Ductal Carcinoma In Situ of Breast: From Molecular Etiology to Therapeutic Management

Abstract

Ductal Carcinoma in Situ (DCIS) makes up a majority of the non-invasive breast cancer cases. DCIS is a neoplastic proliferation of epithelial cells within the ductal structure of the breast. Currently, there is little known about the progression of DCIS to invasive ductal carcinoma (IDC), or the molecular etiology behind each DCIS lesion or grade. The DCIS lesions can be heterogeneous in morphology, genetics, cellular biology, and clinical behavior, posing challenges to our understanding of the molecular mechanisms for progression into an invasive status by approximately half of all DCIS lesions. New strategies that target molecular mechanisms are necessary to overcome this gap in understanding, which is a barrier to more targeted therapy. In this review, we will discuss the etiological factors associated with DCIS, as well as the complexity of each nuclear grade lesion. Moreover, we will discuss the possible molecular features that lead to progression of DCIS to IDC. We will highlight current therapeutic management and areas for improvement.

Northwestern University resource and education development initiatives to advance collaborative artificial intelligence across the learning health system

“…For example, we have collaborated with NMEDW team to deposit results from the NLP pipelines into data marts for breast cancer patients (deployment completed) 38 and cardiovascular disease patients (deployment in progress). The breast cancer data mart has enabled Northwestern investigators to secure federal grants and advance clinical research in breast cancer recurrence adjudication and prediction, 26‐29 genetic risk stratification 39‐41 and intervention, and drug delivery assessment 42‐44 . We have adapted an NLP pipeline for cardiovascular disease patients, successfully extracting key information like left ventricular ejection fraction from echocardiography reports across our adult care network and external institutions 45 .…”

“…The breast cancer data mart has enabled Northwestern investigators to secure federal grants and advance clinical research in breast cancer recurrence adjudication and prediction, 26 , 27 , 28 , 29 genetic risk stratification 39 , 40 , 41 and intervention, and drug delivery assessment. 42 , 43 , 44 We have adapted an NLP pipeline for cardiovascular disease patients, successfully extracting key information like left ventricular ejection fraction from echocardiography reports across our adult care network and external institutions. 45 This also lays the groundwork for downstream tasks like drug repurposing for atrial fibrillation.…”

MurSS: A Multi-Resolution Selective Segmentation Model for Breast Cancer

“…Breast cancer is a prevalent cancer among women worldwide [1]. In the United States, breast cancer is the second leading cause of cancer-related deaths in women [2]. Breast cancer is diagnosed by analyzing pathological features such as tumor growth patterns and cytologic characteristics.…”

Predicting Breast Cancer Events in Ductal Carcinoma In Situ (DCIS) Using Generative Adversarial Network Augmented Deep Learning Model

“…Ductal carcinoma in situ (DCIS) accounts for 20–25% of the cases of breast cancer [ 1 , 2 ] and is estimated to have afflicted over 1 million US women in 2022. Patients with DCIS are at risk of developing recurrent DCIS or invasive carcinoma.…”

“…mRNA-based models have been described [ 9 , 10 , 11 ] but have not been considered to be cost-effective [ 12 ]. Thus, there is an unmet clinical need for novel tools to improve BCEs risk stratification in DCIS [ 1 , 2 , 13 ].…”