Early Detection of Breast Cancer in MRI Using AI

RATIONALE AND OBJECTIVES

To develop and evaluate an AI algorithm that detects breast cancer in MRI scans up to one year before radiologists typically identify it, potentially enhancing early detection in high-risk women.

MATERIALS AND METHODS

A convolutional neural network (CNN) AI model, pre-trained on breast MRI data, was fine-tuned using a retrospective dataset of 3029 MRI scans from 910 patients. These contained 115 cancers that were diagnosed within one year of a negative MRI. The model aimed to identify these cancers, with the goal of predicting cancer development up to one year in advance. The network was fine-tuned and tested with 10-fold cross-validation. Mean age of patients was 52 years (range, 18-88 years), with average follow-up of 4.3 years (range 1-12 years).

RESULTS

The AI detected cancers one year earlier with an area under the ROC curve of 0.72 (0.67-0.76). Retrospective analysis by a radiologist of the top 10% highest risk MRIs as ranked by the AI could have increased early detection by up to 30%. (35/115, CI:22.2-39.7%, 30% sensitivity). A radiologist identified a visual correlate to biopsy-proven cancers in 83 of prior-year MRIs (83/115, CI: 62.1-79.4%). The AI algorithm identified the anatomic region where cancer would be detected in 66 cases (66/115, CI:47.8-66.5%); with both agreeing in 54 cases (54/115, CI:%37.5-56.4%).

CONCLUSION

This novel AI-aided re-evaluation of "benign" breasts shows promise for improving early breast cancer detection with MRI. As datasets grow and image quality improves, this approach is expected to become even more impactful.

Breast Imaging: what women & healthcare professionals need to know

Women in Radiology should be aware of the importance of early detection of breast cancer, the most common cancer in women. This knowledge is essential to advocate for high quality breast imaging for women, including themselves and their patients. The imaging modalities used in breast imaging have dramatically changed the way in which breast cancer may be diagnosed, and their use affects the stage at which it is diagnosed. Breast cancer may be screen-detected, either with mammography, digital breast tomosynthesis, breast ultrasound, breast MRI or contrast-enhanced mammography, and is typically diagnosed at stage 1. Incidental detection with Chest CT, abdominal CT or MRI or by PET CT may also lead to a diagnosis of breast cancer. When detected because of symptoms in women who have not undergone routine screening or as an interval cancer in women after a normal screen typically because of the masking effect of dense breast tissue, breast cancer is typically diagnosed at a more advanced stage, stage IIA or greater. A review of the imaging modalities currently used to diagnose breast cancer is provided and includes the advantages and limitations of each modality and the ways to optimize the imaging quality for detection of breast cancer. Up-to-date recommendations aimed to minimize the harms of delayed diagnosis of breast cancer are summarized to improve the health of women in Radiology and their patients.

Limited impact of adding digital breast tomosynthesis to full field digital mammography in an elevated breast cancer risk population

PURPOSE

To investigate the impact of adding digital breast tomosynthesis (DBT) to full field digital mammography (FFDM) in screening asymptomatic women with an elevated breast cancer life time risk (BCLTR) but without known genetic mutation.

METHODS

This IRB-approved single-institution multi-reader study on prospectively acquired FFDM + DBT images included 429 asymptomatic women (39-69y) with an elevated BC risk on their request form. The BCLTR was calculated for each patient using the IBISrisk calculator v8.0b. The screening protocol and reader study consisted of 4-view FFDM + DBT, which were read by four independent radiologists using the BI-RADS lexicon. Standard of care (SOC) included ultrasound (US) and magnetic resonance imaging (MRI) for women with > 30 % BCLTR. Breast cancer detection rate (BCDR), sensitivity and positive predictive value were assessed for FFDM and FFDM + DBT and detection outcomes were compared with McNemar-test.

RESULTS

In total 7/429 women in this clinically elevated breast cancer risk group were diagnosed with BC using SOC (BCDR 16.3/1000) of which 4 were detected with FFDM. Supplemental DBT did not detect additional cancers and BCDR was the same for FFDM vs FFDM + DBT (9.3/1000, McNemar p = 1). Moderate inter-reader agreement for diagnostic BI-RADS score was found for both study arms (ICC for FFDM and FFDM + DBT was 0.43, resp. 0.46).

CONCLUSION

In this single institution study, supplemental screening with DBT in addition to standard FFDM did not increase BCDR in this higher-than-average BC risk group, objectively documented using the IBISrisk calculator.

ACR Appropriateness Criteria® Female Breast Cancer Screening: 2023 Update

Early detection of breast cancer from regular screening substantially reduces breast cancer mortality and morbidity. Multiple different imaging modalities may be used to screen for breast cancer. Screening recommendations differ based on an individual's risk of developing breast cancer. Numerous factors contribute to breast cancer risk, which is frequently divided into three major categories: average, intermediate, and high risk. For patients assigned female at birth with native breast tissue, mammography and digital breast tomosynthesis are the recommended method for breast cancer screening in all risk categories. In addition to the recommendation of mammography and digital breast tomosynthesis in high-risk patients, screening with breast MRI is recommended. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

How Does Diagnostic Accuracy Evolve with Increased Breast MRI Experience?

Introduction: Our institution is part of a provincial program providing annual breast MRI screenings to high-risk women. We assessed how MRI experience, background parenchymal enhancement (BPE), and the amount of fibroglandular tissue (FGT) affect the biopsy-proven predictive value (PPV3) and accuracy for detecting suspicious MRI findings. Methods: From all high-risk screening breast MRIs conducted between 1 July 2011 and 30 June 2020, we reviewed all BI-RADS 4/5 observations with pathological tissue diagnoses. Overall and annual PPV3s were computed. Radiologists with fewer than ten observations were excluded from performance analyses. PPV3s were computed for each radiologist. We assessed how MRI experience, BPE, and FGT impacted diagnostic accuracy using logistic regression analyses, defining positive cases as malignancies alone (definition A) or malignant or high-risk lesions (definition B). Findings: There were 536 BI-RADS 4/5 observations with tissue diagnoses, including 77 malignant and 51 high-risk lesions. A total of 516 observations were included in the radiologist performance analyses. The average radiologist's PPV3 was 16 ± 6% (definition A) and 25 ± 8% (definition B). MRI experience in years correlated significantly with positive cases (definition B, OR = 1.05, p = 0.03), independent of BPE or FGT. Diagnostic accuracy improved exponentially with increased MRI experience (definition B, OR of 1.27 and 1.61 for 5 and 10 years, respectively, p = 0.03 for both). Lower levels of BPE significantly correlated with increased odds of findings being malignant, independent of FGT and MRI experience. Summary: More extensive MRI reading experience improves radiologists' diagnostic accuracy for high-risk or malignant lesions, even in MRI studies with increased BPE.

Clinical and imaging modality factors impacting radiological interpretation of breast screening in young women with neurofibromatosis type 1

Young women with Neurofibromatosis type 1 (NF1) have a high risk of developing breast cancer and poorer survival following breast cancer diagnosis. International guidelines recommend commencing breast screening between 30 and 35 years; however, the optimal screening modality is unestablished, and previous reports suggest that breast imaging may be complicated by the presence of intramammary and cutaneous neurofibromas (cNFs). The aim of this study was to explore potential barriers to implementation of breast screening for young women with NF1.Twenty-seven women (30-47 years) with NF1 completed breast screening with breast MRI, mammogram and breast ultrasound. Nineteen probably benign/suspicious lesions were detected across 14 women. Despite the presence of breast cNFs, initial biopsy rate for participants with NF1 (37%), were comparable to a BRCA pathogenic variant (PV) cohort (25%) (P = 0.311). No cancers or intramammary neurofibromas were identified. Most participants (89%) returned for second round screening.The presence of cNF did not affect clinician confidence in 3D mammogram interpretation, although increasing breast density, frequently seen in young women, impeded confidence for 2D and 3D mammogram. Moderate or marked background parenchymal enhancement on MRI was higher in the NF1 cohort (70.4%) than BRCA PV carriers (47.3%), which is an independent risk factor for breast cancer.Breast MRI was the preferred mode of screening over mammogram, as the majority (85%) with NF1 demonstrated breast density (BI-RADS 3C/4D), which hinders mammogram interpretation. For those with high breast density and high cNF breast coverage, 3D rather than 2D mammogram is preferred, if MRI is unavailable.

Breast Cancer Screening Utilization and Outcomes in Women With Neurofibromatosis Type 1

INTRODUCTION

Women with neurofibromatosis type 1 (NF1) have up to a 5-fold increased risk for breast cancer before age 50 and a 3.5-fold increased risk of breast cancer overall. The purpose of our study was to assess breast cancer screening utilization and outcomes in this population.

PATIENTS AND METHODS

This IRB approved HIPAA compliant study retrospectively assessed consecutive NF1 patients (January 2012-December 2021) with recorded clinical visits and/or breast imaging. Patient demographics, risk factors, and screening mammogram and breast magnetic resonance imaging (MRI) outcomes were recorded. Descriptive statistics were obtained and standard breast screening measures were calculated.

RESULTS

One hundred and eleven women (median age 43, range 30-82) were eligible for screening based on current NCCN guidelines. A total of 86% (95/111) of all patients and 80% (24/30) of patients under age 40 had at least 1 mammogram. In contrast, 28% (31/111) of all patients and 33% (25/76) of patients ages 30 to 50 had at least 1 screening MRI. Of 368 screening mammograms performed, 38 of 368 (10%) resulted in the recall, and 22 of 368 (6%) resulted in a biopsy. Of 48 screening MRIs performed, 19 of 48 (40%) short-term follow-ups and 12 of 48 (25%) biopsies were recommended. All 6 screen-detected cancers in our cohort were detected initially on screening mammograms.

CONCLUSION

Results confirm the utility and performance of screening mammography in the NF1 population. The low utilization of MRI in our cohort limits the evaluation of outcomes via this modality and suggests there may be an education or interest gap among referrers and patients regarding supplemental screening recommendations.

Testing for Inherited Susceptibility to Breast Cancer

When BRCA1 and BRCA2 were first identified, the initial models for delivering testing were shaped by concepts of genetic exceptionalism and a lack of data regarding therapeutic implications and the effectiveness of risk reduction. Since then, interventions have been effective, and treatment implications have become clear. The sensitivity of guideline-based testing is incomplete, leading to calls for universal testing. Completely universal testing, however, is not necessary to identify the great majority of BRCA1 or BRCA2 variants. Broader testing (both in terms of eligibility and genes tested) will identify more variants, particularly in moderate penetrance genes, but the clinical implications remain less clear for these variants.

High Risk Breast Cancer Screening is a Double Edged Sword: A Qualitative Study of Patient Perspectives on the Ontario High Risk Breast Cancer Screening Program

BACKGROUND

MRI-based breast cancer screening for high-risk women has been associated with false positives. This study explored the benefits and drawbacks of MRI-based screening using in-depth patient interviews.

METHODS

This was a qualitative study of interviews with women participating in the High Risk Ontario Breast Screening Program. Women enrolled at two centers who had completed at least one round of screening were invited to participate. Recruitment was suspended when thematic saturation was reached. Semi-structured telephone interviews were conducted and transcribed verbatim. Emergent themes were identified and a coding framework established.

RESULTS

21 women (median age 41 years) participated in telephone interviews. Women had been in the program a median of 4 years (IQR 1-5), and 71% had experienced at least one abnormal screen. Eight participants (38%) had undergone biopsies. Six women (29%) were BRCA mutation carriers. MRIs were described as intimidating, uncomfortable, and claustrophobic. Participants were concerned about long-term exposure to gadolinium contrast. Compared to MRI, mammography alone was viewed as painful, less sensitive, and a "useless…waste of time." MRI provided a "psychological safety net" that outweighed the distress associated with abnormal screens. Many women accepted this trade-off as a "two-edged sword" that was "worth it" and provided a sense of control. Suggestions for improvement included more information regarding the risks of MRI, and access to counselling.

CONCLUSIONS

Women participating in MRI-based screening strongly value reassurance from a highly sensitive screening test. This outweighed the distress of abnormal screens. There are areas for improvement around patient communication and psychosocial support.

Feasibility Study and Clinical Impact of Incorporating Breast Tissue Density in High-Risk Breast Cancer Screening Assessment

Breast tissue density (BTD) is known to increase the risk of breast cancer but is not routinely used in the risk assessment of the population-based High-Risk Ontario Breast Screening Program (HROBSP). This prospective, IRB-approved study assessed the feasibility and impact of incorporating breast tissue density (BTD) into the risk assessment of women referred to HROBSP who were not genetic mutation carriers. All consecutive women aged 40-69 years who met criteria for HROBSP assessment and referred to Genetics from 1 December 2020 to 31 July 2021 had their lifetime risk calculated with and without BTD using Tyrer-Cuzick model version 8 (IBISv8) to gauge overall impact. McNemar's test was performed to compare eligibility with and without density. 140 women were referred, and 1 was excluded (BRCA gene mutation carrier and automatically eligible). Eight of 139 (5.8%) never had a mammogram, while 17/131 (13%) did not have BTD reported on their mammogram and required radiologist review. Of 131 patients, 22 (16.8%) were clinically impacted by incorporation of BTD: 9/131 (6.9%) became eligible for HROBSP, while 13/131 (9.9%) became ineligible (p = 0.394). It was feasible for the Genetics clinic to incorporate BTD for better risk stratification of eligible women. This did not significantly impact the number of eligible women while optimizing the use of high-risk supplemental MRI screening.

BI-RADS BERT and Using Section Segmentation to Understand Radiology Reports

Radiology reports are one of the main forms of communication between radiologists and other clinicians, and contain important information for patient care. In order to use this information for research and automated patient care programs, it is necessary to convert the raw text into structured data suitable for analysis. State-of-the-art natural language processing (NLP) domain-specific contextual word embeddings have been shown to achieve impressive accuracy for these tasks in medicine, but have yet to be utilized for section structure segmentation. In this work, we pre-trained a contextual embedding BERT model using breast radiology reports and developed a classifier that incorporated the embedding with auxiliary global textual features in order to perform section segmentation. This model achieved 98% accuracy in segregating free-text reports, sentence by sentence, into sections of information outlined in the Breast Imaging Reporting and Data System (BI-RADS) lexicon, which is a significant improvement over the classic BERT model without auxiliary information. We then evaluated whether using section segmentation improved the downstream extraction of clinically relevant information such as modality/procedure, previous cancer, menopausal status, purpose of exam, breast density, and breast MRI background parenchymal enhancement. Using the BERT model pre-trained on breast radiology reports, combined with section segmentation, resulted in an overall accuracy of 95.9% in the field extraction tasks. This is a 17% improvement, compared to an overall accuracy of 78.9% for field extraction with models using classic BERT embeddings and not using section segmentation. Our work shows the strength of using BERT in the analysis of radiology reports and the advantages of section segmentation by identifying the key features of patient factors recorded in breast radiology reports.

Effective Surveillance of High-Risk Women

BACKGROUND

This study addresses the effectiveness of risk models and screening breast magnetic resonance imaging (MRI) in women who have atypical hyperplasia (AH), lobular carcinoma in situ (LCIS), or a family history of breast cancer, but not a genetic mutation.

PATIENTS AND METHODS

A retrospective review of 444 women who had 458 breast screening MRIs at a community teaching hospital over a 12-month period between March 25, 2014 and March 31, 2015 was performed. The patients underwent high risk screening with breast MRIs alternating with mammograms every 6 months. After excluding patients with prior breast or ovarian cancer, genetic mutations, and chest wall radiation, 200 remaining patients constituted the study cohort. Over the following 5 years, the patients were screened with MRIs alternating with mammograms every 6 months. A total of 961 total MRI screenings were performed over the entire 5-year period of the study.

RESULTS

A total of 200 women fit the study criteria. Of these 103 had a prior history of AH or LCIS. Over the 5-year period, 60 women dropped out of the screening regimen, 6 patients were diagnosed with breast cancer on screening MRIs, and 2 additional patients were diagnosed with breast cancer on screening mammograms. Surprisingly, the highest Tyrer-Cuzick (T-C) scores did not correlate with increased development of breast cancers in our population.

CONCLUSIONS

This study shows that there is wide variation in the results of risk assessment models. Risk models may overestimate breast cancer risk, suggesting that re-evaluation of current risk assessment and screening protocols is warranted.

Molecular and cellular mechanisms of melatonin in breast cancer

Breast cancer is considered as one of the most important health problems due to its poor prognosis and high rate of mortality and new diagnosed cases. Annually, a great number of deaths are reported in men and women; this means that despite all the improvements in cancer diagnosis and treatment, still, an intense need for more effective approaches exists. Melatonin is a multivalent compound which has a hand in several cellular and molecular processes and therefore, is an appropriate candidate for treatment of many diseases like cancer. Currently, considerable properties of this agent have oriented the research towards investigating its effects specifically in breast cancer. In this review, we gathered a bunch of evidence in order to give a new sight for breast cancer treatment utilizing melatonin. We expect that in coming years, melatonin will become one of the most common therapeutic drugs with lesser side-effects than other chemotherapeutic drugs.

The Breast Cancer Screening and Timing of Breast MRI—Experience in a Genetic High-Risk Screening Clinic in a Comprehensive Cancer Center

For women with genetic risk of breast cancer, the addition of screening breast MRI to mammography has become a standard. The order and interval of annual imaging can be variable among providers. To evaluate the clinical implications related to the timing, we conducted a chart review on a cohort of women (N = 276) with high-risk (BRCA1, BRCA2, CDH1, PTEN and TP53) and moderate high-risk (ATM and CHEK2) predisposition to breast cancer in a 48-month follow up. The estimated MRI detection rate in the entire group is 1.75% (18 per 1000 MRI tests). For the high-risk group, the estimated rate is 2.98% (30 per 1000 MRI tests). Many women discovered their genetic risk at an age much older (average age of the high-risk group was 48 years) than the age recommended to initiate enhanced screening (age 20 to 25 years). In total, 4 of the 11 primary breast cancers detected were identified by screening MRI within the first month after initial visit, which were not detected by previous mammography, suggesting the benefit of initiating MRI immediately after the discovery of genetic risk. Breast screening findings for women with Lynch syndrome and neurofibromatosis type 1 were also included in this report.

Clinical experience of tensor-valued diffusion encoding for microstructure imaging by diffusional variance decomposition in patients with breast cancer

BACKGROUND

Diffusion-weighted imaging plays a key role in magnetic resonance imaging (MRI) of breast tumors. However, it remains unclear how to interpret single diffusion encoding with respect to its link with tissue microstructure. The purpose of this retrospective cross-sectional study was to use tensor-valued diffusion encoding to investigate the underlying microstructure of invasive ductal carcinoma (IDC) and evaluate its potential value in a clinical setting.

METHODS

We retrospectively reviewed biopsy-proven breast cancer patients who underwent preoperative breast MRI examination from July 2020 to March 2021. We reviewed the MRI of 29 patients with 30 IDCs, including analysis by diffusional variance decomposition enabled by tensor-valued diffusion encoding. The diffusion parameters of mean diffusivity (MD), total mean kurtosis (MKT), anisotropic mean kurtosis (MKA), isotropic mean kurtosis (MKI), macroscopic fractional anisotropy (FA), and microscopic fractional anisotropy (µFA) were estimated. The parameter differences were compared between IDC and normal fibroglandular breast tissue (FGBT), as well as the association between the diffusion parameters and histopathologic items.

RESULTS

The mean value of MD in IDCs was significantly lower than that of normal FGBT (1.07±0.27 vs. 1.34±0.29, P<0.001); however, MKT, MKA, MKI, FA, and µFA were significantly higher (P<0.005). Among all the diffusion parameters, MKI was positively correlated with the tumor size on both MRI and pathological specimen (rs=0.38, P<0.05 vs. rs=0.54, P<0.01), whereas MKT had a positive correlation with the tumor size in the pathological specimen only (rs=0.47, P<0.02). In addition, the lymph node (LN) metastasis group had significantly higher MKT, MKA, and µFA compared to the metastasis negative group (P<0.05).

CONCLUSIONS

Tensor-valued diffusion encoding enables a useful non-invasive method for characterizing breast cancers with information on tissue microstructures. Particularly, µFA could be a potential imaging biomarker for evaluating breast cancers prior to surgery or chemotherapy.

Missed Breast Cancers on MRI in High-Risk Patients: A Retrospective Case–Control Study

Purpose: To determine if MRI features and molecular subtype influence the detectability of breast cancers on MRI in high-risk patients. Methods and Materials: Breast cancers in a high-risk population of 104 patients were diagnosed following MRI describing a BI-RADS 4–5 lesion. MRI characteristics at the time of diagnosis were compared with previous MRI, where a BI-RADS 1–2–3 lesion was described. Results: There were 77 false-negative MRIs. A total of 51 cancers were overlooked and 26 were misinterpreted. There was no association found between MRI characteristics, the receptor type and the frequency of missed cancers. The main factors for misinterpreted lesions were multiple breast lesions, prior biopsy/surgery and long-term stability. Lesions were mostly overlooked because of their small size and high background parenchymal enhancement. Among missed lesions, 50% of those with plateau kinetics on initial MRI changed for washout kinetics, and 65% of initially progressively enhancing lesions then showed plateau or washout kinetics. There were more basal-like tumours in BRCA1 carriers (50%) than in non-carriers (13%), p = 0.0001, OR = 6.714, 95% CI = [2.058–21.910]. The proportion of missed cancers was lower in BRCA carriers (59%) versus non-carriers (79%), p < 0.05, OR = 2.621, 95% CI = [1.02–6.74]. Conclusions: MRI characteristics or molecular subtype do not influence breast cancer detectability. Lesions in a post-surgical breast should be assessed with caution. Long-term stability does not rule out malignancy and multimodality evaluation is of added value. Lowering the biopsy threshold for lesions with an interval change in kinetics for a type 2 or 3 curve should be considered. There was a higher rate of interval cancers in BRCA 1 patients attributed to lesions more aggressive in nature.

Radiologist-Level Performance by Using Deep Learning for Segmentation of Breast Cancers on MRI Scans

Purpose

To develop a deep network architecture that would achieve fully automated radiologist-level segmentation of cancers at breast MRI.

Materials and Methods

In this retrospective study, 38 229 examinations (composed of 64 063 individual breast scans from 14 475 patients) were performed in female patients (age range, 12–94 years; mean age, 52 years ± 10 [standard deviation]) who presented between 2002 and 2014 at a single clinical site. A total of 2555 breast cancers were selected that had been segmented on two-dimensional (2D) images by radiologists, as well as 60 108 benign breasts that served as examples of noncancerous tissue; all these were used for model training. For testing, an additional 250 breast cancers were segmented independently on 2D images by four radiologists. Authors selected among several three-dimensional (3D) deep convolutional neural network architectures, input modalities, and harmonization methods. The outcome measure was the Dice score for 2D segmentation, which was compared between the network and radiologists by using the Wilcoxon signed rank test and the two one-sided test procedure.

Results

The highest-performing network on the training set was a 3D U-Net with dynamic contrast-enhanced MRI as input and with intensity normalized for each examination. In the test set, the median Dice score of this network was 0.77 (interquartile range, 0.26). The performance of the network was equivalent to that of the radiologists (two one-sided test procedures with radiologist performance of 0.69–0.84 as equivalence bounds, P < .001 for both; n = 250).

Conclusion

When trained on a sufficiently large dataset, the developed 3D U-Net performed as well as fellowship-trained radiologists in detailed 2D segmentation of breast cancers at routine clinical MRI.

Keywords: MRI, Breast, Segmentation, Supervised Learning, Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms

Published under a CC BY 4.0 license.

Supplemental material is available for this article.

MRI versus Mammography plus Ultrasound in Women at Intermediate Breast Cancer Risk: Study Design and Protocol of the MRIB Multicenter, Randomized, Controlled Trial

In women at high/intermediate lifetime risk of breast cancer (BC-LTR), contrast-enhanced magnetic resonance imaging (MRI) added to mammography ± ultrasound (MX ± US) increases sensitivity but decreases specificity. Screening with MRI alone is an alternative and potentially more cost-effective strategy. Here, we describe the study protocol and the characteristics of enrolled patients for MRIB feasibility, multicenter, randomized, controlled trial, which aims to compare MRI alone versus MX+US in women at intermediate breast cancer risk (aged 40-59, with a 15-30% BC-LTR and/or extremely dense breasts). Two screening rounds per woman were planned in ten centers experienced in MRI screening, the primary endpoint being the rate of cancers detected in the 2 arms after 5 years of follow-up. From July 2013 to November 2015, 1254 women (mean age 47 years) were enrolled: 624 were assigned to MX+US and 630 to MRI. Most of them were aged below 50 (72%) and premenopausal (45%), and 52% used oral contraceptives. Among postmenopausal women, 15% had used hormone replacement therapy. Breast and/or ovarian cancer in mothers and/or sisters were reported by 37% of enrolled women, 79% had extremely dense breasts, and 41% had a 15-30% BC-LTR. The distribution of the major determinants of breast cancer risk profiles (breast density and family history of breast and ovarian cancer) of enrolled women varied across centers.

"High-Risk Breast Cancer Screening in BRCA1/2 Carriers Leads to Early Detection and Improved Survival After a Breast Cancer Diagnosis"

BACKGROUND

Germline BRCA1/2 pathogenic variant (PV) carriers have high lifetime risk of developing breast cancer and therefore subjected to intense lifetime screening. However, solid data on the effectiveness of high-risk screening of the BRCA1/2 carrier population is limited.

PATIENTS AND METHODS

Retrospectively, we analyzed 346 women diagnosed with breast tumors. Patients were divided according to the timing of BRCA1/2 PVrecognition, before (BRCA-preDx awareness, N = 62) or after (BRCA-postDx awareness group, N = 284) cancer diagnosis.

RESULTS

Median follow-up times were 131.42 and 93.77 months in the BRCA-preDx awareness and BRCA-postDx awareness groups, respectively. In the BRCA-preDx awareness group, 78.7% of the patients had invasive tumors and 21.3% were diagnosed with pure ductal carcinoma in situ. In contrast, in the BRCA-postDx awareness group over 93% of women were diagnosed with invasive cancer and only 6.4% had in situ disease. The mode of tumor detection differed significantly between the groups: 71.9% in the BRCA-postDx awareness group and 26.2% in the BRCA-preDx awareness group were diagnosed after personally palpating a lump. Tumor size and nodal involvement were significantly more favorable in the BRCA-preDx awareness group. T stage was significantly lower in the BRCA-preDx awareness group: 54.84% at T1 and 20.96% at Tis. In the BRCA-postDx awareness group, only 37.54% were at T1 and 6.49% at Tis. The N stage was also significantly lower in the BRCA-preDx awareness group: 71% had no lymph node metastases, compared with 56.1% in the BRCA-postDx awareness group. Additionally, therapeutic procedures varied between the groups: BRCA-preDx awareness group patients underwent more breast conserving surgeries. Axillary lymph node dissection was done in 38% of women in the BRCA-postDx awareness group and in only 8.7% of the BRCA-preDx awareness group patients. Interestingly, improved survival was found among patients who underwent high-risk screening (hazard ratio=0.34).

CONCLUSIONS

High-risk screening might facilitate downstaging of detected breast tumor among BRCA1/2 carrier population.

Personalized Risk Assessment for Prevention and Early Detection of Breast Cancer: Integration and Implementation (PERSPECTIVE I&I)

Early detection of breast cancer through screening reduces breast cancer mortality. The benefits of screening must also be considered within the context of potential harms (e.g., false positives, overdiagnosis). Furthermore, while breast cancer risk is highly variable within the population, most screening programs use age to determine eligibility. A risk-based approach is expected to improve the benefit-harm ratio of breast cancer screening programs. The PERSPECTIVE I&I (Personalized Risk Assessment for Prevention and Early Detection of Breast Cancer: Integration and Implementation) project seeks to improve personalized risk assessment to allow for a cost-effective, population-based approach to risk-based screening and determine best practices for implementation in Canada. This commentary describes the four inter-related activities that comprise the PERSPECTIVE I&I project. 1: Identification and validation of novel moderate to high-risk susceptibility genes. 2: Improvement, validation, and adaptation of a risk prediction web-tool for the Canadian context. 3: Development and piloting of a socio-ethical framework to support implementation of risk-based breast cancer screening. 4: Economic analysis to optimize the implementation of risk-based screening. Risk-based screening and prevention is expected to benefit all women, empowering them to work with their healthcare provider to make informed decisions about screening and prevention.

Cost effectiveness of in vitro fertilisation and preimplantation genetic testing to prevent transmission of BRCA1/2 mutations

STUDY QUESTION

Is it cost-effective to use in vitro fertilisation and preimplantation genetic testing of monogenic defects (IVT/PGT-M) to prevent transmission of BRCA1/2 mutations to second-generation new births in comparison with naturally conceived births?

SUMMARY ANSWER

In this cost-effectiveness analysis, we found that IVF/PGT-M is cost-effective for BRCA1 and BRCA2 mutation carriers if using a willingness to pay of $50 000 per quality-adjusted life-year (QALY).

WHAT IS KNOWN ALREADY

Carriers of a BRCA1 or BRCA2 mutation have a significantly increased risk of several types of cancer throughout their lifetime. The cost of risk reduction, screening and treatment of cancer in this population is high. In addition, there is a 50% chance of passing on this genetic mutation to each child. One option to avoid transmission of an inherited deleterious gene to one's offspring involves in vitro fertilisation with preimplantation genetic testing.

STUDY DESIGN, SIZE, DURATION

We implemented a state transition model comparing the healthcare impact of a cohort of healthy children born after IVF/PGT-M, who have a population risk of developing cancer, to a cohort of naturally conceived live-births, half of whom are carriers of the BRCA mutation. Transition probabilities are based on published sources, a lifetime horizon and a perspective of a provincial Ministry of Health in Canada.

PARTICIPANTS/MATERIALS, SETTING, METHODS

The target population is the second-generation new births who have at least one parent with a known BRCA1 or BRCA2 mutation.

MAIN RESULTS AND THE ROLE OF CHANCE

At a willingness-to-pay threshold of $50 000 per QALY, IVF/PGT-M is a cost-effective intervention for carriers of either BRCA mutation. For BRCA1, the incremental cost-effectiveness ratio (ICER) for IVF/PGT-M is $14 242/QALY. For BRCA2, the ICER of intervention is $12 893/QALY. Probabilistic sensitivity analysis results show that IVF/PGT-M has a 98.4 and 97.3% chance of being cost-effective for BRCA1 and BRCA2 mutation carriers, respectively, at the $50 000/QALY threshold.

LIMITATIONS, REASONS FOR CAUTION

Our model did not include the short-term negative effect of IVF/PGT-M on the woman's quality of life; in addition, our model did not consider any ethical issues related to post-implantation genetic testing.

WIDER IMPLICATIONS OF THE FINDINGS

In countries in which the healthcare of a large segment of the population is covered by a single payer system such as the government, it would be cost-effective for that payer to cover the cost of IVF/PGT-M for couples in which one member has a BRCA mutation, in order to avoid the future costs and disutility of managing offspring with an inherited BRCA mutation.

STUDY FUNDING/COMPETING INTEREST(S)

Dr Wong's research program was supported by the Canadian Institutes of Health Research (CIHR), the Natural Sciences and Engineering Research Council (NSERC), the Canadian Liver Foundation and an Ontario Ministry of Research, Innovation and Science Early Researcher Award. All authors declared no conflict of interests.

Abnormal screens among nonmutation carriers in the High Risk Ontario Breast Screening Program

BACKGROUND

The Ontario Breast Screening Program was expanded in 2011 to offer annual MRI and mammography to women with high-risk genetic mutations (e.g., BRCA1/2) and women with strong family histories and ≥25% estimated lifetime risk of breast cancer. Data to support high-risk screening is less clear in the nonmutation carrier group, as MRI has lower specificity among this population. The potential unintended consequences may be considerable and need to be explored. We aimed to describe the frequency of abnormal screens and biopsies.

METHODS

Demographic surveys and chart review consent were sent to a sample of 441 individuals enrolled in a high-risk screening program at two tertiary care hospitals in Toronto, Ontario. Retrospective cross-sectional chart review was undertaken for clinicopathologic data. The frequencies of abnormal screens and biopsies were calculated.

RESULTS

One hundred sixty-nine nonmutation carriers were included. The majority were white, employed, and highly educated. The median International Breast Cancer Intervention Study lifetime risk of breast cancer was 28.0% (range 24.5%-89.0%). 108 individuals (64%) experienced at least 1 abnormal screen and 13 (8%) had 3 or more over a median 3 years of screening (range 1-6 years). Of 55 biopsies, 3 (5.5%) were malignant. The cancer detection rate was 8.4/1000 screens (95% CI 3.2-22.4).

CONCLUSIONS

An MRI-based screening program for nonmutation carriers was effective at diagnosing breast cancer. However, this population experienced a high rate of abnormal screens and intervention. Further research is needed to improve the performance of MRI-based screening in these women.

Optimal Screening in Breast Cancer Survivors With Dense Breasts on Mammography

The Oncology Grand Rounds series is designed to place original reports published in the Journal into clinical context. A case presentation is followed by a description of diagnostic and management challenges, a review of the relevant literature, and a summary of the authors' suggested management approaches. The goal of this series is to help readers better understand how to apply the results of key studies, including those published in Journal of Clinical Oncology, to patients seen in their own clinical practice.

Imaging-based prostate cancer screening among BRCA mutation carriers-results from the first round of screening

BACKGROUND

Male-carriers of BRCA1/2 gene mutations have an increased risk of prostate cancer (PCa) with a more aggressive phenotype. Current screening-guidelines suggest the use of prostate-specific antigen (PSA) only among BRCA2 carriers. Female carriers have extensive guidelines that include imaging. Our objective was to test the prevalence of PCa among BRCA carriers and examine screening strategies, using PSA and multiparametric magnetic resonance imaging (mpMRI).

PATIENTS AND METHODS

We recruited men aged 40-70 years with BRCA1/2 germline mutations and no prior history of prostate biopsy. All men underwent an initial round of screening which included PSA, and prostate mpMRI. PSA was considered elevated using an age-stratified threshold of ≥1 ng/ml for 40-50 years of age, ≥2 ng/ml for 50-60 years of age, and 2.5 ng/ml for 60-70 years of age. Men with elevated PSA and/or suspicious lesion on mpMRI were offered a prostate biopsy. PSA levels, MRI findings, PCa incidence, and tumor characteristics were evaluated. Decision curve analysis was used to compare screening strategies.

RESULTS

We recruited 188 men (108 BRCA1, 80 BRCA2), mean age 54 years (9.8). One hundred and ten (57%) had either elevated age-stratified PSA (75; 40%), a suspicious MRI lesion (67; 36%), or both (32; 17%). Of these, 92 (85%) agreed to perform a prostate biopsy. Sixteen (8.5%) were diagnosed with PCa; 44% of the tumors were classified as intermediate- or high-risk disease. mpMRI-based screening missed only one of the cancers (6%), while age-stratified PSA would have missed five (31%). Decision curve analysis showed that mpMRI screening, regardless of PSA, had the highest net benefit for PCa diagnosis, especially among men younger than 55 years of age. We found no difference in the risk of PCa between BRCA1 and BRCA2 (8.3% versus 8.7%, P = 0.91). Ninety percent had a Jewish founder mutation, thus the results cannot be generalized to all ethnic groups.

CONCLUSIONS

PCa is prevalent among BRCA carriers. Age may affect screening strategy for PCa in this population. Young carriers could benefit from initial MRI screening. BRCA carriers aged older than 55 years should use PSA and be referred to mpMRI if elevated.

TRIAL REGISTRATION

ClinicalTrial.gov ID: NCT02053805.

Radiologic Technologist and Radiologist Knowledge Gaps about Breast Density Revealed by an Online Continuing Education Course

OBJECTIVE

We sought to identify provider knowledge gaps and their predictors, as revealed by a breast density continuing education course marketed to the radiology community.

METHODS

The course, continually available online during the study period of November 2, 2016 and December 31, 2018, includes demographics collection; a monograph on breast density, breast cancer risk, and screening; and a post-test. Four post-test questions were modified during the study period, resulting in different sample sizes pre- and postmodification. Multiple logistic regression was used to identify predictors of knowledge gaps (defined as > 25% of responses incorrect).

RESULTS

Of 1649 analyzable registrants, 1363 (82.7%) were radiologic technologists, 226 (13.7%) were physicians, and 60 (3.6%) were other nonphysicians; over 90% of physicians and over 90% of technologists/nonphysicians specialized in radiology. Sixteen of 49 physicians (32.7%) and 80/233 (34.3%) technologists/nonphysicians mistakenly thought the Gail model should be used to determine "high-risk" status for recommending MRI or genetic testing. Ninety-nine of 226 (43.8%) physicians and 682/1423 (47.9%) technologists/nonphysicians misunderstood the inverse relationship between increasing age and lifetime breast cancer risk. Fifty-two of 166 (31.3%) physicians and 549/1151 (47.7%) technologists/nonphysicians were unaware that MRI should be recommended for women with a family history of BRCA1/BRCA2 mutations. Tomosynthesis effectiveness was overestimated, with 18/60 (30.0%) physicians and 95/272 (34.9%) technologists/nonphysicians believing sensitivity nearly equaled MRI. Knowledge gaps were more common in technologists/nonphysicians.

CONCLUSIONS

Important knowledge gaps about breast density, breast cancer risk assessment, and screening exist among radiologic technologists and radiologists. Continued education efforts may improve appropriate breast cancer screening recommendations.

Performance Measures of Magnetic Resonance Imaging Plus Mammography in the High Risk Ontario Breast Screening Program

BACKGROUND

The Ontario Breast Screening Program expanded in July 2011 to screen high-risk women age 30-69 years with annual magnetic resonance imaging (MRI) and digital mammography. This study examined the benefits of screening with mammography and MRI by age and risk criteria.

METHODS

This prospective cohort study included 8782 women age 30-69 years referred to the High Risk Ontario Breast Screening Program from July 2011 to June 2015, with final results to December 2016. Cancer detection rates, sensitivity, and specificity of MRI and mammography combined were compared with each modality individually within risk groups stratified by age using generalized estimating equation models. Prognostic features of screen-detected breast cancers were compared by modality using Fisher exact test. All P values are two-sided.

RESULTS

Among 20 053 screening episodes, there were 280 screen-detected breast cancers (cancer detection rate = 14.0 per 1000, 95% confidence interval [CI] = 12.4 to 15.7). The sensitivity of mammography was statistically significantly lower than that of MRI plus mammography (40.8%, 95% CI = 29.3% to 53.5% vs 96.0%, 95% CI = 92.2% to 98.0%, P < .001). In mutation carriers age 30-39 years, sensitivity of the combination was comparable with MRI alone (100.0% vs 96.8%, 95% CI = 79.2% to 100.0%, P = .99) but with statistically significantly decreased specificity (78.0%, 95% CI = 74.7% to 80.9% vs 86.2%, 95% CI = 83.5% to 88.5%, P < .001). In women age 50-69 years, combining MRI and mammography statistically significantly increased sensitivity compared with MRI alone (96.3%, 95% CI = 90.6% to 98.6% vs 90.9%, 95% CI = 83.6% to 95.1%, P = .02), with a small but statistically significant decrease in specificity (84.2%, 95% CI = 83.1% to 85.2% vs 90.0%, 95% CI = 89.2% to 90.9%, P < .001).

CONCLUSIONS

Screening high risk women age 30-39 years with annual MRI only may be sufficient for cancer detection and should be evaluated further, particularly for mutation carriers. Among women age 50-69 years, detection is most effective when mammography is included with annual MRI.

Clinical Feasibility of Reduced Field-of-View Diffusion-Weighted Magnetic Resonance Imaging with Computed Diffusion-Weighted Imaging Technique in Breast Cancer Patients

Background: We evaluated the feasibility of the reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) with computed DWI technique by comparison and analysis of the inter-method agreement among acquired rFOV DWI (rFOVA), rFOV DWI with computed DWI technique (rFOVS), and dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) in patients with breast cancer. Methods: A total of 130 patients with biopsy-proven breast cancers who underwent breast MRI from April 2017 to December 2017 were included in this study. The rFOVS were reformatted by calculation of the apparent diffusion coefficient curve obtained from rFOVA b = 0 s/mm² and b = 500 s/mm². Visual assessment of the image quality of rFOVA b = 1000 s/mm², rFOVS, and DCE MRI was performed using a four-point grading system. Morphologic analyses of the index cancer was performed on rFOVA, rFOVS, and DCE MRI. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and contrast of tumor-to-parenchyma (TPC) were calculated. Results: Image quality scores with rFOVA, rFOVS, and DCE MRI were not significantly different (p = 0.357). Lesion analysis of shape, margin, and size of the index cancer also did not show significant differences among the three sequences (p = 0.858, p = 0.242, and p = 0.858, respectively). SNR, CNR, and TPC of DCE MRI were significantly higher than those of rFOVA and rFOVS (p < 0.001, p = 0.001, and p = 0.016, respectively). Significant differences were not found between the SNR, CNR, and TPC of rFOVA and those of rFOVS (p > 0.999, p > 0.999, and p > 0.999, respectively). Conclusion: The rFOVA and rFOVS showed nearly equivalent levels of image quality required for morphological analysis of the tumors and for lesion conspicuity compared with DCE MRI.

Screening in patients with increased risk of breast cancer (part 2). Where are we now? Actual MRI screening controversies

For women with a high risk of breast cancer, early detection plays an important role. Due to the high incidence of breast cancer, and at a younger age than in the general population, screening begins earlier, and there is considerable evidence that magnetic resonance is the most sensitive diagnostic tool, and the principal American and European guidelines agree on the recommendation to perform annual magnetic resonance (with supplemental annual mammography) as an optimal mode of screening. In addition to the absence of current consensus on which patients should be included in the recommendation for magnetic resonance screening (widely discussed in the introduction of part 1 of this work), there are other aspects that are different between guidelines, that are not specified, or that are susceptible to change based on the evidence of several years of experience, that we have called «controversies», such as the age to begin screening, the possible advisability of using a different strategy in different subgroups, performing alternate versus synchronous magnetic resonance and mammography, the age at which to terminate the two techniques, or how to follow up after risk reduction surgery.The aim of the second part of the paper is, by reviewing the literature, to provide an update in relation to some of the main «controversies» in high risk screening with magnetic resonance. And finally, based on all this, to propose a possible model of optimal and updated screening protocol.

Solving the preoperative breast MRI conundrum: design and protocol of the MIPA study

Despite its high diagnostic performance, the use of breast MRI in the preoperative setting is controversial. It has the potential for personalized surgical management in breast cancer patients, but two of three randomized controlled trials did not show results in favor of its introduction for assessing the disease extent before surgery. Meta-analyses showed a higher mastectomy rate in women undergoing preoperative MRI compared to those who do not. Nevertheless, preoperative breast MRI is increasingly used and a survey from the American Society of Breast Surgeons showed that 41% of respondents ask for it in daily practice. In this context, a large-scale observational multicenter international prospective analysis (MIPA study) was proposed under the guidance of the European Network for the Assessment of Imaging in Medicine (EuroAIM). The aims were (1) to prospectively and systematically collect data on consecutive women with a newly diagnosed breast cancer, not candidates for neoadjuvant therapy, who are offered or not offered breast MRI before surgery according to local practice; (2) to compare these two groups in terms of surgical and clinical endpoints, adjusting for covariates. The underlying hypotheses are that MRI does not cause additional mastectomies compared to conventional imaging, while reducing the reoperation rate in all or in subgroups of patients. Ninety-six centers applied to a web-based call; 36 were initially selected based on volume and quality standards; 27 were active for enrollment. On November 2018, the target of 7000 enrolled patients was reached. The MIPA study is presently at the analytic phase. Key Points • Breast MRI has a high diagnostic performance but its utility in the preoperative setting is controversial. • A large-scale observational multicenter prospective study was launched to compare women receiving with those not receiving preoperative MRI. • Twenty-seven centers enrolled more than 7000 patients. The study is presently at the analytic phase.

The supplemental value of mammographic screening over breast MRI alone in BRCA2 mutation carriers

Purpose

BRCA2 mutation carriers are offered annual breast screening with MRI and mammography. The aim of this study was to investigate the supplemental value of mammographic screening over MRI screening alone.

Methods

In this multicenter study, proven BRCA2 mutation carriers, who developed breast cancer during screening using both digital mammography and state-of-art breast MRI, were identified. Clinical data were reviewed to classify cases in screen-detected and interval cancers. Imaging was reviewed to assess the diagnostic value of mammography and MRI, using the Breast Imaging and Data System (BI-RADS) classification allocated at the time of diagnosis.

Results

From January 2003 till March 2019, 62 invasive breast cancers and 23 ductal carcinomas in situ were diagnosed in 83 BRCA2 mutation carriers under surveillance. Overall screening sensitivity was 95.2% (81/85). Four interval cancers occurred (4.7% (4/85)). MRI detected 73 of 85 breast cancers (sensitivity 85.8%) and 42 mammography (sensitivity 49.9%) (p < 0.001).

Eight mammography-only lesions occurred. In 1 of 17 women younger than 40 years, a 6-mm grade 3 DCIS, retrospectively visible on MRI, was detected with mammography only in a 38-year-old woman. The other 7 mammography-only breast cancers were diagnosed in women aged 50 years and older, increasing sensitivity in this subgroup from 79.5% (35/44) to 95.5% (42/44) (p ≤ 0.001).

Conclusions

In BRCA2 mutation carriers younger than 40 years, the benefit of mammographic screening over MRI was very small. In carriers of 50 years and older, mammographic screening contributed significantly. Hence, we propose to postpone mammographic screening in BRCA2 mutation carriers to at least age 40.

Image-based screening for men at high risk for breast cancer: Benefits and drawbacks

Male breast cancer is a rare malignancy. Due to low prevalence and limited data to support male breast cancer screening, there are currently no recommendations for image-based screening in asymptomatic men and few recommendations for men at high risk for breast cancer. However, symptomatically diagnosed cancers in men are typically advanced, suggesting that earlier detection may improve outcomes. In this article we briefly review the risk factors for male breast cancer, and discuss the potential benefits and possible drawbacks of routine image-based screening for men at high risk for breast cancer.

Does Medicare-eligible high-risk breast cancer screening MRI target the right women?

INTRODUCTION

MRI is the most sensitive modality to screen for breast cancer, but it is expensive with somewhat limited access. Audit of screening performance should reflect appropriate population targeting.

METHODS

An observational study on consecutively screened high-risk women, assessment of the contralateral breast staging a new cancer, or surveillance in women with prior breast cancer or high-risk lesion in Perth, Western Australia. All breast MRI studies from 1 January 2015 to 7 September 2018 were included. Studies were 3T comprising T2, DWI, ADC and T1-weighted +/- fat saturation +/- IV gadolinium, +/- subtraction. DCE was read on the dynamics or DynaCAD (Invivo, Gainesville, FL, USA). Fellowship-trained breast radiologists blindly double-read by consensus; additional reader/s arbitrated. The reference standard was the histopathology result or cancer registry notification for cancer diagnoses and benign biopsies, benign follow-up imaging or subsequent screening MRI.

RESULTS

Of 993 MRI studies in 554 women, 870 eligible MRI were performed in 471 women, and 706 had a reference standard. Median age was 44 years (range 18-80). The majority of studies (65% 457/706) were screening Medicare rebate-eligible high familial risk; 26% for surveillance after a breast cancer or contralateral staging; 6% screened BRCA carriers. Eleven cancers were diagnosed, eight were MRI-detected. Only two of these were at high-risk screening MRI. Five were detected at staging contralateral ILC, after negative 2D mammography and ultrasound. Cancer prevalence was highest for staging contralateral ILC, at 600/10,000 MRI, for high-risk screening 77/10,000 MRI and surveillance 116/10,000 MRI.

CONCLUSIONS

Cancers were predominantly detected in women undergoing preoperative staging of new invasive lobular carcinoma in the contralateral breast, rather than the Medicare rebate-eligible high-risk screening group.

The Utility of Short-Interval Follow-Up for Baseline High-Risk Screening Breast MRI

OBJECTIVE

Patients with a high lifetime risk of developing breast cancer undergo annual screening magnetic resonance imaging (MRI) starting at age 30. A proportion of these patients are subsequently required to undergo short-interval follow-up MRI 6 months after their baseline examination. The purpose of this study was to evaluate the utility and outcome of these short-interval follow-up examinations in this population.

METHODS

A retrospective review was performed of 523 patients who received their baseline high-risk breast screening MRI at our institution between 2013 and 2017. The proportion of patients who received a short-interval follow-up MRI at 6 months was recorded. The findings at baseline and follow-up MRI were recorded, as well as the outcomes and results of any interventions performed.

RESULTS

Ninety-six (17.6%) patients (age range: 25-67, mean age: 41) received a short-interval follow-up MRI following their baseline screening examination. Indications for follow-up included moderate to marked background parenchymal enhancement, nonmass enhancement, and likely benign enhancing mass. Of the 92 patients, 5 (5.4%) went on to have a biopsy, with none revealing a malignant pathology. The remainder either returned to routine screening (91.3%) or received further imaging in the form of ultrasound or additional follow-up MRI (3.3%).

CONCLUSION

Short-interval follow-up breast MRI in high-risk patients after a baseline screening study with likely benign findings is unlikely to yield clinically significant findings. This retrospective study can be considered a starting point for additional future work looking at the rate, indications, and yield of short-interval follow-up following baseline high-risk screening breast MRI studies.

Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images

BACKGROUND

The usefulness of 3D deep learning-based classification of breast cancer and malignancy localization from MRI has been reported. This work can potentially be very useful in the clinical domain and aid radiologists in breast cancer diagnosis.

PURPOSE

To evaluate the efficacy of 3D deep convolutional neural network (CNN) for diagnosing breast cancer and localizing the lesions at dynamic contrast enhanced (DCE) MRI data in a weakly supervised manner.

STUDY TYPE

Retrospective study.

SUBJECTS

A total of 1537 female study cases (mean age 47.5 years ±11.8) were collected from March 2013 to December 2016. All the cases had labels of the pathology results as well as BI-RADS categories assessed by radiologists.

FIELD STRENGTH/SEQUENCE

1.5 T dynamic contrast-enhanced MRI.

ASSESSMENT

Deep 3D densely connected networks were trained under image-level supervision to automatically classify the images and localize the lesions. The dataset was randomly divided into training (1073), validation (157), and testing (307) subsets.

STATISTICAL TESTS

Accuracy, sensitivity, specificity, area under receiver operating characteristic curve (ROC), and the McNemar test for breast cancer classification. Dice similarity for breast cancer localization.

RESULTS

The final algorithm performance for breast cancer diagnosis showed 83.7% (257 out of 307) accuracy (95% confidence interval [CI]: 79.1%, 87.4%), 90.8% (187 out of 206) sensitivity (95% CI: 80.6%, 94.1%), 69.3% (70 out of 101) specificity (95% CI: 59.7%, 77.5%), with the area under the curve ROC of 0.859. The weakly supervised cancer detection showed an overall Dice distance of 0.501 ± 0.274.

DATA CONCLUSION

3D CNNs demonstrated high accuracy for diagnosing breast cancer. The weakly supervised learning method showed promise for localizing lesions in volumetric radiology images with only image-level labels.

LEVEL OF EVIDENCE

4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;50:1144-1151.

Association of Retrospective Peer Review and Positive Predictive Value of Magnetic Resonance Imaging-Guided Vacuum-Assisted Needle Biopsies of Breast

Objective

To evaluate the association between retrospective peer review of breast magnetic resonance imaging-guided vacuum-assisted needle biopsies and positive predictive value of subsequent magnetic resonance imaging-guided biopsies.

Materials and Methods

In January, 2015, a weekly conference was initiated in our institution to evaluate all breast magnetic resonance imaging-guided vacuum-assisted needle biopsies performed over January 1, 2014-December 31, 2015. During this weekly conferences, breast dynamic contrast-enhanced magnetic resonance imaging findings of 6 anonymized cases were discussed and then the faculty voted on whether they agree with the biopsy indication, accurate sampling and radiology-pathology correlation. We retrospectively reviewed and compared the magnetic resonance imaging indication, benign or malignant pathology rates, lesion types and the positive predictive value of magnetic resonance imaging-guided vacuum-assisted needle biopsy in the years before and after initiating this group peer review.

Results

The number of dynamic contrast-enhanced magnetic resonance imaging and magnetic resonance imaging-guided vacuum-assisted needle biopsies before and after initiating the review were 1447 vs 1596 (p=0.0002), and 253 (17.5%) vs 203 (12.7%) (p=0.04), respectively. There was a significant decrease in the number of benign biopsies in 2015 (n=104) compared to 2014 (n=154, p=0.04). The positive predictive value of magnetic resonance imaging-guided biopsy significantly increased after group review was implemented (Positive predictive value in 2014=%39.1 and positive predictive value in 2015=%48.8) (p=0.03), although the indications (p=0.49), history of breast cancer (p=0.14), biopsied magnetic resonance imaging lesion types (p=0.53) were not different. Less surgical excision was performed on magnetic resonance imaging-guided vacuum-assisted needle biopsy identified high-risk lesions in 2015 (p=0.25).

Conclusion

Our study showed an association between retrospective peer review of past biopsies and increased positive predictive value of magnetic resonance imaging-guided vacuum-assisted needle biopsies in our institution.

Promoting Breast Cancer Surveillance: The EMPOWER Study, a Randomized Clinical Trial in the Childhood Cancer Survivor Study

PURPOSE

The aim of the current study was to increase the uptake of screening mammography among high-risk women who were treated for a childhood cancer with chest radiotherapy.

PATIENTS AND METHODS

Two hundred four female survivors in the Childhood Cancer Survivor Study who were treated with chest radiotherapy with 20 Gy or greater, age 25 to 50 years, and without breast imaging in the past 24 months were randomly assigned 2:1 to receive a mailed informational packet followed by a tailored telephone-delivered brief motivational interview (intervention) versus an attention control. Primary outcome was the difference in the proportion of participants who completed a screening mammogram by 12 months as evaluated in an intent-to-treat analysis. Stratum-adjusted relative risk (RR) and 95% CI were estimated using the Cochran-Mantel-Haenszel method. Secondary outcomes included the completion of screening breast magnetic resonance imaging (MRI) and barriers to screening and moderating factors.

RESULTS

Women in the intervention group were significantly more likely than those in the control group to report a mammogram (45 [33.1%] of 136 v 12 [17.6%] of 68; RR, 1.9; 95% CI, 1.1 to 3.3). The intervention was more successful among women age 25 to 39 years (RR, 2.2; 95% CI, 1.1 to 4.7) than among those age 40 to 50 years (RR, 1.4; 95% CI, 0.6 to 3.2). The proportion of women who reported a breast MRI at 12 months was similar between the two groups: 16.2% (intervention) compared with 13.2% (control; RR, 1.2; 95% CI, 0.6 to 2.5). Primary barriers to completing a screening mammogram and/or breast MRI included lack of physician recommendation, deferred action by survivor, cost, and absence of symptoms.

CONCLUSION

Use of mailed materials followed by telephone-delivered counseling increased mammography screening rates in survivors at high risk for breast cancer; however, this approach did not increase the rate of breast MRI. Cost of imaging and physician recommendation were important barriers that should be addressed in future studies.

Subsequent Breast Cancer in Female Childhood Cancer Survivors in the St Jude Lifetime Cohort Study (SJLIFE)

PURPOSE

Anthracycline-associated risk for subsequent breast cancer in childhood cancer survivors is hypothesized to be mediated by TP53 mutation-related gene-environment interactions. We characterized treatment/genetic risks and the impact of screening for breast cancer in the St Jude Lifetime Cohort.

PATIENTS AND METHODS

Female participants underwent risk-based assessments, prior health event validation, chest radiation dosimetry, and whole genome sequencing. Breast biopsy reports were reviewed. A subgroup (n = 139) underwent both breast magnetic resonance imaging and mammography. Multivariable regression was used to calculate hazard ratios (HRs) and 95% CIs.

RESULTS

Among 1,467 women, 56 developed 68 breast cancers at a median age 38.6 (range, 24.5 to 53.0) years. Cumulative incidences at age 35 years were 1% (no chest radiation) and 8% (≥ 10 Gy of chest radiation). In adjusted models, breast cancer was associated with 20 Gy or more of chest radiation versus none (HR, 7.6; 95% CI, 2.9 to 20.4), anthracycline exposure versus none (1 to 249 mg/m2: HR, 2.6; 95% CI, 1.1 to 6.2; ≥ 250 mg/m2: HR, 13.4, 95% CI, 5.5 to 32.5), and having a breast cancer predisposition gene mutation (HR, 23.0; 95% CI, 7.3 to 72.2). Anthracyclines 250 mg/m2 or greater remained significantly associated with increased risk of breast cancer in models excluding survivors with cancer predisposition gene mutations, chest radiation 10 Gy or greater, or both. Sensitivity/specificity were 53.8%/96.3% for mammography, 69.2%/91.4% for magnetic resonance imaging, and 85.8%/99.7% for dual imaging. Breast cancers detected by imaging and/or prophylactic mastectomy compared with physical findings were more likely to be in situ carcinomas, smaller, without lymph node involvement, and treated without chemotherapy.

CONCLUSION

Higher doses of anthracyclines are associated with increased risk of breast cancer independent of mutations in known cancer predisposition genes. Surveillance imaging identifies breast cancers less likely to require chemotherapy than those detected by physical findings.

NF1 Patients Receiving Breast Cancer Screening: Insights from The Ontario High Risk Breast Screening Program

Neurofibromatosis Type I (NF1) is caused by variants in neurofibromin (NF1). NF1 predisposes to a variety of benign and malignant tumor types, including breast cancer. Women with NF1 <50 years of age possess an up to five-fold increased risk of developing breast cancer compared with the general population. Impaired emotional functioning is reported as a comorbidity that may influence the participation of NF1 patients in regular clinical surveillance despite their increased risk of breast and other cancers. Despite emphasis on breast cancer surveillance in women with NF1, the uptake and feasibility of high-risk screening programs in this population remains unclear. A retrospective chart review between 2014–2018 of female NF1 patients seen at the Elizabeth Raab Neurofibromatosis Clinic (ERNC) in Ontario was conducted to examine the uptake of high-risk breast cancer screening, radiologic findings, and breast cancer characteristics. 61 women with pathogenic variants in NF1 enrolled in the high-risk Ontario breast screening program (HR-OBSP); 95% completed at least one high-risk breast screening modality, and four were diagnosed with invasive breast cancer. Our findings support the integration of a formal breast screening programs in clinical management of NF1 patients.

BRCA Mutation Carriers: Breast and Ovarian Cancer Screening Guidelines and Imaging Considerations

Patients who carry the BRCA1 and BRCA2 gene mutations have an underlying genetic predisposition for breast and ovarian cancers. These deleterious genetic mutations are the most common genes implicated in hereditary breast and ovarian cancers. This monograph summarizes the evidence behind current screening recommendations, reviews imaging protocols specific to this patient population, and illustrates some of the imaging nuances of breast and ovarian cancers in this clinical setting.

Intensive Surveillance with Biannual Dynamic Contrast-Enhanced Magnetic Resonance Imaging Downstages Breast Cancer in BRCA1 Mutation Carriers

PURPOSE

To establish a cohort of high-risk women undergoing intensive surveillance for breast cancer.Experimental Design: We performed dynamic contrast-enhanced MRI every 6 months in conjunction with annual mammography (MG). Eligible participants had a cumulative lifetime breast cancer risk ≥20% and/or tested positive for a pathogenic mutation in a known breast cancer susceptibility gene.

RESULTS

Between 2004 and 2016, we prospectively enrolled 295 women, including 157 mutation carriers (75 BRCA1, 61 BRCA2); participants' mean age at entry was 43.3 years. Seventeen cancers were later diagnosed: 4 ductal carcinoma in situ (DCIS) and 13 early-stage invasive breast cancers. Fifteen cancers occurred in mutation carriers (11 BRCA1, 3 BRCA2, 1 CDH1). Median size of the invasive cancers was 0.61 cm. No patients had lymph node metastasis at time of diagnosis, and no interval invasive cancers occurred. The sensitivity of biannual MRI alone was 88.2% and annual MG plus biannual MRI was 94.1%. The cancer detection rate of biannual MRI alone was 0.7% per 100 screening episodes, which is similar to the cancer detection rate of 0.7% per 100 screening episodes for annual MG plus biannual MRI. The number of recalls and biopsies needed to detect one cancer by biannual MRI were 2.8 and 1.7 in BRCA1 carriers, 12.0 and 8.0 in BRCA2 carriers, and 11.7 and 5.0 in non-BRCA1/2 carriers, respectively.

CONCLUSIONS

Biannual MRI performed well for early detection of invasive breast cancer in genomically stratified high-risk women. No benefit was associated with annual MG screening plus biannual MRI screening.See related commentary by Kuhl and Schrading, p. 1693.

An investigation of the effect of fat suppression and dimensionality on the accuracy of breast MRI segmentation using U-nets

PURPOSE

Accurate segmentation of the breast is required for breast density estimation and the assessment of background parenchymal enhancement, both of which have been shown to be related to breast cancer risk. The MRI breast segmentation task is challenging, and recent work has demonstrated that convolutional neural networks perform well for this task. In this study, we have investigated the performance of several two-dimensional (2D) U-Net and three-dimensional (3D) U-Net configurations using both fat-suppressed and nonfat-suppressed images. We have also assessed the effect of changing the number and quality of the ground truth segmentations.

MATERIALS AND METHODS

We designed eight studies to investigate the effect of input types and the dimensionality of the U-Net operations for the breast MRI segmentation. Our training data contained 70 whole breast volumes of T1-weighted sequences without fat suppression (WOFS) and with fat suppression (FS). For each subject, we registered the WOFS and FS volumes together before manually segmenting the breast to generate ground truth. We compared four different input types to the U-nets: WOFS, FS, MIXED (WOFS and FS images treated as separate samples), and MULTI (WOFS and FS images combined into a single multichannel image). We trained 2D U-Nets and 3D U-Nets with these data, which resulted in our eight studies (2D-WOFS, 3D-WOFS, 2D-FS, 3D-FS, 2D-MIXED, 3D-MIXED, 2D-MULTI, and 3D-MULT). For each of these studies, we performed a systematic grid search to tune the hyperparameters of the U-Nets. A separate validation set with 15 whole breast volumes was used for hyperparameter tuning. We performed Kruskal-Walis test on the results of our hyperparameter tuning and did not find a statistically significant difference in the ten top models of each study. For this reason, we chose the best model as the model with the highest mean dice similarity coefficient (DSC) value on the validation set. The reported test results are the results of the top model of each study on our test set which contained 19 whole breast volumes annotated by three readers fused with the STAPLE algorithm. We also investigated the effect of the quality of the training annotations and the number of training samples for this task.

RESULTS

The study with the highest average DSC result was 3D-MULTI with 0.96 ± 0.02. The second highest average is 2D WOFS (0.96 ± 0.03), and the third is 2D MULTI (0.96 ± 0.03). We performed the Kruskal-Wallis one-way ANOVA test with Dunn's multiple comparison tests using Bonferroni P-value correction on the results of the selected model of each study and found that 3D-MULTI, 2D-MULTI, 3D-WOFS, 2D-WOFS, 2D-FS, and 3D-FS were not statistically different in their distributions, which indicates that comparable results could be obtained in fat-suppressed and nonfat-suppressed volumes and that there is no significant difference between the 3D and 2D approach. Our results also suggested that the networks trained on single sequence images or multiple sequence images organized in multichannel images perform better than the models trained on a mixture of volumes from different sequences. Our investigation of the size of the training set revealed that training a U-Net in this domain only requires a modest amount of training data and results obtained with 49 and 70 training datasets were not significantly different.

CONCLUSIONS

To summarize, we investigated the use of 2D U-Nets and 3D U-Nets for breast volume segmentation in T1 fat-suppressed and without fat-suppressed volumes. Although our highest score was obtained in the 3D MULTI study, when we took advantage of information in both fat-suppressed and nonfat-suppressed volumes and their 3D structure, all of the methods we explored gave accurate segmentations with an average DSC on >94% demonstrating that the U-Net is a robust segmentation method for breast MRI volumes.

Contrast-enhanced MRI for breast cancer screening

Multiple studies in the first decade of the 21^st century have established contrast-enhanced breast MRI as a screening modality for women with a hereditary or familial increased risk for the development of breast cancer. In recent studies, in women with various risk profiles, the sensitivity ranges between 81% and 100%, which is approximately twice as high as the sensitivity of mammography. The specificity increases in follow-up rounds to around 97%, with positive predictive values for biopsy in the same range as for mammography. MRI preferentially detects the more aggressive/invasive types of breast cancer, but has a higher sensitivity than mammography for any type of cancer. This performance implies that in women screened with breast MRI, all other examinations must be regarded as supplemental. Mammography may yield ~5% additional cancers, mostly ductal carcinoma in situ, while slightly decreasing specificity and increasing the costs. Ultrasound has no supplemental value when MRI is used. Evidence is mounting that in other groups of women the performance of MRI is likewise superior to more conventional screening techniques. Particularly in women with a personal history of breast cancer, the gain seems to be high, but also in women with a biopsy history of lobular carcinoma in situ and even women at average risk, similar results are reported. Initial outcome studies show that breast MRI detects cancer earlier, which induces a stage-shift increasing the survival benefit of screening. Cost-effectiveness is still an issue, particularly for women at lower risk. Since costs of the MRI scan itself are a driving factor, efforts to reduce these costs are essential. The use of abbreviated MRI protocols may enable more widespread use of breast MRI for screening. Level of Evidence: 1 Technical Efficacy: Stage 5 J. Magn. Reson. Imaging 2019;50:377-390.

Assessment of Quantitative Magnetic Resonance Imaging Background Parenchymal Enhancement Parameters to Improve Determination of Individual Breast Cancer Risk

OBJECTIVES

The aims of this study were to identify optimal quantitative breast magnetic resonance imaging background parenchymal enhancement (BPE) parameters associated with breast cancer risk and compare performance to qualitative assessments.

METHODS

Using a matched case-control cohort of 46 high-risk women who underwent screening magnetic resonance imaging (23 who developed breast cancer matched to 23 who did not), fibroglandular tissue area, BPE area, and intensity metrics (mean, SD, quartiles, skewness, and kurtosis) were quantitatively measured at varying enhancement thresholds. Optimal thresholds for discriminating between cancer and control cohorts were identified for each metric and performance summarized using area under the receiver operating characteristic curve.

RESULTS

Women who developed breast cancer exhibited greater BPE area (adjusted P = 0.004) and higher intensity statistics (adjusted P < 0.004, except skewness and kurtosis with P > 0.99) than did control subjects, with areas under the receiver operating characteristic curve ranging from 0.75 to 0.78 at optimized thresholds.

CONCLUSIONS

Elevated quantitative BPE parameters, related to both area and intensity of enhancement, are associated with breast cancer development.

Screening BRCA1 and BRCA2 Mutation Carriers for Breast Cancer

Women with BRCA mutations, who choose to decline or defer risk-reducing mastectomy, require a highly sensitive breast screening regimen they can begin by age 25 or 30. Meta-analysis of multiple observational studies, in which both mammography and magnetic resonance imaging (MRI) were performed annually, demonstrated a combined sensitivity of 94% for MRI plus mammography compared to 39% for mammography alone. There was negligible benefit from adding screening ultrasound or clinical breast examination to the other two modalities. The great majority of cancers detected were non-invasive or stage I. While the addition of MRI to mammography lowered the specificity from 95% to 77%, the specificity improved significantly after the first round of screening. The median follow-up of women with screen-detected breast cancer in the above observational studies now exceeds 10 years, and the long-term breast cancer-free survival in most of these studies is 90% to 95%. However, ongoing follow-up of these study patients, as well of women screened and treated more recently, is necessary. Advances in imaging technology will make highly sensitive screening accessible to a greater number of high-risk women.