typing. MethodsFrom January 2020 to January 2023, 120 BC patients diagnosed in our hospital were studied. According to the pathological results, the patients were divided into Luminal and non-Luminal BC groups. Both groups underwent contrast-enhanced ultrasoun. The time-intensity curve (TIC) is obtained, and the relevant characteristic parameters are obtained, including peak intensity (PI), peak time (TTP), area under the curve (AUC), and mean transit time (MTT). Serum tumor marker CA153 was detected in both groups. Combined with CEUS characteristic parameters and serum CA153 of two groups of BC patients, a multiple Logistic regression model was constructed, and a nomogram prediction model was constructed based on the model. Calibration curve and receiver operating characteristic (ROC) curve were used to analyze the value of this model in the prediction of BC molecular classification. ResultsThere were no significant differences between Luminal BC patients and non-Luminal BC patients in clinical parameters and qualitative parameters of contrast-enhanced ultrasound, while there were statistical differences between quantitative parameters PI, AUC and serum tumor marker CA153. The AUC of the combined diagnosis of three parameters (PI, AUC and CA153) was significantly higher than that of the single index diagnosis group. The ROC curve AUC of BC molecular typing was predicted to be 0.94 based on the three-parameter nomogram, and the fitting of the actual curve and the ideal curve in the calibration curve was close. ConclusionsThe nomogram model based on breast contrast-enhanced ultrasound (CEUS) parameters combined with serum CA153 can effectively predict the molecular classification of BC.