Abstract: Anti–vascular endothelial growth factor (VEGF) injections in eyes with nonproliferative diabetic retinopathy (NPDR) without center-involved diabetic macular edema (CI-DME) reduce development of vision-threatening complications from diabetes over at least 2 years, but whether this treatment has a longer-term benefit on visual acuity is unknown. Objective To compare the primary 4-year outcomes of visual acuity and rates of vision-threatening complications in eyes with moderate to severe NPDR treated with intravitreal aflibercept compared with sham. The primary 2-year analysis of this study has been reported. Design, Setting, and Participants Randomized clinical trial conducted at 64 clinical sites in the US and Canada from January 2016 to March 2018, enrolling 328 adults (399 eyes) with moderate to severe NPDR (Early Treatment Diabetic Retinopathy Study [ETDRS] severity level 43-53) without CI-DME. Interventions Eyes were randomly assigned to 2.0 mg aflibercept (n = 200) or sham (n = 199). Eight injections were administered at defined intervals through 2 years, continuing quarterly through 4 years unless the eye improved to mild NPDR or better. Aflibercept was given in both groups to treat development of high-risk proliferative diabetic retinopathy (PDR) or CI-DME with vision loss. Main Outcomes and Measures Development of PDR or CI-DME with vision loss (≥10 letters at 1 visit or ≥5 letters at 2 consecutive visits) and change in visual acuity (best corrected ETDRS letter score) from baseline to 4 years. Results Among participants (mean age 56 years; 42.4% female; 5% Asian, 15% Black, 32% Hispanic, 45% White), the 4-year cumulative probability of developing PDR or CI-DME with vision loss was 33.9% with aflibercept vs 56.9% with sham (adjusted hazard ratio, 0.40 [97.5% CI, 0.28 to 0.57] ; P   & amp;lt; .001). The mean (SD) change in visual acuity from baseline to 4 years was −2.7 (6.5) letters with aflibercept and −2.4 (5.8) letters with sham (adjusted mean difference, −0.5 letters [97.5% CI, −2.3 to 1.3]; P  = .52). Antiplatelet Trialists’ Collaboration cardiovascular/cerebrovascular event rates were 9.9% (7 of 71) in bilateral participants, 10.9% (14 of 129) in unilateral aflibercept participants, and 7.8% (10 of 128) in unilateral sham participants. Conclusions and Relevance Among patients with NPDR but without CI-DME, at 4 years treatment with aflibercept vs sham, initiating aflibercept treatment only if vision-threatening complications developed, resulted in statistically significant anatomic improvement but no improvement in visual acuity. Aflibercept as a preventive strategy, as used in this trial, may not be generally warranted for patients with NPDR without CI-DME. Trial Registration ClinicalTrials.gov Identifier: NCT02634333

Abstract: Background: High percent mammographic density (MD), which reflects the relative fibroglandular tissue content of the breast, is one of the strongest breast cancer risk factors; however, the pathologic mediators of this risk are unknown. We hypothesize that analysis of breast tissue sections using deep learning approaches may characterize histologic features that underpin risk associated with high MD. Methods: Non-targeted H & E stained breast tissue sections of diagnostic image-guided breast biopsies were evaluated among 588 women enrolled following an abnormal mammogram in the Breast Radiology Evaluation and Study of Tissues (BREAST) Stamp Project (2007-2010). Overall volumetric percent MD for the biopsied breast and localized volumetric percent MD surrounding the biopsy site were determined for each participant. A deep convolutional neural network (CNN) model was trained to identify and quantitatively assess breast epithelial, stroma and fat tissue and their organizational and spatial arrangements. Least absolute shrinkage and selection operator (Lasso) regression was used to determine relationships between MD measures and pathological features. To ensure reliability of the model, a cross-validation strategy was employed to build and assess the performance of the fitted model. Finally, Spearman correlation coefficients were estimated to test the association between the predicted density values by each model (predicting overall or localized MD) and the actual MD measurements. We report the average and standard deviation (SD) of the correlation coefficients. Results: In an independent validation set, the CNN model was 95.5% accurate in classifying epithelial, stromal and fat tissue. The mean (SD) correlations between the predicted model and the actual measurements for overall and localized MD were 0.70 (0.06) and 0.65 (0.06) respectively. The amount of stroma identified (normalized to tissue area) had the highest selection probability (P-value) by the Lasso model and thus the strongest positive relationship with MD (P-value & gt;0.9 for each MD measurement). In contrast, the amount of normalized epithelial tissue was not related to MD (P-value=0.01 for each MD measurement). No association was observed for the total normalized fat area with MD (P-value & lt;0.31 for each MD measurement). In addition, the number of distributed epithelial regions was positively associated, whereas the distance between epithelial regions was inversely associated with overall MD (P-value & lt;0.87 and 0.62, respectively). Conclusions: These results show that greater stromal tissue amount and spatial distribution patterns of epithelial regions, rather than total epithelial amounts, had the strongest relationships with elevated MD. Future work will determine the relationship of these MD features with biopsy diagnosis. Citation Format: Maeve Mullooly, Babak Ehteshami Bejnordi, Maya Palakal, Pamela M. Vacek, Donald L. Weaver, John A. Shepherd, Bo Fan, Amir Pasha Mahmoudzadeh, Jeff Wang, Jason M. Johnson, Sally D. Herschorn, Brian L. Sprague, Ruth M. Pfeiffer, Louise A. Brinton, Mark E. Sherman, Andrew Beck, Gretchen L. Gierach. Application of convolutional neural networks to breast biopsies to uncover tissue correlates of mammographic breast density [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4235. doi:10.1158/1538-7445.AM2017-4235

Abstract: Introduction: The etiology of breast cancer remains an area of ongoing investigation. Improving our understanding of factors associated with breast cancer development will strengthen the utility of risk prediction strategies and improve risk stratification. Mammographic breast density remains one of the most influential breast cancer risk factors, with 4 to 6 fold elevated risk observed among women with the highest compared to the lowest levels. These associations have consistently been observed irrespective of the method (visual or automated), used to quantify breast density, showing the robustness of the associations. An understudied breast cancer risk factor is the bilateral asymmetry of mammographic features, with prior studies suggesting that women with higher levels of breast asymmetry may be at elevated breast cancer risk. Increasingly, studies are recognizing the potential of bilateral breast asymmetry defined by mammographic breast density in determining breast cancer risk. Recently, Eriksson and colleagues highlighted a strong influence of breast asymmetry within a model they developed to predict short-term breast cancer risk among women attending breast screening. They observed that the contribution to the model of asymmetry in mammographic breast density, microcalcifications, and masses between the breasts was as substantial as the total number of microcalcifications and masses found within a mammogram, indicating that differences between the breasts may be an important risk marker. To help further understand associations between breast cancer risk factors and asymmetry of breast density, particularly among women at elevated risk for breast cancer, we evaluated risk factor relationships with the bilateral asymmetry of volumetric measures of global and local breast density across the spectrum of premalignant and invasive breast cancer diagnoses. Methods: This study included 882 women enrolled as part of the National Cancer Institute's Breast Radiology Evaluation and Study of Tissues (BREAST)-Stamp Project (2007-2010). The BREAST-Stamp Project is a cross-sectional molecular epidemiologic study that aims to understand how novel breast density measures are related to breast cancer etiology. Women were enrolled if they were referred for diagnostic image-guided breast biopsy following an abnormal mammogram at the University of Vermont Medical Center, and had not previously been diagnosed or treated for cancer, undergone breast surgery within one year or received chemoprevention. Risk factor data were collected at study enrolment via interview and self-administered questionnaires. Breast density measures were estimated using Single X-ray Absorptiometry (SXA), a technique in which an SXA breast density phantom was affixed to the compression paddle of the mammography machine during routine mammography so that it was included in the X-ray field. It served as a reference standard to estimate volumetric breast density. Breast density was assessed using pre-biopsy craniocaudal full-field digital mammograms of both the ipsilateral (affected) and contralateral (unaffected) breast. Firstly, global density from each laterality was determined as percent fibroglandular volume (%FGV). Secondly, localized volumetric density measures were estimated following identification of the biopsy site on the ipsilateral pre-biopsy mammogram by the study radiologist and identifying the corresponding site on the contralateral mammogram. The SXA estimated %FGV in a perilesional volume twice the size of, but excluding, the biopsy target, centered at the biopsy site. Previous estimates of reproducibility for the SXA test phantoms demonstrated a repeatability SD of 2%, with a ±2% accuracy for the entire thickness and density ranges. Breast density asymmetry was defined as an absolute bilateral difference in measures when subtracting the breast density measures of the contralateral breast from the ipsilateral measures. Spearman's correlations (rho) examined associations between breast density measures of the left and right breasts. To determine relationships between breast cancer risk factors (defined as categorical variables) and measures of bilateral breast density asymmetry, analysis of covariance (ANCOVA) models (PROC GLM) were used. Analyses were conducted at the per woman level using SAS. Probability values of & lt;0.05 were considered statistically significant, and all tests were two-tailed. Results:We initially investigated asymmetry within each woman by examining correlations between breast density measures in the left and right breasts. Strong, positive correlations between the ipsilateral and contralateral breast were observed for each breast density measure (rho for global %FGV=0.89, p-value & lt;0.0001; rho for local %FGV=0.79, p-value & lt;0.0001). Breast asymmetry was observed in the majority of the study population. For global %FGV, 76% of women had a bilateral difference ≥2%, with 43% of women having higher %FGV in their ipsilateral affected breast and 33% having higher in their contralateral unaffected breast. For localized %FGV, the majority of women had differences between their two breasts (89%), with 61% of women having higher localized %FGV in their ipsilateral affected breast compared to the remaining 28% who had higher localized breast density in their unaffected breast. We next examined relationships between breast cancer risk factors and breast asymmetry. Overall, no associations were observed between any of the risk factors examined, including age, race, body mass index, education, menopausal status, menopausal hormone therapy use with absolute bilateral differences in global or localized %FGV. Among the study population, most women had a benign breast disease diagnosis, with 33% and 43% being diagnosed with benign non-proliferative and benign proliferative lesions, respectively. Of the study population, 15% were diagnosed with invasive breast cancer. Overall, no differences were observed in bilateral differences in global %FGV according to diagnosis; however, higher mean bilateral differences in localized %FGV were observed for women with invasive compared to other diagnoses including benign and in-situ lesions (p=0.056). Discussion and conclusions: Our findings showed that breast asymmetry, defined by bilateral differences in global and localized mammographic breast density, was mostly unrelated to breast cancer risk factors among women undergoing an image guided breast biopsy. We observed that localized measures, defined according to within-woman bilateral differences in localized %FGV surrounding a suspicious lesion as compared with localized %FVG in a comparable location in the contralateral breast, may be an indication of cancer. This investigation is currently ongoing and efforts are underway to replicate findings using evolving image analysis methods such as Volpara Density Maps, an FDA-approved breast imaging tool which provides volumetric estimates of local glandular tissue distribution. Further, the ascertainment of 10-year follow-up data within this study population is in progress, which will facilitate prospective investigations of the relationship between breast asymmetry and breast cancer risk. Ongoing work will also extend our understanding of whether breast asymmetry is related to individualized breast cancer risk, defined using tailored breast cancer risk assessment tools. In conclusion, further understanding of breast asymmetry is needed to better exploit how bilateral differences in mammographic features may be used to inform breast cancer etiology as well as future breast cancer risk. Citation Format: Maeve Mullooly, Shaoqi Fan, Ruth M. Pfeiffer, Brian Sprague, Pamela M. Vacek, Donald L. Weaver, John A. Shepherd, Amir Pasha Mahmoudzadeh, Jeff Wang, Serghei Malkov, Jason M. Johnson, Sally D. Herschorn, Gretchen L. Gierach. Investigation of relationships between breast cancer risk factors and bilateral mammographic breast density asymmetry among women undergoing diagnostic image-guided breast biopsies [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr NG15.

Abstract: Immunodominance refers to the restricted peptide specificity of T cells that are detectable after an adaptive immune response. For CD4 T cells, many of the mechanisms used to explain this selectivity suggest that events related to Ag processing play a major role in determining a peptide’s ability to recruit CD4 T cells. Implicit in these models is the prediction that the molecular context in which an antigenic peptide is contained will impact significantly on its immunodominance. In this study, we present evidence that the selectivity of CD4 T cell responses to peptides contained within protein Ags is not detectably influenced by the location of the peptide in a given protein or the primary sequence of the protein that bears the test peptide. We have used molecular approaches to change the location of peptides within complex protein Ags and to change the flanking sequences that border the peptide epitope to now include a protease site, and find that immunodominance or crypticity of a peptide observed in its native protein context is preserved. Collectively, these results suggest immunodominance of peptides contained in complex Ags is due to an intrinsic factor of the peptide, based upon the affinity of that peptide for MHC class II molecules. These findings are discussed with regard to implications for vaccine design.