To identify influencing factors for recurrence and metastasis in patients with HER-2 positive breast cancer, and to establish a risk prediction model.

According to the inclusion and exclusion criteria, a total of 209 HER-2 positive breast cancer patients from four tertiary hospitals in Shandong Province between January 2021 and December 2024 were retrospectively enrolled as the training set, and 121 patients from two tertiary hospitals between January 2024 and May 2025 were included as the external validation set. The clinicopathological data of patients were collected and analyzed. Univariate analysis and LASSO regression were performed to screen for predictive factors, and multivariate logistic regression was adopted to determine the final variables for constructing a nomogram. Receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis (DCA) were used to evaluate the discrimination, calibration and clinical practicability of the model.

Seven predictive variables were finally screened out by multivariate logistic regression, including age, time since diagnosis, tumor size, family history, lymphedema, completion of targeted therapy and completion of chemotherapy. The results showed that time since diagnosis ≥1 year (OR=2.520, 95%CI: 1.017-6.245, P=0.046) and tumor diameter >5 cm (OR=2.998, 95%CI: 1.209-7.431, P=0.018) were risk factors for recurrence and metastasis in patients with HER-2 positive breast cancer; age ≥50 years (OR=0.366, 95%CI: 0.149-0.898, P=0.028) , completion of targeted therapy (OR=0.334, 95%CI: 0.128-0.872, P=0.025) and completion of chemotherapy (OR=0.028, 95%CI: 0.005-0.150, P=0.001) were protective factors. A nomogram was constructed based on the 5 statistically significant variables. In the training set, the area under the ROC curve (AUC) of the model was 0.865 (95%CI: 0.808-0.922) with a specificity of 0.963; the Hosmer-Lemeshow test yielded a good calibration (χ²=4.632, P=0.462). The AUC of the external validation set was 0.817 (95%CI: 0.715-0.919) with a specificity of 0.946 and the results of Hosmer-Lemeshow test showed a good calibration (χ²=6.833, P=0.555). DCA indicated stable and considerable clinical net benefit by applying this model within the risk threshold range of 0.2-0.8.

Patient age, time since diagnosis, tumor size, chemotherapy completion and targeted therapy completion are influencing factors for recurrence and metastasis in patients with HER-2 positive breast cancer. The prediction model established based on the above-mentioned factors can provide references for individualized clinical risk assessment.