International Interobserver Variability of Breast Density Assessment.

PURPOSE: The aim of this study was to determine variability in visually assessed mammographic breast density categorization among radiologists practicing in Indonesia, the Netherlands, South Africa, and the United States. METHODS: Two hundred consecutive 2-D full-field digital screening mammograms obtained from September to December 2017 were selected and retrospectively reviewed from four global locations, for a total of 800 mammograms. Three breast radiologists in each location (team) provided consensus density assessments of all 800 mammograms using BI-RADS® density categorization. Interreader agreement was compared using Gwet's AC2 with quadratic weighting across all four density categories and Gwet's AC1 for binary comparison of combined not dense versus dense categories. Variability of distribution among teams was calculated using the Stuart-Maxwell test of marginal homogeneity across all four categories and using the McNemar test for not dense versus dense categories. To compare readers from a particular country on their own 200 mammograms versus the other three teams, density distribution was calculated using conditional logistic regression. RESULTS: For all 800 mammograms, interreader weighted agreement for distribution among four density categories was 0.86 (Gwet's AC2 with quadratic weighting; 95% confidence interval, 0.85-0.88), and for not dense versus dense categories, it was 0.66 (Gwet's AC1; 95% confidence interval, 0.63-0.70). Density distribution across four density categories was significantly different when teams were compared with one another and one team versus the other three teams combined (P < .001). Overall, all readers placed the largest number of mammograms in the scattered and heterogeneous categories. CONCLUSIONS: Although reader teams from four different global locations had almost perfect interreader agreement in BI-RADS density categorization, variability in density distribution across four categories remained statistically significant.

Clinical Diagnoses and Outcomes After Diagnostic Breast Ultrasound by Nurses and General Practitioner Physicians in Rural Rwanda.

PURPOSE: To scale up early detection of breast cancer in low- and middle-income countries, research is needed to inform the role of diagnostic breast ultrasound performed by nonradiologists in resource-constrained settings. The authors examined 2-year clinical follow-up and outcomes among women who underwent diagnostic breast ultrasound performed by nonradiologist clinicians participating in a breast ultrasound training and mentorship program at a rural Rwandan hospital. METHODS: Imaging findings, management plans, and pathologic results were prospectively collected during the training using a standardized form. Data on follow-up and outcomes for patients receiving breast ultrasound between January 2016 and March 2017 were retrospectively collected through medical record review. RESULTS: Two hundred twenty-nine breast palpable findings (199 patients) met the study's eligibility criteria. Of 104 lesions initially biopsied, 38 were malignant on initial biopsy; 3 lesions were identified as malignant on repeat biopsy. All 34 patients ultimately diagnosed with cancer received initial recommendations for either biopsy or aspiration by trainees. The positive predictive value of trainee biopsy recommendation was 34.8% (95% confidence interval, 24.8%-45.0%). The sensitivity of trainees' biopsy recommendation for identifying malignant lesions was 92.7% (95% confidence interval, 84.2%-100%). Of 46 patients who did not receive biopsy and were told to return for clinical or imaging follow-up, 37.0% did not return. CONCLUSIONS: Trained nonradiologist clinicians in Rwanda successfully identified suspicious breast lesions on diagnostic breast ultrasound. Loss to follow-up was common among patients instructed to return for surveillance, so lower biopsy thresholds, decentralized surveillance, or patient navigation should be considered for patients with low- or intermediate-suspicion lesions.

Downstream Mammary and Extramammary Cascade Services and Spending Following Screening Breast Magnetic Resonance Imaging vs Mammography Among Commercially Insured Women.

Importance: Increasing use of screening breast magnetic resonance imaging (MRI), including among women at low or average risk of breast cancer, raises concerns about resulting mammary and extramammary cascades (downstream services and new diagnoses) of uncertain value. Objective: To estimate rates of cascade events (ie, laboratory tests, imaging tests, procedures, visits, hospitalizations, and new diagnoses) and associated spending following screening breast MRI vs mammography among commercially insured US women. Design, Setting, and Participants: This cohort study used 2016 to 2018 data from the MarketScan research database (IBM Corporation), which includes claims and administrative data from large US employers and commercial payers. Participants included commercially insured women aged 40 to 64 years without prior breast cancer who received an index bilateral screening breast MRI or mammogram between January 1, 2017, and June 30, 2018. We used propensity scores based on sociodemographic, clinical, and utilization variables to match MRI recipients to mammogram recipients in each month of index service use. Data were analyzed from October 8, 2020, to October 28, 2021. Exposures: Breast MRI vs mammography. Main Outcomes and Measures: Mammary and extramammary cascade event rates and associated total and patient out-of-pocket spending in the 6 months following the index test. Results: In this study, 9208 women receiving breast MRI were matched with 9208 women receiving mammography (mean [SD] age, 51.4 [6.7] years). Compared with mammogram recipients, breast MRI recipients had 39.0 additional mammary cascade events per 100 women (95% CI, 33.7-44.2), including 5.0 additional imaging tests (95% CI, 3.8-6.2), 17.3 additional procedures (95% CI, 15.5-19.0), 13.0 additional visits (95% CI, 9.4-17.2), 0.34 additional hospitalizations (95% CI, 0.18-0.50), and 3.0 additional new diagnoses (95% CI, 2.5-3.6). For extramammary cascades, breast MRI recipients had 19.6 additional events per 100 women (95% CI, 8.6-30.7) including 15.8 additional visits (95% CI, 10.2-21.4) and no statistically significant differences in other events. Breast MRI recipients had higher total spending for mammary events ($564 more per woman; 95% CI, $532-$596), extramammary events ($42 more per woman; 95% CI, $16-$69), and overall ($1404 more per woman; 95% CI, $1172-$1636). They also had higher overall out-of-pocket spending ($31 more per woman; 95% CI, $6-$55). Conclusions and Relevance: In this cohort study of commercially insured women, breast MRI was associated with more mammary and extramammary cascade events and spending relative to mammography. These findings can inform cost-benefit assessments and coverage policies to ensure breast MRI is reserved for patients for whom benefits outweigh harms.

How Often Does Retrieval of a Clipped Lymph Node Change Adjuvant Therapy Recommendations? A Prospective, Consecutive, Patient Cohort Study.

BACKGROUND: Prior studies examining sentinel lymph node biopsy (SLNB) after neoadjuvant chemotherapy (NAC) for cN1 patients have demonstrated that 20% of biopsied, clipped lymph nodes (cLNs) are nonsentinel lymph nodes (non-SLNs). Our goal was to determine how often the cLN was a non-SLN among both cN0 and cN1 patients and how often cLN pathology impacted management. METHODS: Overall, 238 patients treated with NAC and surgery January 2019 to June 2020 were prospectively examined. Patients underwent routine axillary ultrasound, biopsy of suspicious nodes, and clip placement. Radioactive iodine-125 seed localization of the cLN was performed in cN1 patients only. Isolated tumor cells (ITCs) were considered node positive (ypN+) for both cN0 and cN1 cohorts. Chart review was performed to determine if cLNs were non-SLN and their ypN status. RESULTS: Of 118 cN0 patients, 115 of 118 (97%) underwent successful SLNB, 33 of whom had a cLN present; 21 of 33 (64%) cLNs were non-SLNs. Overall, 9 of 118 (8%) were ypN+; no cLN was ypN+ without additional +SLNs. Of 120 cN1 patients, 104 of 120 (87%) converted to cN0, 98 of 104 (94%) of which had attempted SLNB, and 95 of 98 (97%) successfully mapped. The cLN was a non-SLN in 18 of 95 (19%). Overall, 58 of 104 (56%) cN1 patients were ypN+. One patient had a positive cLN in the absence of +SLNs. This patient underwent axillary lymph node dissection (ALND); adjuvant treatment recommendations were unchanged. CONCLUSIONS: The cLN was a non-SLN in 19% of cN1 patients. cLN pathology did not impact adjuvant therapy recommendations, calling into question the utility of routinely clipping biopsied lymph nodes.

Persistent inter-observer variability of breast density assessment using BI-RADS® 5th edition guidelines.

OBJECTIVES: Due to most states' legislation, mammographic density categorization has potentially far-reaching implications, but remains subjective based on BIRADS® guidelines. We aimed to determine 1) effect of BI-RADS® 5th edition (5th-ed) vs 4th-edition (4th-ed) guidelines on reader agreement regarding density assessment; 2) 5th-ed vs 4th-ed density distribution, and visual vs quantitative assessment agreement; 3) agreement between experienced vs less experienced readers. METHODS: In a retrospective review, six breast imaging radiologists (BIR) (23-30 years' experience) visually assessed density of 200 screening mammograms performed September 2012-January 2013 using 5th-ed guidelines. Results were compared to 2016 data of the same readers evaluating the same mammograms using 4th-ed guidelines after a training module. 5th-ed density categorization by seven junior BIR (1-5 years' experience) was compared to eight experienced BIR. Nelson et al.'s kappas (κm, κw), Fleiss' κF, and Cohen's κ were calculated. Quantitative density using Volpara was compared with reader assessments. RESULTS: Inter-reader weighted agreement using 5th-ed is moderately strong, 0.73 (κw, s.e. = 0.01), similar to 4th-ed, 0.71 (κw, s.e. = 0.03). Intra-reader Cohen's κ is 0.23-0.34, similar to 4th-ed. Binary not-dense vs dense categorization, using 5th-ed results in higher dense categorization vs 4th-ed (p < 0.001). 5th-ed density distribution results in higher numbers in categories B/C vs 4th-ed (p < 0.001). Distribution for 5th-ed does not differ based on reader experience (p = 0.09). Reader vs quantitative weighted agreement is similar (5th-ed, Cohen's κ = 0.76-0.85; 4th-ed, Cohen's κ = 0.68-0.83). CONCLUSION: There is persistent subjectivity of visually assessed mammographic density using 5th-ed guidelines; experience does not correlate with better inter-reader agreement.

Methods of assessing categorical agreement between correlated screening tests in clinical studies.

Advances in breast imaging and other screening tests have prompted studies to evaluate and compare the consistency between experts' ratings of existing with new screening tests. In clinical settings, medical experts make subjective assessments of screening test results such as mammograms. Consistency between experts' ratings is evaluated by measures of inter-rater agreement or association. However, conventional measures, such as Cohen's and Fleiss' kappas, are unable to be applied or may perform poorly when studies consist of many experts, unbalanced data, or dependencies between experts' ratings exist. Here we assess the performance of existing approaches including recently developed summary measures for assessing the agreement between experts' binary and ordinal ratings when patients undergo two screening procedures. Methods to assess consistency between repeated measurements by the same experts are also described. We present applications to three large-scale clinical screening studies. Properties of these agreement measures are illustrated via simulation studies. Generally, a model-based approach provides several advantages over alternative methods including the ability to flexibly incorporate various measurement scales (i.e. binary or ordinal), large numbers of experts and patients, sparse data, and robustness to prevalence of underlying disease.

Lesion conspicuity on synthetic screening mammography compared to full field digital screening mammography.

OBJECTIVE: To compare lesion conspicuity on synthetic screening mammography (SM) plus digital breast tomosynthesis (DBT) versus full field digital mammography (FFDM) plus DBT. MATERIALS AND METHODS: Seven breast imagers each prospectively evaluated 107-228 screening mammograms (FFDM, DBT, and SM; total 1206 examinations) over 12 weeks in sets of 10-50 consecutive examinations. Interpretation sessions alternated as follows: SM + DBT, then FFDM, or FFDM + DBT, then SM. Lesion conspicuity on SM versus FFDM (equal/better versus less) was assessed using proportions with 95% confidence intervals. DBT-only findings were excluded. RESULTS: Overall 1082 of 1206 (89.7%) examinations were assessed BI-RADS 1/2, and 124 of 1206 (10.3%) assessed BI-RADS 0. There were 409 evaluated findings, including 134 masses, 119 calcifications, 72 asymmetries, 49 architectural distortion, and 35 focal asymmetries. SM conspicuity compared to FFDM conspicuity for lesions was rated 1) masses: 77 (57%) equal or more conspicuous, 57 (43%) less conspicuous; 2) asymmetries/focal asymmetries: 61 (57%) equal or more conspicuous, and 46 (43%) less conspicuous; 3) architectural distortion: 46 (94%) equal or more conspicuous, 3 (6%) less conspicuous; 4) calcifications: 115 (97%) equal or more conspicuous, 4 (3%) less conspicuous. SM had better conspicuity than FFDM for calcifications and architectural distortion and similar conspicuity for most masses and asymmetries. CONCLUSION: Compared to FFDM, SM has better conspicuity for calcifications and architectural distortion and similar conspicuity for most masses and asymmetries.

Training Nonradiologist Clinicians in Diagnostic Breast Ultrasound in Rural Rwanda: Impact on Knowledge and Skills.

OBJECTIVE: To assess the effectiveness of diagnostic breast ultrasound training provided for general practitioners and nurses in Rwanda via intensive in-person and subsequent online supervision and mentorship. METHODS: Four breast radiologists from Brigham and Women's Hospital trained two general practitioner physicians and five nurses in Rwanda over 9 total weeks of in-person training and 20 months of remote mentorship using electronic image review with emailed feedback. Independently recorded assessments were compared to calculate the sensitivity and specificity of trainee assessments, with radiologist assessments as the gold standard. We compared performance in the first versus second half of the training. RESULTS: Trainees' performance on written knowledge assessments improved after training (57.7% versus 98.1% correct, P = .03). Mean sensitivity of trainee-performed ultrasound for identifying a solid breast mass was 90.6% (SD 4.2%) in the first half of the training (period 1) and 94.0% (SD 6.7%) in period 2 (P = .32). Mean specificity was 94.7% (SD 5.4%) in period 1 and 100.0% (SD 0) in period 2 (P = .10). Mean sensitivity for identifying a medium- or high-suspicion solid mass increased from 79.2% (SD 11.0%) in period 1 to 96.3% (SD 6.4%) in period 2 (P = .03). Specificity was 84.4% (SD 15.0%) in period 1 and 96.7% (SD 5.8%) in period 2 (P = .31). DISCUSSION: Nonradiologist clinicians (doctors and nurses) in a rural sub-Saharan African hospital built strong skills in diagnostic breast ultrasound over 23 months of combined in-person training and remote mentorship. The sensitivity of trainees' assessments in identifying masses concerning for malignancy improved after sustained mentorship. Assessment of impact on patient care and outcomes is ongoing.

Epigenetic Reprogramming of Cancer-Associated Fibroblasts Deregulates Glucose Metabolism and Facilitates Progression of Breast Cancer.

The mechanistic contributions of cancer-associated fibroblasts (CAFs) in breast cancer progression remain to be fully understood. While altered glucose metabolism in CAFs could fuel cancer cells, how such metabolic reprogramming emerges and is sustained needs further investigation. Studying fibroblasts isolated from patients with benign breast tissues and breast cancer, in conjunction with multiple animal models, we demonstrate that CAFs exhibit a metabolic shift toward lactate and pyruvate production and fuel biosynthetic pathways of cancer cells. The depletion or suppression of the lactate production of CAFs alter the tumor metabolic profile and impede tumor growth. The glycolytic phenotype of the CAFs is in part sustained through epigenetic reprogramming of HIF-1α and glycolytic enzymes. Hypoxia induces epigenetic reprogramming of normal fibroblasts, resulting in a pro-glycolytic, CAF-like transcriptome. Our findings suggest that the glucose metabolism of CAFs evolves during tumor progression, and their breast cancer-promoting phenotype is partly mediated by oxygen-dependent epigenetic modifications.

A Randomized Cross-Over Trial Focused on Clinical Breast Exam Skill Acquisition Using High Fidelity versus Low Fidelity Simulation Models in Rwanda.

OBJECTIVE: Breast cancer incidence is rising for women in low and middle income country (LMIC)s. Growing the health care workforce trained in clinical breast exam (CBE) is critical to mitigating breast cancer globally. We developed a CBE simulation training course and determined whether training on a low-fidelity (LF) simulation model results in similar skill acquisition as training on high-fidelity (HF) models in Rwanda. DESIGN: A single-center randomized educational crossover trial was implemented. A preintervention baseline exam (exam 1), followed by a lecture series (exam 2), and training sessions with assigned simulation models was implemented (exam 3)-participants then crossed over to their unassigned model (exam 4). The primary outcome of this study determined mean difference in CBE exam scores between HF and LF groups. Secondary outcomes identified any provider level traits and changes in overall scores. SETTING: The study was implemented at the University Teaching Hospital, Kigali (CHUK) in Rwanda, Africa from July 2014 to March 2015 PARTICIPANTS: Medical students, residents in surgery, obstetrics and gynecology, and internal medicine residents participated in a 1-day CBE simulation training course. RESULTS: A total of 107 individuals were analyzed in each arm of the study. Mean difference in exam scores between HF and LF models in exam 1 to 4 was not significantly different (exam 1 0.08 standard error (SE) = 0.47, p = 0.42; exam 2 0.86, SE = 0.69, p = 0.16; exam 3 0.03, SE = 0.38, p = 0.66; exam 4 0.10 SE = 0.37, p = 0.29). Overall exam scores improved from pre- to post-intervention. CONCLUSIONS: Mean difference in exams scores were not significantly different between participants trained with HF versus LF models. LF models can be utilized as cost effective teaching tools for CBE skill acquisition, in resource poor areas.

A Randomized Cross-Over Trial Focused on Breast Core Needle Biopsy Skill Acquisition and Safety Using High Fidelity Versus Low Fidelity Simulation Models in Rwanda.

OBJECTIVE: Breast cancer is the most common cancer diagnosed in low and middle-income countries. Growing the number of health care personnel trained in diagnostic procedures like breast core needle biopsy (BCNB) is critical. We developed a BCNB simulation-training course that evaluated skill acquisition, confidence, and safety, comparing low-cost low fidelity (LF) models to expensive high fidelity (HF) models. DESIGN: A single-center randomized education crossover trial was implemented. Participants were randomized to HF or LF groups. A preintervention baseline exam followed by lectures and training sessions with a HF or LF model was implemented. A postintervention simulation exam was conducted, and participants crossed over to the other simulation model. SETTING: The study was implemented at the University Teaching Hospital, Kigali (CHUK) in Rwanda, Africa from October 2014 to March 2015. PARTICIPANTS: Residents training in surgery or obstetrics and gynecology participated in a 1-day BCNB training course. RESULTS: A total of 36 residents were analyzed, 19 in the HF arm and 17 in the LF arm. Mean difference in exam scores for HF and LF groups in the baseline exam (exam 1) (0.067, p = 0.94, standard error [SE] of 1.57) postintervention exam (exam 2) (1.85, SE 1.46, p = 0.33), and the crossover exam (exam 3) (4.39, SE = 1.90, p = 0.11) were not significantly different between HF and LF. Overall exam scores improved from pre- to postintervention. CONCLUSIONS: Our results indicate that mean difference in exams scores were not significantly different between residents trained with HF versus LF models. In resources poor areas-LF models can be utilized as effective teaching tools for skill acquisition for diagnostic surgical procedures.

Optimizing Success and Avoiding Mishaps in the Most Difficult Image-guided Breast Biopsies.

Breast cancer is an increasing challenge in developed and limited resource areas of the world. Early detection of breast cancer offers the best chance for optimal care and best outcomes. A critical step in early detection is to obtain efficient and accurate tissue diagnoses. Although image-guided core needle breast biopsies are usually straightforward for experienced breast imagers, there are some not uncommon scenarios that present particular challenges. In this review article we will discuss these difficult situations and offer our tried and true methods to ensure safe and successful biopsies, while using stereotactic, ultrasound, and MRI guidance.

Avoiding Pitfalls, Maximizing Success at Image-guided Breast Interventions: A Pictorial Review.

Imaging and image-guided interventions have become increasingly important in the workup and treatment of breast lesions in the past 2 decades. Radiologists should be aware of potential pitfalls during the workup, the procedure itself, and in the postprocedure follow-up. In this pictorial review, we illustrate challenges related to technique and interpretation related to breast interventions, and suggest ways to maximize success.

Effect of Training on Qualitative Mammographic Density Assessment.

PURPOSE: The aim of this study was to evaluate the accuracy of visual mammographic breast density assessment and determine if training can improve this assessment, to compare the accuracy of qualitative density assessment before and after training with a quantitative assessment tool, and to evaluate agreement between qualitative and quantitative density assessment methods. METHODS: Consecutive screening mammograms performed over a 4-month period were visually assessed by two study breast radiologists (the leads), who selected 200 cases equally distributed among the four BI-RADS density categories. These 200 cases were shown to 20 other breast radiologists (the readers) before and after viewing a training module on visual density assessment. Agreement between reader assessment and lead radiologist assessment was calculated for both reading sessions. Quantitative volumetric density of the 200 mammograms, determined using a commercially available tool, was compared with both sets of reader assessment and with lead radiologist assessment. RESULTS: Compared with lead radiologist assessment, reader accuracy of breast density assessment increased from 65% before training to 72% after training (odds ratio, 1.41; P < .0001). Training specifically improved assignment to BI-RADS categories 1 (P < .0001) and 4 (P < .10). Compared with quantitative assessment, reader accuracy showed statistically nonsignificant improvement with training (odds ratio, 1.1; P = .26). Substantial agreement between qualitative and quantitative breast density assessment was demonstrated (κ = 0.78). CONCLUSIONS: Training may improve the accuracy of mammographic breast density assessment. Substantial agreement between qualitative and quantitative breast density assessment exists.

The Breast Imager's Approach to Nonmammary Masses at Breast and Axillary US: Imaging Technique, Clues to Origin, and Management.

Ultrasonography (US) of the breast and axilla is primarily used to evaluate a symptomatic patient or to further investigate findings identified with other imaging modalities. Breast imagers are generally familiar with US evaluation of level I, II, and III axillary lymph nodes in the diagnosis and staging of breast cancer. However, the axilla contains nonlymphatic tissue as well, including muscle, fat, and vascular and neurologic structures, and anatomically the breast lies on the chest wall. Therefore, lesions of nonmammary and non-lymph node origin in the axilla or chest wall are not infrequently encountered during US evaluation of the breast or axilla. In fact, such lesions may be the reason that the patient presents to the breast imaging department for evaluation. Understanding the anatomy of the chest wall and axilla and using a systematic US approach will help radiologists expedite accurate diagnosis, suggest optimal additional imaging, and streamline appropriate clinical referral. Key imaging features of nonmammary non-lymph node masses are highlighted, and case examples are provided to illustrate these features. Appropriate patient management is critical in these cases because referral to a breast surgeon may not be the best next step. Depending on institutional referral patterns, other subspecialty surgeons will be involved. Online supplemental material is available for this article.

Recurrence of breast carcinoma as Paget disease of the skin at a prior core needle biopsy site: Case report and review of the literature.

INTRODUCTION: Core needle biopsy has become the preferred method of diagnosing breast carcinomas prior to definitive surgery. The possibility of displacing tumor cells into the needle track is a concern. PRESENTATION OF CASE: A 38 year old woman was diagnosed with right breast ductal carcinoma in situ (DCIS) with microinvasion by core needle biopsy. Bilateral skin sparing mastectomies with immediate autologous reconstruction were performed. One and a half years later the patient noted erythema and a scaling crust on the skin of the right breast that progressed over several months. Punch biopsy revealed Paget disease restricted to the epidermis. Subsequent comparison to initial clinical photographs confirmed the cancer was associated with the skin puncture site of the needle biopsy. The patient underwent complete excision with skin grafting and remains free of disease three years later. DISCUSSION: Only 13 cases of needle track recurrences have been reported. The majority presented as invasive carcinoma forming a subcutaneous mass. In the current case, detection was delayed due to not initially noting that a skin lesion was at the puncture site of the original needle biopsy. This is the only case of recurrence as tumor limited to the epidermis. CONCLUSION: Although recurrence in a needle track occurs very infrequently, clinicians should be aware of this phenomenon and investigate any changes, particularly when occurring at a needle biopsy site. Recording the skin puncture site can aid in early detection of recurrences. Recognition of a recurrence is important for prompt treatment and optimal prognosis.