Suspicious Mammograms Research Articles

Architectural distortion on digital breast tomosynthesis mammograms in symptomatic breast clinics: what are the result outcomes?

In April 2020, standard two-dimensional (2D) full-field digital mammograms were replaced with digital breast tomosynthesis (DBT) and synthesised 2D views for symptomatic breast clinics. This study aimed to evaluate the positive predictive value (PPV) for malignancy in DBT-detected Architectural distortion (AD). All mammogram reports with the word "distortion" were assessed between April 2020 and October 2022. There were 458 mammograms with the word "distortion." After excluding mammograms with no distortion (n = 128), post-surgical distortion (n = 173), distortion with mass (n = 33), and unchanged distortion (n = 14), there were 111 patients with pure distortion. Correlation with histopathology was obtained where possible. All patients were followed for a minimum of 2 years. Forty-two out of 111 patients (37.84%) with AD had a normal ultrasound (US) and were discharged. Fifty-five (49.5%) patients had sonographic correlation corresponding to the distortion, leading to US-guided biopsy. Thirteen (23.6%) had tomosynthesis-guided biopsy, and one had a skin biopsy. The PPV for malignancy was 42.34%. Malignancy diagnoses were higher with US-guided biopsies than tomosynthesis-guided biopsies, 78.1% and 30%, respectively. With a total malignancy rate of 42.34%, DBT-detected AD has a high enough PPV for malignancy to justify selective tissue sampling if a sonographic correlate is present or with suspicious mammograms. The chances of malignancy are higher when a sonographic correlate corresponding to AD is present. AD on DBT/synthesized mammograms views in symptomatic breast clinic patients justifies selective sampling.

Effect of age of initiation of mammography breast cancer screening.

Mammography is an excellent resource to reduce the burden of premature death associated with breast cancer; however, screening is only recommended between the ages of 50 and 69 years. To compare the frequency of suspicious and non-diagnostic mammograms for breast cancer when screening Chilean women between the ages of 40 and 50 years. Cross-sectional study nested in a breast cancer screening programme in Chilean women >40 years old, conducted between 2017 and 2021. Demographic variables and risk factors are described. To establish the effect of age on screening, we calculated the number needed to screen for a Breast Imaging Reporting and Data Reporting System 4 or 5 mammogram when screening over 40 years or over 50 years. We studied 137,690 women screened for breast cancer since the age of 40 years. The median age was 54 years (range 40-93 years). 64.7% of women were postmenopausal, 4.79% were nulliparous and 14% of post-menopausal women were receiving hormone replacement therapy. To find a suspicious mammogram, 170 women over 40 years and 149 women over 50 years would have to be screened. By changing the starting age of screening from 50 to 40 years 21 more women would have to be screened to detect a suspicious mammogram and if screened from age 50 and not from age 40 years 21% of total suspicious mammograms would remain unidentified.

Abstract P1-05-02: Reducing Delays of Breast Cancer Care Using An Innovative Digital Health Platform

Abstract Background: For breast cancer patients, prolonged time from core needle biopsy to initiation of treatment is associated with an incremental decrease in overall survival. This overall survival reduction disproportionately affects patients with stage I and II breast cancer (hazard ratio: 1.09-1.16, p< 0.001), the subgroup of breast cancer patients for which a survival reduction would be least expected due to its early stage and least accepted due to its large population size and potential increased cost of healthcare. Materials and Methods: We conducted an IRB-approved quality improvement initiative among patients with Medicare Advantage at a large integrated healthcare organization in Nevada to assess the ability of an innovative digital health platform to reduce the time interval between positive core needle biopsy to initiation of breast cancer treatment compared to a historical cohort of Medicare Advantage patients diagnosed and treated at the same facility over the preceding 6-month period. The study was restricted to women evaluated at a single contracted breast imaging center for inconclusive (Bi-RADS 0) and suspicious (BI-RADS 4 & 5) mammograms, ultrasound, and/or breast MRI. Time to treatment (TTT) calculations were limited to the subset of patients diagnosed with in situ and invasive breast cancer by core needle biopsy. The primary goal was to reduce the historical TTT by greater than 50%. A secondary goal was to reduce TTT to < 30 days, well below the interval where overall survival would be adversely affected. Results: Between September 2021 and April 2022, 552 patients with BI-RADS 0, 4, and 5 breast imaging were enrolled in the quality improvement initiative and managed on the XpediteMD digital health platform. 208 patients were designated as BI-RADS 4 or 5 (i.e., suspicious imaging findings), all of whom ultimately underwent a diagnostic core needle biopsy. Of these, 75 patients were found to have invasive or in situ breast cancer and were subsequently referred to surgery or medical oncology where treatment was initiated. Among these patients, average TTT was 32 days with use of the digital health platform compared to 74 days in the historical cohort, a statistically significant 57% reduction in TTT, exceeding the 50% TTT goal. Although the study did not meet the target goal of average TTT < 30 days from needle biopsy to initiation of treatment, failure to achieve this goal was largely attributable to 3 patients who elected to delay cancer therapy for personal or health reasons. Excluding these 3 patients would have yielded an average TTT of 28.75 days, which is within our < 30-day goal. Conclusions: This quality improvement initiative demonstrated the successful deployment of a novel digital health platform which achieved a 57% reduction in the time interval from a positive core needle biopsy to performance of breast cancer surgery or initiation of systemic therapy. By reducing TTT to 32 days, the initiative eliminated any delay in initiation of cancer therapy that could have compromised patients’ overall survival. Citation Format: Dennis R. Holmes, Riya Pulicharam. Reducing Delays of Breast Cancer Care Using An Innovative Digital Health Platform [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P1-05-02.

Integrating age, BMI, and serum N-glycans detected by MALDI mass spectrometry to classify suspicious mammogram findings as benign lesions or breast cancer

While mammograms are the standard tool for breast cancer screening, there remains challenges for mammography to effectively distinguish benign lesions from breast cancers, leading to many unnecessary biopsy procedures. A blood-based biomarker could provide a minimally invasive supplemental assay to increase the specificity of breast cancer screening. Serum N-glycosylation alterations have associations with many cancers and several of the clinical characteristics of breast cancer. The current study utilized a high-throughput mass spectrometry workflow to identify serum N-glycans with differences in intensities between patients that had a benign lesion from patients with breast cancer. The overall N-glycan profiles of the two patient groups had no differences, but there were several individual N-glycans with significant differences in intensities between patients with benign lesions and ductal carcinoma in situ (DCIS). Many N-glycans had strong associations with age and/or body mass index, but there were several of these associations that differed between the patients with benign lesions and breast cancer. Accordingly, the samples were stratified by the patient’s age and body mass index, and N-glycans with significant differences between these subsets were identified. For women aged 50–74 with a body mass index of 18.5–24.9, a model including the intensities of two N-glycans, 1850.666 m/z and 2163.743 m/z, age, and BMI were able to clearly distinguish the breast cancer patients from the patients with benign lesions with an AUROC of 0.899 and an optimal cutoff with 82% sensitivity and 84% specificity. This study indicates that serum N-glycan profiling is a promising approach for providing clarity for breast cancer screening, especially within the subset of healthy weight women in the age group recommended for mammograms.

Breast cancer: Three‐class masses classification in mammograms using Apriori dynamic selection

SummaryMammography is a commonly used screening technique for early diagnosis of breast cancer. However, the early detection of abnormalities remains challenging, particularly for dense breast categories. In this context, the automated classification of breast masses assists radiologists in their diagnosis and give them a second opinion. In this paper, we propose a machine learning‐based method for the classification of breast masses. First, the shape and texture features are extracted from the suspicious mammogram patches. These features are then fed to the Principal Component Analysis (PCA) to keep the relevant features only and are classified using the Support Vector Machine (SVM) and Random Forest (RF). Lastly, the Apriori dynamic selection method is applied for the final test predictions using the appropriate classifier for each test sample. The classification of breast masses patches into normal and abnormal attains accuracy of 96.43%, F1‐score of 95.76%, precision of 96.29%, recall of 95.27%, specificity of 96.43%, and AUC of 0.963. Whereas the one‐stage multi‐classification of breast masses into normal, benign, and malignant achieves accuracy of 75.81%, F1‐score of 76.47%, precision of 76.85%, recall of 78.77%, specificity of 87.91%, and AUC of 0.829.

Abstract 189: Evaluating the influence of Lactobacillus rhamnosus-derived extracellular vesicles on calcium deposition in early breast cancer

Abstract Microcalcifications are an important mammographic feature of ductal carcinoma in situ (DCIS) and are detected in 85-90% of abnormal mammograms. Calcium transport and calcium channels regulate microcalcifications and are essential for breast cancer proliferation. The breast duct intraductal lumen, which is the origin of ductal carcinomas, has been shown to be a non-sterile environment. We hypothesize that the intraductal microbiome influences calcium handling and mineral deposition in breast cancer. Bacteria have been known to enhance the formation of mineral deposits, including calcium salts, in various non mammalian microenvironments via secreted factors or extracellular vesicle mediated nucleation. We performed a clinical study on a cohort of 150 women who received a suspicious mammogram and were subjected to a follow up biopsy, and we identified differences in calcium handling microbiota between patients who harbored a malignant disease with respect to benign controls. To mechanistically study the role of microbiota in breast cancer calcium handling, we developed a 3D spheroid model of ductal carcinoma in situ using the BT-474 cell line. When exposed to high concentrations of calcium in the culture medium (from 1.75 mM to 2.05 mM), spheroids developed calcium mineral deposits in a calcium dose dependent manner; deposits were identified in internal regions of the spheroid, which correspond to hypoxic areas of DCIS in vivo. Extracellular calcium deposition correlated with cell survival in face of a toxic calcium level. Spheroids were cultured under a continuous flow of Lactobacillus rhamnosus LMS2-1-derived EVs over the course of two weeks. EVs were isolated by sequential ultracentrifugation; the 100K fraction was retained and characterized in terms of size, concentration, and protein content. EV treatment induced increased calcium export and mineral deposition in a dose dependent manner. Our findings support the hypothesis that L. rhamnosus-derived extracellular vesicles influence calcium regulation and mineral deposition in the hypoxic breast cancer milieu. More studies will be undertaken in the future to unveil molecular mechanisms that underlie this interaction, both at the cancer cell and the microbiota level. Citation Format: Ngoc Bao Vuong, Melany Alomia, Intisar Alruwaili, Fahad Alsaab, James L. Erickson, Weidong Zhou, Alessandra Luchini, Monique Van Hoek, Virginia Espina, Richard A. Hoefer, Lance Liotta. Evaluating the influence of Lactobacillus rhamnosus-derived extracellular vesicles on calcium deposition in early breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 189.

An Artificial Intelligence-based Mammography Screening Protocol for Breast Cancer: Outcome and Radiologist Workload.

Background Developments in artificial intelligence (AI) systems to assist radiologists in reading mammograms could improve breast cancer screening efficiency. Purpose To investigate whether an AI system could detect normal, moderate-risk, and suspicious mammograms in a screening sample to safely reduce radiologist workload and evaluate across Breast Imaging Reporting and Data System (BI-RADS) densities. Materials and Methods This retrospective simulation study analyzed mammographic examination data consecutively collected from January 2014 to December 2015 in the Danish Capital Region breast cancer screening program. All mammograms were scored from 0 to 10, representing the risk of malignancy, using an AI tool. During simulation, normal mammograms (score < 5) would be excluded from radiologist reading and suspicious mammograms (score > recall threshold [RT]) would be recalled. Two radiologists read the remaining mammograms. The RT was fitted using another independent cohort (same institution) by matching to the radiologist sensitivity. This protocol was further applied to each BI-RADS density. Screening outcomes were measured using the sensitivity, specificity, workload, and false-positive rate. The AI-based screening was tested for noninferiority sensitivity compared with radiologist screening using the Farrington-Manning test. Specificities were compared using the McNemar test. Results The study sample comprised 114 421 screenings for breast cancer in 114 421 women, resulting in 791 screen-detected, 327 interval, and 1473 long-term cancers and 2107 false-positive screenings. The mean age of the women was 59 years ± 6 (SD). The AI-based screening sensitivity was 69.7% (779 of 1118; 95% CI: 66.9, 72.4) and was noninferior (P = .02) to the radiologist screening sensitivity of 70.8% (791 of 1118; 95% CI: 68.0, 73.5). The AI-based screening specificity was 98.6% (111 725 of 113 303; 95% CI: 98.5, 98.7), which was higher (P < .001) than the radiologist specificity of 98.1% (111 196 of 113 303; 95% CI: 98.1, 98.2). The radiologist workload was reduced by 62.6% (71 585 of 114 421), and 25.1% (529 of 2107) of false-positive screenings were avoided. Screening results were consistent across BI-RADS densities, although not significantly so for sensitivity. Conclusion Artificial intelligence (AI)-based screening could detect normal, moderate-risk, and suspicious mammograms in a breast cancer screening program, which may reduce the radiologist workload. AI-based screening performed consistently across breast densities. © RSNA, 2022 Online supplemental material is available for this article.

Biomarkers for predicting cancer in women with a suspicious mammogram.

e15533 Background: Biomarker identification for early breast cancer diagnosis is confounded by comparing healthy control patients to patients undergoing surgical procedures and stress of a potential cancer diagnosis. We implemented a clinical research protocol that combines biomarker harvesting and identification with Breast Imaging-Reporting and Data System (BIRADS) results, within a cohort of women with a suspicious mammogram who donated samples prior to biopsy. The primary goals were to discover candidate novel plasma markers for stage I breast cancer versus benign lesions, and validate the markers by mass spectrometry and immunohistochemistry. Methods: 150 women found on screening mammography to have a BIRADS IV or V mammographic abnormality were enrolled in the IRB approved study, with one year follow-up. After informed consent, serum, plasma, and saliva specimens were obtained and frozen. The patient underwent image guided biopsy of the mammographic abnormality. Hydrogel nanoparticles were used to harvest and concentrate low abundance protein biomarkers from plasma. Proteins were identified by mass spectrometry. The BIRADS score and biopsy outcome were blinded to the laboratory researchers. Results: 37/150 women (median age 64, 73% ER+, 70% PR+) were diagnosed with biopsy-proven breast cancer. 15/37 had a family history of breast cancer. Within the context of stress of an abnormal mammogram and invasive biopsy, we identified 5478 plasma peptides. A model to predict endpoints that discriminate cancer vs no cancer was developed using cross-validation and lasso shrinkage method. The best fit multi-analyte ROC/AUC model of peptide spectral matches revealed 10 candidate peptides, including PLAA, TRAPPC9, PROS1, DDX41, ANKRD63, EGFLAM (AUC = 0.81), that discriminated cancer versus no cancer. The functional mechanisms of these proteins are calcium metabolism, GPI anchor biosynthesis, neural-immune crosstalk, DNA repair, and ubiquitin-mediated protein trafficking. Conclusions: Molecular profiling of blood can potentially complement imaging to improve diagnostic specificity in the setting of a suspicious mammogram. This unique trial design, enhanced by nanotechnology protein harvesting, identified potential novel cancer biomarkers in the presence of a suspicious mammogram. A confirmation trial is underway.

Abstract P2-02-03: Assessment of an immune response panel of serum protein biomarkers for the non-invasive detection of breast cancer

Abstract Background: Breast cancer screening guidelines (ACS) recommend annual mammography for asymptomatic women ages 45 to 54 and once every two years for women age 55 and older. Women with suspicious screening mammograms are recommended for a diagnostic mammogram and may also undergo MRI or ultrasound. Ultimately, unresolved suspicious findings frequently result in the recommendation of a breast biopsy. Approximately 10% of suspicious diagnostic mammograms are recommended for breast biopsies and a large majority of these biopsies yield benign results. We conducted a screen for serum protein biomarkers and identified a novel panel for the non-invasive detection of breast cancer with the goal of developing a diagnostic test that can reduce the number of patients with benign pathology undergoing invasive biopsies. Methods: Serum samples were collected at 4 US sites from women with suspicious diagnostic mammogram findings (primarily BI-RADS category 4 and breast composition b/c) undergoing biopsy for evaluation of a potential malignancy. Serum samples from 136 patients (87 benign pathology and 49 malignant pathology) were evaluated on the olink® Proteomics Immune Response Panel (92 analytes). Statistical screening methodologies, such as individual t-tests with control for false discovery, were used to identify markers with the potential to distinguish benign from malignant pathology. The candidate markers were further studied and combined using generalized linear modeling to develop potential diagnostic models. Results: A 19-marker model resulted in an AUC of 0.94 with a sensitivity of 90% and a specificity of 80%. A 12-marker model resulted in an AUC of 0.93, yielding a sensitivity of 90% with a specificity of 77%. Conclusions: This study reveals a novel panel of serum protein biomarkers that may allow for the non-invasive and sensitive detection of breast cancer in patients presenting with suspicious findings on mammography, thus reducing the need for invasive biopsies. Citation Format: McQuary PR, Copeland KAF, Sheibani NM, Friedman LA, Korn RL, Fiorica JV, Lourenco AP, Struthers SE, Hesterberg LK. Assessment of an immune response panel of serum protein biomarkers for the non-invasive detection of breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P2-02-03.

Factors affecting the under-diagnosis of atypical ductal hyperplasia diagnosed by core needle biopsies – A 10-year retrospective study and review of the literature

IntroductionDue to the possibility of underestimation, surgical excision is usually offered to patients with atypical ductal hyperplasia (ADH) diagnosed with core needle biopsy (CNB). Here we review the 10-year data of patients with ADH diagnosed by CNB, aiming to identify the factors associated with under-diagnosis. MethodsRetrospective review of database from 1st Jan 2005 to 31st Dec 2014 was performed; patients with ADH diagnosed by CNB were identified. Diagnosis upgrade rate and its risk factors were evaluated. Results104 patients were found to have ADH on CNB, 101 patients received excisional biopsy while 3 patients refused operation. 34 patients had ductal carcinoma in situ (DCIS) after excision, 6 had invasive ductal carcinoma, 1 had lobular carcinoma in situ and 1 had angiosarcoma. CNB under-diagnosed up to 41.6% of malignant lesions. Breast mass on presentation and suspicious mammograms (BIRADS ≥ 4) are associated with diagnosis upgrade (P = 0.0005, 0.0001).Literature review of 39 studies between 1997 and 2017 revealed 3125 excision procedures performed for ADH diagnosed by CNB, the pooled median diagnosis upgrade rate was 25% (Range 4–54%). ConclusionWe recommend excision in all patients with ADH diagnosed by CNB, especially in patients with suspicious mammographic features.

Abstract P5-03-05: Development of a panel of serum-based protein biomarkers for the non-invasive detection of breast cancer in BI-RADS category 4 patients

Abstract Background: Current breast cancer screening guidelines call for annual mammography for asymptomatic women age 45 to 54 and once every two years for women age 55 and older. Women with suspicious screening mammograms are recommended for a diagnostic mammogram and may also undergo MRI or ultrasound. Ultimately, suspicious findings unresolved by imaging typically result in the recommendation of a breast biopsy. Approximately 10% of suspicious diagnostic mammograms are recommended for breast biopsies and 67% to 95% of these biopsies yield negative results. With the goal of reducing the number of patients with benign pathology undergoing invasive biopsies, we conducted a screen for serum protein biomarkers and identified a novel panel for the non-invasive detection of breast cancer. Methods: Serum samples were collected at two sites from women with suspicious diagnostic mammogram findings (primarily BI-RADS category 4) undergoing biopsy for the evaluation of a potential malignancy. Serum samples from 100-patients (50 benign pathology and 50 malignant pathology) were evaluated on the SOMAscan Assay 1.3k, which measures levels of 1,310 different protein analytes. Statistical screening methodologies, such as individual t-tests with control for false discovery, were used to identify markers with the potential to distinguish benign from malignant pathology. The candidate markers were further studied and combined using generalized linear modeling to develop three potential diagnostic models. K-fold cross validation was used to guard against over fitting of the models. Results: A 15-marker model resulted in an AUC of 0.92 with a sensitivity of 90% and specificity of 76%. Two 6-marker models (with 4 markers in common) each resulted in AUC of 0.85, yielding a sensitivity of 90% with a specificity of 56% or 64%. Conclusions: This study reveals a novel panel of serum protein biomarkers that may allow for the non-invasive and sensitive detection of breast cancer in BI-RADS category 4 patients. A multicenter study is underway to further refine and validate this panel in a larger set of prospectively collected patient samples. Citation Format: Chapman KB, Copeland K, Kidd J, Qiu L, Sheibani N, Tam O, Friedman L, Korn R, Fiorica J, Lourenco A, Suthers S, Hesterberg L. Development of a panel of serum-based protein biomarkers for the non-invasive detection of breast cancer in BI-RADS category 4 patients [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P5-03-05.

Screening Mammography Outcomes in One Community-Based Initiative in Lebanon

Introduction and objectives: The outcomes of a screening mammography initiative implemented by a Lebanese NGO over several years are analyzed to assess 1. the socio-demographic and reproductive characteristics associated with repeating a screening mammography after receiving reassuring results from a first one, and 2, to estimate the proportion of suspicious mammograms resulting from a mass screening activity. Methods: A total of 2885 women participated in the screening initiative between 1997 and 2013, during which they were able to obtain repeated annual mammograms free of charge. Results: A steady significant decrease in mean age at first mammography was found from 2002 (59.1 years) to 2013 (48.4 years). Given enough time span, 44% repeated the test at least once after receiving reassuring results. Factors associated with test repetition were older age and retired employment status. Mammography results were suspicious (ACR4 or 5) in 2.3% of cases. Discussion and conclusions: The proportion of suspicious mammography results is close to those reported from countries with similar incidence rates of breast cancer. This suggests that mammography quality in Lebanon is acceptable globally, although gaps may be detected for specific centers. The percentage of annual test repetition (44%) was much lower than in other developing (Brazil: 90%) or developed countries (USA: 82%), even when financial barriers were cancelled. Among non-retired women, household worries and/or professional activities may be competing with self-care, compared to retired women. Mammography centers should be encouraged to adopt flexible opening hours to facilitate the access to working women.

Abstract P3-01-04: Mammographic changes after oncoplastic reduction mammoplasty

Abstract BACKGROUND: Reconstruction of partial lumpectomy defects with reduction mammaplasty techniques can improve aesthetic outcomes. However, the impact of the significant tissue rearrangement on post-operative mammographic findings and subsequent recommendations for biopsy has not been well-studied. METHODS: A retrospective review of 50 patients who underwent partial mastectomy with immediate oncoplastic reduction mammaplasty reconstruction from 2001 to 2008 was performed. Mammography reports at 6 months, 1 year, 2 years, and 3 years post-operatively were reviewed for Breast Imaging Reporting and Data System (BI-RADS) scores, predominant findings, and recommendations for subsequent imaging or biopsy. RESULTS: At six months post-operatively, 49 patients (98%) had benign findings of post-surgical changes, while one patient had microcalcifications and underwent subsequent surgical re-excision with residual DCIS on pathologic analysis. At one year, 94% of patients continued to have benign mammograms; of these mammograms, 94% reported only post-surgical scarring, while 6% described benign-appearing scattered or dystrophic calcifications. Of the three patients with suspicious mammograms at one year, all underwent core biopsies with benign results. At two years, rates of fat necrosis (2%) and scattered or dystrophic calcifications (10%) increased, though all mammographic findings were considered benign and none required additional imaging or biopsies. By three years post-operatively, an additional two patients (4%) developed suspicious findings and underwent biopsies confirming local recurrence (one invasive, one in situ). The remaining 96% continued to have benign mammographic findings and were subsequently followed with routine biannual mammographic surveillance. Overall, 88% of patients required no additional intervention in the three-year period following oncoplastic reduction mammaplasty. Of patients recommended for biopsy, malignancy was discovered in 50%, including two local recurrences (4% of total patients). This data is similar to a previously published large study (1841 women) of mammographic surveillance after partial mastectomy without oncoplastic reconstruction at our institution, which found a 6% malignancy rate at 5 years. CONCLUSIONS: Although substantial tissue rearrangement is performed at the time of oncoplastic reduction mammaplasty, our results demonstrate low rates of abnormal post-operative mammograms and subsequent biopsies over the first three years following the procedure. These findings support the use of oncoplastic reduction mammaplasty as a strategy for improving reconstructive outcomes in patients undergoing partial mastectomy. Citation Format: Anne Warren Peled, Merisa Piper, Laura J Esserman, Robert D Foster, Hani Sbitany, Elissa R Price. Mammographic changes after oncoplastic reduction mammoplasty [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-01-04.

Multi-scale textural feature extraction and particle swarm optimization based model selection for false positive reduction in mammography

The high number of false positives and the resulting number of avoidable breast biopsies are the major problems faced by current mammography Computer Aided Detection (CAD) systems. False positive reduction is not only a requirement for mass but also for calcification CAD systems which are currently deployed for clinical use. This paper tackles two problems related to reducing the number of false positives in the detection of all lesions and masses, respectively. Firstly, textural patterns of breast tissue have been analyzed using several multi-scale textural descriptors based on wavelet and gray level co-occurrence matrix. The second problem addressed in this paper is the parameter selection and performance optimization. For this, we adopt a model selection procedure based on Particle Swarm Optimization (PSO) for selecting the most discriminative textural features and for strengthening the generalization capacity of the supervised learning stage based on a Support Vector Machine (SVM) classifier. For evaluating the proposed methods, two sets of suspicious mammogram regions have been used. The first one, obtained from Digital Database for Screening Mammography (DDSM), contains 1494 regions (1000 normal and 494 abnormal samples). The second set of suspicious regions was obtained from database of Mammographic Image Analysis Society (mini-MIAS) and contains 315 (207 normal and 108 abnormal) samples. Results from both datasets demonstrate the efficiency of using PSO based model selection for optimizing both classifier hyper-parameters and parameters, respectively. Furthermore, the obtained results indicate the promising performance of the proposed textural features and more specifically, those based on co-occurrence matrix of wavelet image representation technique.

Abstract P4-03-09: Time points to diagnosis and treatment of invasive breast cancer in southwestern Ontario

Abstract Background Delays in breast cancer diagnosis and treatment are associated with increased tumour size at presentation, higher incidence of lymph node metastasis, higher relapse rates and lower 5-year survival rates in some studies. Diagnosis and treatment of breast cancer is complex, involving many health care practitioners at multiple locations. New advances in diagnostic and prognostic indicators make the process ever more complex. We reviewed the timing of crucial events in breast cancer diagnosis and treatment in Southwestern Ontario. Methods: Two time periods (2001, 2011) were assessed. Patients with new early invasive breast cancer with specific TNM criteria (T1c or greater plus any N, or any T plus N1-3; M0) were included. Data collection and analysis explored significant time points from first suspicion of malignancy, to diagnosis and definitive treatment, as well as patient demographic information. Results: 300 and 451 eligible patients were identified in 2001 and 2011 cohorts. Distribution of T and N status by cohort was compared by Chi-square test and time from first suspicion to diagnosis by Wilcoxon testing. The proportion of patients in which the time from first suspicion (clinical by lay person or health professional or by imaging) to pathological diagnosis exceeded two months increased from 20.4% in 2001 to 42.1% in 2011. There were no differences in time from first suspicion to diagnosis when analyzed by age (p = 0.54) or location (p = 0.50). Pathological T2-4 status at diagnosis increased from 48.4% in 2001 to 56.8% in 2011, and N2-3 status increased from 7.3% to 12.6%. Patients who had mammograms increased from 52.7% to 59.4%. A positive or suspicious mammogram was the first sign in 36% and 39.1% of cases. There was a trend towards more pathological diagnoses and definitive surgeries at tertiary centers compared to community hospitals. Conclusions: There is a longer time interval from first suspicion of malignancy to pathological diagnosis in 2011 compared to 2001 for both urban and rural populations. The number of non-low risk, non-metastatic cancers at diagnosis increased by 50% over the time interval studied (300 v 451), but number of T2-T4 non-metastatic cancers increased by 77% (145 v 256) and N2,3 cancers by 136% (22 v 52). This is despite more patients receiving mammograms (52.7% v 59.4%). Delays in breast cancer treatment are multifactorial, including both system-, and patient-related factors. This is an under-researched area and more investigation is needed to understand the reasons for the diagnostic delays and more serious cancers at presentation in order to improve outcomes. Reasons for higher proportions of advanced cancers will be discussed. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-03-09.

Using a new patient process to improve breast care access.

171 Background: Access to timely care is of major focus in oncology. Clinicians and administration collaborated to improve the training and support for non medical staff involved in the new patient process at the breast center. The goal of the process was to improve upon the surgical triage of patient appointments for those with suspicious mammograms or newly diagnosed with breast cancer in less than one week, use the nurse practitioner clinics most effectively, and to triage other diagnoses and patients per updated guidelines, thereby increasing access to care. Methods: A new patient intake form was created with the assistance of IT and added to the electronic medical record as a tool for use by clinical and nonclinical staff. Clinicians and administration also met to update the new patient process forming a decision tree, which ultimately was developed into a manual. A series of in-services were held with scheduling and clinical staff for education and training. Education was also provided to refer patients with complex needs to the nurse navigator. Outcomes were measured by clinical and administrative staff. Results: Access to surgery clinics has improved for new breast patients since the process updates were implemented. Patients can be seen an average of 3.5 days sooner for both malignant and benign diagnoses. Access to the nurse practitioner clinic also improved by 9.4 days. Patients that are diagnosed within the health system can be seen within 3 days, and those that are diagnosed outside the system, within 7 days. Conclusions: It was the goal to improve access to the breast center by updating the intake and scheduling guidelines to see patients with a new diagnosis of breast cancer or other breast problems more quickly. Having guidelines that can be followed by clinical and non clinical staff promotes communication and collaboration, as well as a consistent approach among staff. In the future, the continued refinement of the new patient process and ongoing collaboration will further improve access to care.

Mammographic Changes after Oncoplastic Reduction Mammaplasty

INTRODUCTION: Reconstruction of partial lumpectomy defects with reduction mammaplasty techniques can improve aesthetic outcomes. However, the impact of the significant tissue rearrangement on post-operative mammographic findings and subsequent recommendations for biopsy has not been well-studied. METHODS: A retrospective review of 50 patients who underwent partial mastectomy with immediate oncoplastic reduction mammaplasty reconstruction from 2001 to 2008 was performed. Mammography reports at 6 months, 1 year, 2 years, and 3 years post-operatively were reviewed for Breast Imaging Reporting and Data System (BI-RADS) scores, predominant findings, and recommendations for subsequent imaging or biopsy. RESULTS: At six months post-operatively, 49 patients (98%) had benign findings of post-surgical changes, while one patient had microcalcifications and underwent subsequent surgical re-excision with residual DCIS on pathologic analysis. At one year, 94% of patients continued to have benign mammograms; of these mammograms, 94% reported only post-surgical scarring, while 6% described benign-appearing scattered or dystrophic calcifications. Of the three patients with suspicious mammograms at one year, all underwent core biopsies with benign results. At two years, rates of fat necrosis (2%) and scattered or dystrophic calcifications (10%) increased, though all mammographic findings were considered benign and none required additional imaging or biopsies. By three years post-operatively, an additional two patients (4%) developed suspicious findings and underwent biopsies confirming local recurrence (one invasive, one in situ). The remaining 96% continued to have benign mammographic findings and were subsequently followed with routine biannual mammographic surveillance. Overall, 88% of patients required no additional intervention in the three-year period following oncoplastic reduction mammaplasty. Of patients recommended for biopsy, malignancy was discovered in 50%, including two local recurrences (4% of total patients). This data is similar to a previously published large study (1841 women) of mammographic surveillance after partial mastectomy without oncoplastic reconstruction at our institution, which found a 6% malignancy rate at 5 years.1 CONCLUSION: Although substantial tissue rearrangement is performed at the time of oncoplastic reduction mammaplasty, our results demonstrate low rates of abnormal post-operative mammograms and subsequent biopsies over the first three years following the procedure. These findings support the use of oncoplastic reduction mammaplasty as a strategy for improving reconstructive outcomes in patients undergoing partial mastectomy.

Mobile mammography in underserved populations: analysis of outcomes of 3,923 women.

Mobile health units are increasingly utilized to address barriers to mammography screening. Despite the existence of mobile mammography outreach throughout the US, there is a paucity of data describing the populations served by mobile units and the ability of these programs to reach underserved populations, address disparities, and report on outcomes of screening performance. To evaluate the association of variables associated with outcomes for women undergoing breast cancer screening and clinical evaluation on a mobile unit. Retrospective analysis of women undergoing mammography screening during the period 2008–2010. Logistic regression was fitted using generalized estimating equations to account for potential repeat annual visits to the mobile unit. In total, 4,543 mammograms and/or clinical breast exams were conducted on 3,923 women with a mean age of 54.6, 29 % of whom had either never been screened or had not had a screening in 5 years. Age < 50 years, lack of insurance, Hispanic ethnicity, current smoking, or having a family relative (<50 years of age) with a diagnosis of cancer were associated with increased odds of a suspicious mammogram finding (BIRADS 4,5,6). Thirty-one breast cancers were detected. The mobile outreach initiative successfully engaged many women who had not had a recent mammogram. Lack of insurance and current smoking were modifiable variables associated with abnormal screens requiring follow up.

PS1-12: Suspicious Mammogram (BIRADS 4) Outcome and Breast Biopsy: Preliminary Findings from a Cohort of 6,198 Women

Background Annually about 1,700,000 women undergo breast biopsies with an estimated cost of $3.5 billion. About 80% of women are diagnosed with benign conditions (BCs). Other concerns are psychological impact and possible complications of radiographic evaluation of future mammograms. A recent report in the New York Times described the high rate of breast biopsies as diagnostic overkill, drastic and expensive. We report the preliminary findings on pathologic outcomes and predictive probability of sub-classifications of Breast Imaging Reporting and Data System-4 (BIRADS 4).

Mammographic Findings after Intraoperative Radiotherapy of the Breast

Intraoperative Radiotherapy (IORT) is a form of accelerated partial breast radiation that has been shown to be equivalent to conventional whole breast external beam radiotherapy (EBRT) in terms of local cancer control. However, questions have been raised about the potential of f IORT to produce breast parenchymal changes that could interfere with mammographic surveillance of cancer recurrence. The purpose of this study was to identify, quantify, and compare the mammographic findings of patients who received IORT and EBRT in a prospective, randomized controlled clinical trial of women with early stage invasive breast cancer undergoing breast conserving therapy between July 2005 and December 2009. Treatment groups were compared with regard to the 1, 2 and 4-year incidence of 6 post-operative mammographic findings: architectural distortion, skin thickening, skin retraction, calcifications, fat necrosis, and mass density. Blinded review of 90 sets of mammograms of 15 IORT and 16 EBRT patients demonstrated a higher incidence of fat necrosis among IORT recipients at years 1, 2, and 4. However, none of the subjects were judged to have suspicious mammogram findings and fat necrosis did not interfere with mammographic interpretation.