Ductal carcinoma in situ: a risk prediction model for the underestimation of invasive breast cancer

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.

Machine learning and new insights for breast cancer diagnosis

Breast cancer (BC) is the most prominent form of cancer among females all over the world. The current methods of BC detection include X-ray mammography, ultrasound, computed tomography, magnetic resonance imaging, positron emission tomography and breast thermographic techniques. More recently, machine learning (ML) tools have been increasingly employed in diagnostic medicine for its high efficiency in detection and intervention. The subsequent imaging features and mathematical analyses can then be used to generate ML models, which stratify, differentiate and detect benign and malignant breast lesions. Given its marked advantages, radiomics is a frequently used tool in recent research and clinics. Artificial neural networks and deep learning (DL) are novel forms of ML that evaluate data using computer simulation of the human brain. DL directly processes unstructured information, such as images, sounds and language, and performs precise clinical image stratification, medical record analyses and tumour diagnosis. Herein, this review thoroughly summarizes prior investigations on the application of medical images for the detection and intervention of BC using radiomics, namely DL and ML. The aim was to provide guidance to scientists regarding the use of artificial intelligence and ML in research and the clinic.

Survival Outcomes Based on Axillary Surgery in Ductal Carcinoma In Situ: A Nationwide Study From the Korean Breast Cancer Society

PURPOSE

In total mastectomy (TM), sentinel lymph node biopsy (SLNB) is recommended but can be omitted for breast-conserving surgery (BCS) in patients with ductal carcinoma in situ (DCIS). However, concerns regarding SLNB-related complications and their impact on quality of life exist. Consequently, further research is required to evaluate the role of axillary surgeries, including SLNB, in the treatment of TM. We aimed to explore the clinicopathological factors and outcomes associated with axillary surgery in patients with a final diagnosis of pure DCIS who underwent BCS or TM.

METHODS

We retrospectively analyzed large-scale data from the Korean Breast Cancer Society registration database, highlighting on patients diagnosed with pure DCIS who underwent surgery and were categorized into two groups: BCS and TM. Patients were further categorized into surgery and non-surgery groups according to their axillary surgery status. The analysis compared clinicopathological factors and outcomes according to axillary surgery status between the BCS and TM groups.

RESULTS

Among 18,196 patients who underwent surgery for DCIS between 1981 and 2022, 11,872 underwent BCS and 6,324 underwent TM. Both groups leaned towards axillary surgery more frequently for large tumors. In the BCS group, clinical lymph node status was associated with axillary surgery (odds ratio, 11.101; p = 0.003). However, in the TM group, no significant differences in these factors were observed. Survival rates did not vary between groups according to axillary surgery performance.

CONCLUSION

The decision to perform axillary surgery in patients with a final diagnosis of pure DCIS does not affect the prognosis, regardless of the breast surgical method. Furthermore, regardless of the breast surgical method, axillary surgery, including SLNB, should be considered for high-risk patients, such as those with large tumors. This may reduce unnecessary axillary surgery and enhance the patients' quality of life.

Prediction of Surgical Upstaging Risk of Ductal Carcinoma In Situ Using Machine Learning Models

Objective

The purpose of this study was to build machine learning models to predict surgical upstaging risk of ductal carcinoma in situ (DCIS) to invasive cancer and to compare model performance to eligibility criteria used by the Comparison of Operative versus Monitoring and Endocrine Therapy (COMET) active surveillance trial.

Methods

Medical records were retrospectively reviewed of all women with DCIS at core-needle biopsy who underwent surgery from 2007 to 2016 at an academic medical center. Multivariable regression and machine learning models were developed to evaluate upstaging-related features and their performance was compared with that achieved using the COMET trial eligibility criteria.

Results

Of 1387 women (mean age, 57 years; range, 27-89 years), the upstaging rate of DCIS was 17% (235/1387). On multivariable analysis, upstaging-associated features were presentation of DCIS as a palpable area of concern, imaging finding of a mass, and nuclear grades 2 or 3 at biopsy (P < 0.05). If COMET trial eligibility criteria were applied to our study cohort, then 496 women (42%, 496/1175) would have been eligible for the trial, with an upstaging rate of 12% (61/496). Of the machine learning models, none had a significantly lower upstaging rate than 12%. However, if using the models to determine eligibility, then a significantly larger proportion of women (56%-87%) would have been eligible for active surveillance.

Conclusion

Use of machine learning models to determine eligibility for the COMET trial identified a larger proportion of women eligible for surveillance compared with current eligibility criteria while maintaining similar upstaging rates.

Unveiling the mechanisms underlying photothermal efficiency of gold shell-isolated nanoparticles (AuSHINs) on ductal mammary carcinoma cells (BT-474)

Gold nanoparticles are valuable photothermal agents owing to their efficient photothermal conversion, photobleaching resistance, and potential surface functionalization. Herein, we combined bioinspired membranes with in vitro assays to elicit the molecular mechanisms of gold shell-isolated nanoparticles (AuSHINs) on ductal mammary carcinoma cells (BT-474). Langmuir and Langmuir-Schaefer (LS) films were handled to build biomembranes from BT-474 lipid extract. AuSHINs incorporation led to surface pressure-area (π-A) isotherms expansion, increasing membrane flexibility. Fourier-transform infrared spectroscopy (FTIR) of LS multilayers revealed electrostatic AuSHINs interaction with head portions of BT-474 lipid extract, causing lipid chain disorganization. Limited AuSHINs insertion into monolayer contributed to hydroperoxidation of the unsaturated lipids upon irradiation, consistently with the surface area increments of ca. 2.0%. In fact, membrane disruption of irradiated BT-474 cells containing AuSHINs was confirmed by confocal microscopy and LDH leakage, with greater damage at 2.2 × 1013 AuSHINs/mL. Furthermore, the decrease in nuclei dimensions indicates cell death through photoinduced damage.

Ductal Carcinoma In Situ (DCIS) and Microinvasive DCIS: Role of Surgery in Early Diagnosis of Breast Cancer

Advances in treatments, screening, and awareness have led to continually decreasing breast cancer-related mortality rates in the past decades. This achievement is coupled with early breast cancer diagnosis. Ductal carcinoma in situ (DCIS) and microinvasive breast cancer have increasingly been diagnosed in the context of mammographic screening. Clinical management of DCIS is heterogenous, and the clinical significance of microinvasion in DCIS remains elusive, although microinvasive DCIS (DCIS-Mi) is distinct from "pure" DCIS. Upfront surgery has a fundamental role in the overall treatment of these breast diseases. The growing number of screen-detected DCIS diagnoses with clinicopathological features of low risk for local recurrence (LR) allows more conservative surgical options, followed by personalised adjuvant radiotherapy plans. Furthermore, studies are underway to evaluate the validity of surgery omission in selected low-risk categories. Nevertheless, the management, the priority of axillary surgical staging, and the prognosis of DCIS-Mi remain the subject of debate, demonstrating how the paucity of data still necessitates adequate studies to provide conclusive guidelines. The current scientific scenario for DCIS and DCIS-Mi surgical approach consists of highly controversial and diversified sources, which this narrative review will delineate and clarify.

Crossroads: the role of biomarkers in the management of lumps in the breast

Premalignant lesions in the breast pose a difficult decision-making problem, whether to treat proactively and accept the side effects or to engage in watchful waiting and possibly encounter a later diagnosis of invasive cancer. A biomarker or set of biomarkers to inform on the individual progression risk would be beneficial to the patient and cost-effective for the healthcare system. The gene products of tumor progression may be expressed in early non-cancerous ("premalignant") lesions, where they are associated with a high probability for full transformation in breast cancers. One such molecule is the OPN splice variant-c. OPN-c is also present in a fraction of the premalignant lesions, where it reflects an elevated risk for progression to cancer within 5 years, regardless of the lesion's subtype. This marker has the properties needed to facilitate decisions to treat at the premalignant stage.

Breast cancer vaccines: New insights into immunomodulatory and nano-therapeutic approaches

Breast cancer (BC) is known to be a highly heterogeneous disease that is clinically subdivided into four primary molecular subtypes, each having distinct morphology and clinical implications. These subtypes are principally defined by hormone receptors and other proteins involved (or not involved) in BC development. BC therapeutic vaccines [including peptide-based vaccines, protein-based vaccines, nucleic acid-based vaccines (DNA/RNA vaccines), bacterial/viral-based vaccines, and different immune cell-based vaccines] have emerged as an appealing class of cancer immunotherapeutics when used alone or combined with other immunotherapies. Employing the immune system to eliminate BC cells is a novel therapeutic modality. The benefit of active immunotherapies is that they develop protection against neoplastic tissue and readjust the immune system to an anti-tumor monitoring state. Such immunovaccines have not yet shown effectiveness for BC treatment in clinical trials. In recent years, nanomedicines have opened new windows to increase the effectiveness of vaccinations to treat BC. In this context, some nanoplatforms have been designed to efficiently deliver molecular, cellular, or subcellular vaccines to BC cells, increasing the efficacy and persistence of anti-tumor immunity while minimizing undesirable side effects. Immunostimulatory nano-adjuvants, liposomal-based vaccines, polymeric vaccines, virus-like particles, lipid/calcium/phosphate nanoparticles, chitosan-derived nanostructures, porous silicon microparticles, and selenium nanoparticles are among the newly designed nanostructures that have been used to facilitate antigen internalization and presentation by antigen-presenting cells, increase antigen stability, enhance vaccine antigenicity and remedial effectivity, promote antigen escape from the endosome, improve cytotoxic T lymphocyte responses, and produce humoral immune responses in BC cells. Here, we summarized the existing subtypes of BC and shed light on immunomodulatory and nano-therapeutic strategies for BC vaccination. Finally, we reviewed ongoing clinical trials on BC vaccination and highlighted near-term opportunities for moving forward.