Mass-forming ductal carcinoma in situ: An ultrasonographic and histopathologic correlation study

Ultrasound characteristics comparison and development of a predictive nomogram for intraductal papilloma and ductal carcinoma in situ: a retrospective cohort study

Background

Intraductal Papilloma (IDP) and Ductal Carcinoma In Situ (DCIS) are significant benign and pre-invasive breast lesions, respectively. This study aimed to investigate ultrasound features and develop a predictive nomogram for discriminating between IDP and DCIS.

Methods

Conducted at Quanzhou First Hospital over a three-year period, 389 patients were enrolled with detailed ultrasound examinations and confirmed pathological diagnoses. IDP was classified into Grades 3, 4, and 5, whereas DCIS presented with a mass-like morphology. Patients meeting the inclusion criteria underwent rigorous analysis, with exclusion criteria eliminating those with incomplete imaging data or confounding comorbidities. Ultrasound characteristics, including lesion size, shape, margin, and echogenicity, etc., were systematically evaluated and compared between the two groups. Univariate and multivariate logistic regression analyses were conducted to identify significant risk factors. Subsequently, based on these characteristics, both static and dynamic nomograms were developed. The performance of the nomograms was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA).

Results

The study cohort included 272 patients in the training set and 117 in the validation set. Significant differences were observed between IDP and DCIS in age, size, shape, aspect ratio, margin, duct dilatation, and microcalcification (P < 0.05). Logistic regression analyses identified age, size, aspect ratio, margin, microcalcification, and duct dilatation as independent risk factors. Compared to DCIS, IDP is associated with younger age, smaller size, clearer margins, fewer microcalcifications, and more ductal dilation. The performance of the nomogram developed to predict IDP and DCIS showed an AUC of 0.918 in the training set and 0.888 in the validation set. The calibration curve indicates a strong fit of the predictive model in the validation set, with the Hosmer-Lemeshow test showing high consistency between predicted and actual probabilities (training set, P = 0.875; validation set, P = 0.751). Additionally, DCA confirms the clinical utility of the model.

Conclusion

The nomogram incorporating key predictors provides a valuable tool for differentiating between IDP and DCIS based on ultrasound characteristics. This approach aids in clinical decision-making and potentially reduces unnecessary biopsies.

Deep-learning model to improve histological grading and predict upstaging of atypical ductal hyperplasia / ductal carcinoma in situ on breast biopsy

AIMS

Risk stratification of atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS), diagnosed using breast biopsy, has great clinical significance. Clinical trials are currently exploring the possibility of active surveillance for low-risk lesions, whereas axillary lymph node staging may be considered during surgical planning for high-risk lesions. We aimed to develop a machine-learning algorithm based on whole-slide images of breast biopsy specimens and clinical information to predict the risk of upstaging to invasive breast cancer after wide excision.

METHODS AND RESULTS

Patients diagnosed with ADH/DCIS on breast biopsy were included in this study, comprising 592 (740 slides) and 141 (198 slides) patients in the development and independent testing cohorts, respectively. Histological grading of the lesions was independently evaluated by two pathologists. Clinical information, including biopsy method, lesion size, and Breast Imaging Reporting and Data System (BI-RADS) classification of ultrasound and mammograms, were collected. Deep DCIS consisted of three deep neural networks to evaluate nuclear grade, necrosis, and stromal reactivity. Deep DCIS output comprised five parameters: total patches, lesion extent, Deep Grade, Deep Necrosis, and Deep Stroma. Deep DCIS highly correlated with the pathologists' evaluations of both slide- and patient-level labels. All five parameters of Deep DCIS were significantly associated with upstaging to invasive carcinoma in subsequent wide excisional specimens. Using multivariate logistic regression, Deep DCIS predicted upstaging to invasive carcinoma with an area under the curve (AUC) of 0.81, outperforming pathologists' evaluation (AUC, 0.71 and 0.69). After including clinical and hormone receptor status information, performance further improved (AUC, 0.87). This combined model retained its predictive power in two subgroup analyses: the first subgroup included unequivocal DCIS (excluding cases of ADH and DCIS suspicious for microinvasion) (AUC, 0.83), while the second excluded cases of high-grade DCIS (AUC, 0.81). The model was validated in an independent testing cohort (AUC, 0.81).

CONCLUSION

This study demonstrated that deep-learning models can refine histological evaluation of ADH and DCIS on breast biopsies, which may help guide future treatment planning.