Breast cancer screening for women at high risk: review of current guidelines from leading specialty societies

Prediction of mammographic breast density based on clinical breast ultrasound images using deep learning: a retrospective analysis

Background

Breast density, as derived from mammographic images and defined by the Breast Imaging Reporting & Data System (BI-RADS), is one of the strongest risk factors for breast cancer. Breast ultrasound is an alternative breast cancer screening modality, particularly useful in low-resource, rural contexts. To date, breast ultrasound has not been used to inform risk models that need breast density. The purpose of this study is to explore the use of artificial intelligence (AI) to predict BI-RADS breast density category from clinical breast ultrasound imaging.

Methods

We compared deep learning methods for predicting breast density directly from breast ultrasound imaging, as well as machine learning models from breast ultrasound image gray-level histograms alone. The use of AI-derived breast ultrasound breast density as a breast cancer risk factor was compared to clinical BI-RADS breast density. Retrospective (2009-2022) breast ultrasound data were split by individual into 70/20/10% groups for training, validation, and held-out testing for reporting results.

Findings

405,120 clinical breast ultrasound images from 14,066 women (mean age 53 years, range 18-99 years) with clinical breast ultrasound exams were retrospectively selected for inclusion from three institutions: 10,393 training (302,574 images), 2593 validation (69,842), and 1074 testing (28,616). The AI model achieves AUROC 0.854 in breast density classification and statistically significantly outperforms all image statistic-based methods. In an existing clinical 5-year breast cancer risk model, breast ultrasound AI and clinical breast density predict 5-year breast cancer risk with 0.606 and 0.599 AUROC (DeLong's test p-value: 0.67), respectively.

Interpretation

BI-RADS breast density can be estimated from breast ultrasound imaging with high accuracy. The AI model provided superior estimates to other machine learning approaches. Furthermore, we demonstrate that age-adjusted, AI-derived breast ultrasound breast density provides similar predictive power to mammographic breast density in our population. Estimated breast density from ultrasound may be useful in performing breast cancer risk assessment in areas where mammography may not be available.

Funding

National Cancer Institute.

[Analysis of Factors That Promote Awareness of Breast MRI Surveillance for Carriers of Hereditary Breast Cancer Risk Genes ( BRCA1/2)]

PURPOSE

Hereditary breast and ovarian cancers (HBOC) carry a high risk of breast cancer, and detailed screening with contrast-enhanced breast MRI (breast MRI surveillance) is recommended. With the increase in the number of individuals diagnosed with HBOC, the demand for breast MRI surveillance is also rising. However, the current system is inadequate, with factors such as lack of knowledge and indifference among healthcare professionals, and insufficient understanding of breast MRI surveillance being cited. This study aims to investigate the knowledge of HBOC and the awareness of breast MRI surveillance among radiological technologists, and to analyze the factors that promote these practices.

METHODS

A web-based survey was conducted among radiological technologists at 1278 facilities with MRI installations.

RESULTS

Responses were obtained from 433 individuals. The knowledge of HBOC was insufficient, with 49.6% unaware that breast MRI surveillance is recommended. Factors promoting awareness included the amount of knowledge about HBOC, age, and the presence of MRI specialists and mammography screening specialists.

CONCLUSION

By enhancing the acquisition of knowledge about HBOC and raising awareness of breast MRI surveillance, it is expected that discussions towards building a robust system will deepen.

The diagnostic performance of ultrafast MRI to differentiate benign from malignant breast lesions: a systematic review and meta-analysis

OBJECTIVES

To assess the diagnostic performance of ultrafast magnetic resonance imaging (UF-DCE MRI) in differentiating benign from malignant breast lesions.

MATERIALS AND METHODS

A comprehensive search was conducted until September 1, 2023, in Medline, Embase, and Cochrane databases. Clinical studies evaluating the diagnostic performance of UF-DCE MRI in breast lesion stratification were screened and included in the meta-analysis. Pooled summary estimates for sensitivity, specificity, diagnostic odds ratio (DOR), and hierarchic summary operating characteristics (SROC) curves were pooled under the random-effects model. Publication bias and heterogeneity between studies were calculated.

RESULTS

A final set of 16 studies analyzing 2090 lesions met the inclusion criteria and were incorporated into the meta-analysis. Using UF-DCE MRI kinetic parameters, the pooled sensitivity, specificity, DOR, and area under the curve (AUC) for differentiating benign from malignant breast lesions were 83% (95% CI 79-88%), 77% (95% CI 72-83%), 18.9 (95% CI 13.7-26.2), and 0.876 (95% CI 0.83-0.887), respectively. We found no significant difference in diagnostic accuracy between the two main UF-DCE MRI kinetic parameters, maximum slope (MS) and time to enhancement (TTE). DOR and SROC exhibited low heterogeneity across the included studies. No evidence of publication bias was identified (p = 0.585).

CONCLUSIONS

UF-DCE MRI as a stand-alone technique has high accuracy in discriminating benign from malignant breast lesions.

CLINICAL RELEVANCE STATEMENT

UF-DCE MRI has the potential to obtain kinetic information and stratify breast lesions accurately while decreasing scan times, which may offer significant benefit to patients.

KEY POINTS

• Ultrafast breast MRI is a novel technique which captures kinetic information with very high temporal resolution. • The kinetic parameters of ultrafast breast MRI demonstrate a high level of accuracy in distinguishing between benign and malignant breast lesions. • There is no significant difference in accuracy between maximum slope and time to enhancement kinetic parameters.

Polygenic Risk Score (PRS) Combined with NGS Panel Testing Increases Accuracy in Hereditary Breast Cancer Risk Estimation

Breast cancer (BC) is the most prominent tumor type among women, accounting for 32% of newly diagnosed cancer cases. BC risk factors include inherited germline pathogenic gene variants and family history of disease. However, the etiology of the disease remains occult in most cases. Therefore, in the absence of high-risk factors, a polygenic basis has been suggested to contribute to susceptibility. This information is utilized to calculate the Polygenic Risk Score (PRS) which is indicative of BC risk. This study aimed to evaluate retrospectively the clinical usefulness of PRS integration in BC risk calculation, utilizing a group of patients who have already been diagnosed with BC. The study comprised 105 breast cancer patients with hereditary genetic analysis results obtained by NGS. The selection included all testing results: high-risk gene-positive, intermediate/low-risk gene-positive, and negative. PRS results were obtained from an external laboratory (Allelica). PRS-based BC risk was computed both with and without considering additional risk factors, including gene status and family history. A significantly different PRS percentile distribution consistent with higher BC risk was observed in our cohort compared to the general population. Higher PRS-based BC risks were detected in younger patients and in those with FH of cancers. Among patients with a pathogenic germline variant detected, reduced PRS values were observed, while the BC risk was mainly determined by a monogenic etiology. Upon comprehensive analysis encompassing FH, gene status, and PRS, it was determined that 41.90% (44/105) of the patients demonstrated an elevated susceptibility for BC. Moreover, 63.63% of the patients with FH of BC and without an inherited pathogenic genetic variant detected showed increased BC risk by incorporating the PRS result. Our results indicate a major utility of PRS calculation in women with FH in the absence of a monogenic etiology detected by NGS. By combining high-risk strategies, such as inherited disease analysis, with low-risk screening strategies, such as FH and PRS, breast cancer risk stratification can be improved. This would facilitate the development of more effective preventive measures and optimize the allocation of healthcare resources.

Ultrafast Dynamic Contrast-enhanced MRI of the Breast: How Is It Used?

Ultrafast dynamic contrast-enhanced (UF-DCE) MRI is a new approach to capture kinetic information in the very early post-contrast period with high temporal resolution while keeping reasonable spatial resolution. The detailed timing and shape of the upslope in the time–intensity curve are analyzed. New kinetic parameters obtained from UF-DCE MRI are useful in differentiating malignant from benign lesions and in evaluating prognostic markers of the breast cancers. Clinically, UF-DCE MRI contributes in identifying hypervascular lesions when the background parenchymal enhancement (BPE) is marked on conventional dynamic MRI. This review starts with the technical aspect of accelerated acquisition. Practical aspects of UF-DCE MRI include identification of target hypervascular lesions from marked BPE and diagnosis of malignant and benign lesions based on new kinetic parameters derived from UF-DCE MRI: maximum slope (MS), time to enhance (TTE), bolus arrival time (BAT), time interval between arterial and venous visualization (AVI), and empirical mathematical model (EMM). The parameters derived from UF-DCE MRI are compared in terms of their diagnostic performance and association with prognostic markers. Pitfalls of UF-DCE MRI in the clinical situation are also covered. Since UF-DCE MRI is an evolving technique, future prospects of UF-DCE MRI are discussed in detail by citing recent evidence. The topic covers prediction of treatment response, multiparametric approach using DWI-derived parameters, evaluation of tumor-related vessels, and application of artificial intelligence for UF-DCE MRI. Along with comprehensive literature review, illustrative clinical cases are used to understand the value of UF-DCE MRI.