Abstract
BACKGROUND: Ductal carcinoma in situ (DCIS) cases diagnosed by ultrasound-guided core needle biopsy (US-CNB) carry a risk of postoperative upstaging to invasive breast carcinoma, complicating clinical management. This study aimed to investigate clinicopathological and ultrasound (US) predictors for postoperative upstaging and to develop a nomogram for individualized risk prediction. METHODS: A total of 240 patients with 246 DCIS lesions diagnosed by US-CNB were enrolled in this retrospective study from May 2013 to January 2025. Clinicopathological and US features were compared using the Student's t-test for continuous variables and the Chi-squared or Fisher's exact test for categorical variables. Multivariate logistic regression identified predictors of upstaging to DCIS with invasive components (DCIS-IC). A nomogram was developed and internally validated. Discrimination was assessed using the area under the receiver operating characteristic curve (AUC-ROC) with 1,000 bootstrap replicates. Calibration was evaluated through calibration curves and the Hosmer-Lemeshow (H-L) test, and clinical utility was examined using decision curve analysis (DCA). RESULTS: Among all the lesions, 161 (65.4%) were diagnosed as pure DCIS, while 85 (34.6%) were upstaged to DCIS-IC, including 37 (15.0% of total) with microinvasive carcinoma. Age [per 1-year increase, odds ratio (OR) =1.04; 95% confidence interval (CI): 1.01-1.06; P=0.01], Ki-67 >20% (OR =2.56; 95% CI: 1.35-4.86; P=0.004), and suspicious axillary lymph node (ALN) on US (OR =3.00; 95% CI: 1.07-8.45; P=0.04) were independent predictors of postoperative upstaging to DCIS-IC. The nomogram showed moderate discrimination with an apparent area under the curve (AUC) of 0.72 (95% CI: 0.65-0.78), which was internally validated as 0.70 (95% CI: 0.66-0.72) using 1,000 bootstrap replicates. It demonstrated good calibration (H-L test, P=0.86). The DCA showed that the nomogram provided net benefit across a threshold probability range of 20% to 88% compared to default strategies. Although larger tumor size (>1 cm; P=0.02) and non-circumscribed mass margins (P=0.03) were associated with upstaging in univariate analysis, they were not retained as independent predictors in the multivariate model. CONCLUSIONS: The nomogram incorporating age, Ki-67, and suspicious ALN on US effectively predicts DCIS upstaging risk in cases diagnosed by US-CNB and may assist in clinical decision-making. US characteristics (size >1 cm, non-circumscribed mass margins) may provide supplementary information but require further validation.