Conant, Emily F.

Three-dimensional radiomic analysis of digital breast tomosynthesis outperformed two-dimensional measures in breast cancer risk assessment.

Radiomic parenchymal phenotypes on mammograms were used to predict breast cancer risk among both Black and White women, notably for false-negative findings and symptomatic interval cancer.

PURPOSE Breast density is a widely established independent breast cancer risk factor. With the increasing utilization of digital breast tomosynthesis (DBT) in breast cancer screening, there is an opportunity to estimate volumetric breast density (VBD) routinely. However, current available methods extrapolate VBD from two-dimensional (2D) images acquired using DBT and/or depend on the existence of raw DBT data, which is rarely archived by clinical centers because of storage constraints. METHODS We retrospectively analyzed 1,080 nonactionable three-dimensional (3D) reconstructed DBT screening examinations acquired between 2011 and 2016. Reference tissue segmentations were generated using previously validated software that uses 3D reconstructed slices and raw 2D DBT data. We developed a deep learning (DL) model that segments dense and fatty breast tissue from background. We then applied this model to estimate %VBD and absolute dense volume (ADV) in cm 3 in a separate case-control sample (180 cases and 654 controls). We created two conditional logistic regression models, relating each model-derived density measurement to likelihood of contralateral breast cancer diagnosis, adjusted for age, BMI, family history, and menopausal status. RESULTS The DL model achieved unweighted and weighted Dice scores of 0.88 (standard deviation [SD] = 0.08) and 0.76 (SD = 0.15), respectively, on the held-out test set, demonstrating good agreement between the model and 3D reference segmentations. There was a significant association between the odds of breast cancer diagnosis and model-derived VBD (odds ratio [OR], 1.41 [95 % CI, 1.13 to 1.77]; P = .002), with an AUC of 0.65 (95% CI, 0.60 to 0.69). ADV was also significantly associated with breast cancer diagnosis (OR, 1.45 [95% CI, 1.22 to 1.73]; P < .001) with an AUC of 0.67 (95% CI, 0.62 to 0.71). CONCLUSION DL-derived density measures derived from 3D reconstructed DBT images are associated with breast cancer diagnosis.

The NCCN Guidelines for Breast Cancer Screening and Diagnosis provide health care providers with a practical, consistent framework for screening and evaluating a spectrum of clinical presentations and breast lesions. The NCCN Breast Cancer Screening and Diagnosis Panel is composed of a multidisciplinary team of experts in the field, including representation from medical oncology, gynecologic oncology, surgical oncology, internal medicine, family practice, preventive medicine, pathology, diagnostic and interventional radiology, as well as patient advocacy. The NCCN Breast Cancer Screening and Diagnosis Panel meets at least annually to review emerging data and comments from reviewers within their institutions to guide updates to existing recommendations. These NCCN Guidelines Insights summarize the panel’s decision-making and discussion surrounding the most recent updates to the guideline’s screening recommendations.

Background: Image-derived artificial intelligence (AI) risk models have shown promise in identifying high-risk women in the short term. The long-term performance of image-derived risk models expanded with clinical factors has not been investigated. Methods: We performed a case–cohort study of 8110 women aged 40–74 randomly selected from a Swedish mammography screening cohort initiated in 2010 together with 1661 incident BCs diagnosed before January 2022. The imaging-only AI risk model extracted mammographic features and age at screening. Additional lifestyle/familial risk factors were incorporated into the lifestyle/familial-expanded AI model. Absolute risks were calculated using the two models and the clinical Tyrer–Cuzick v8 model. Age-adjusted model performances were compared across the 10-year follow-up. Results: The AUCs of the lifestyle/familial-expanded AI risk model ranged from 0.75 (95%CI: 0.70–0.80) to 0.68 (95%CI: 0.66–0.69) 1–10 years after study entry. Corresponding AUCs were 0.72 (95%CI: 0.66–0.78) to 0.65 (95%CI: 0.63–0.66) for the imaging-only model and 0.62 (95%CI: 0.55–0.68) to 0.60 (95%CI: 0.58–0.61) for Tyrer–Cuzick v8. The increased performances were observed in multiple risk subgroups and cancer subtypes. Among the 5% of women at highest risk, the PPV was 5.8% using the lifestyle/familial-expanded model compared with 5.3% using the imaging-only model, p < 0.01, and 4.6% for Tyrer–Cuzick, p < 0.01. Conclusions: The lifestyle/familial-expanded AI risk model showed higher performance for both long-term and short-term risk assessment compared with imaging-only and Tyrer–Cuzick models.

Abstract Objective The purpose of this study is to assess the “real-world” impact of an artificial intelligence (AI) tool designed to detect breast cancer in digital breast tomosynthesis (DBT) screening exams following 12 months of utilization in a subspecialized academic breast center. Methods Following IRB approval, mammography audit reports, as specified in the BI-RADS atlas, were retrospectively generated for five radiologists reading at three locations during a 12-month time frame. One location had the AI tool (iCAD ProFound AI v2.0), and the other two locations did not. The co-primary endpoints were cancer detection rate (CDR) and abnormal interpretation rate (AIR). Secondary endpoints included positive predictive values (PPVs) for cancer among screenings with abnormal interpretations (PPV1) and for biopsies performed (PPV3). Odds ratios (OR) with two-sided 95% confidence intervals (CIs) summarized the impact of AI across radiologists using generalized estimating equations. Results Nonsignificant differences were observed in CDR, AIR, and PPVs. The CDR was 7.3 with AI and 5.9 without AI (OR 1.3, 95% CI: 0.9–1.7). The AIR was 11.7% with AI and 11.8% without AI (OR 1.0, 95% CI: 0.8–1.3). The PPV1 was 6.2% with AI and 5.0% without AI (OR 1.3, 95% CI: 0.97–1.7). The PPV3 was 33.3% with AI and 32.0% without AI (OR 1.1, 95% CI: 0.8–1.5). Conclusion Although we are unable to show statistically significant changes in CDR and AIR outcomes in the two groups, the results are consistent with prior reader studies. There is a nonsignificant trend toward improvement in CDR with AI, without significant increases in AIR.

PURPOSE Image-derived artificial intelligence–based short-term risk models for breast cancer have shown high discriminatory performance compared with traditional lifestyle/familial-based risk models. The long-term performance of image-derived risk models has not been investigated. METHODS We performed a case-cohort study of 8,604 randomly selected women within a mammography screening cohort initiated in 2010 in Sweden for women age 40-74 years. Mammograms, age, lifestyle, and familial risk factors were collected at study entry. In all, 2,028 incident breast cancers were identified through register matching in May 2022 (206 incident breast cancers were found in the subcohort). The image-based model extracted mammographic features (density, microcalcifications, masses, and left-right breast asymmetries of these features) and age from study entry mammograms. The Tyrer-Cuzick v8 risk model incorporates self-reported lifestyle and familial risk factors and mammographic density to estimate risk. Absolute risks were estimated, and age-adjusted AUC model performances (aAUCs) were compared across the 10-year period. RESULTS The aAUCs of the image-based risk model ranged from 0.74 (95% CI, 0.70 to 0.78) to 0.65 (95% CI, 0.63 to 0.66) for breast cancers developed 1-10 years after study entry; the corresponding Tyrer-Cuzick aAUCs were 0.62 (95% CI, 0.56 to 0.67) to 0.60 (95% CI, 0.58 to 0.61). For symptomatic cancers, the aAUCs for the image-based model were ≥0.75 during the first 3 years. Women with high and low mammographic density showed similar aAUCs. Throughout the 10-year follow-up, 20% of all women with breast cancers were deemed high-risk at study entry by the image-based risk model compared with 7.1% using the lifestyle familial-based model ( P < .01). CONCLUSION The image-based risk model outperformed the Tyrer-Cuzick v8 model for both short-term and long-term risk assessment and could be used to identify women who may benefit from supplemental screening and risk reduction strategies.

Abstract Breast parenchymal patterns on radiologic images are associated with breast cancer risk. Radiomic features have been proposed as quantitative measures of parenchymal patterns. We defined intrinsic imaging phenotypes of breast parenchymal patterns based on radiomic features extracted from full field digital mammography (FFDM) in breast screening populations and assessed whether these phenotypes are associated with breast cancer risk and masking. We selected 30,000 women with 4-view FFDM exams from Hologic machines from three institutions (Hospital of the University of Pennsylvania, Mayo Clinic, and San Francisco Mammography Registry), randomly split into a training (20,000 women) and test set (10,000 women). In total, 390 radiomic features were automatically extracted from each image using a validated software pipeline, standardized, and adjusted for site differences using ComBat. We used two methods, hierarchical clustering and Principal Components (PCs) analysis, to classify significant variation among the features in the training set and replicate among the test set. Next, we applied the replicated clusters and PCs to an independent nested case-control set [1082 invasive breast cancer (BC) cases (of which 151 were Black and 893 White women, 38 other race) matched to 2837 controls (411 Black and 2345 White women, 81 other race) on age, race, timing of images, and site]. We examined associations of the clusters and PCs with invasive breast cancer risk, as well as masking [defined as a false-negative (FN) screen (124 cases and 319 matched controls) and additionally the subset with symptomatic interval cancer (IC) within 12 months of negative screen (88 cases and 223 matched controls)] using conditional logistic regression. We evaluated their association with breast cancer alone, and with adjustment for age, body mass index (BMI) and breast density assessed by Breast Imaging Reporting and Data System (BI-RADS) using likelihood ratio tests. We estimated discrimination using area under the curve (AUC) and compared AUCs for models that included the radiomic clusters and PCs with the model that included only age, BMI and density. We also stratified analyses by race (Black/White). From hierarchical clustering, we defined six statistically significant phenotype clusters (each of at least 1000 women) in the training set which were replicated in the test set. For PC Analysis, we identified six PCs in the training set, explaining 85% of the variation in texture features and reproduced these in the test set. The six radiomic phenotype clusters (P&lt; 0.001) and six PCs (P&lt; 0.001) were both associated with invasive BC, including after adjusting for age, BMI, and density (cluster P=0.004; PCs P&lt; 0.001). Improvements in discrimination of invasive BC with inclusion of PCs or clusters were more pronounced among Black women (Table). Further, the PCs (P&lt; 0.001) and clusters (P&lt; 0.001) were significantly associated with FN overall and for symptomatic IC (PCs P&lt; 0.001; clusters P=0.001), but only PCs remained significant after adjusting for age, BMI and density (PCs P=0.004 for FN; PCs P=0.007 for symptomatic IC). Discrimination of masking also improved with inclusion of both clusters and PCs (Table). We identified reproducible radiomic phenotypes that are associated with invasive BC risk, above and beyond breast density with the strongest associations for invasive BC among Black women and symptomatic interval cancers. Citation Format: Stacey Winham, Anne Marie McCarthy, Aimilia Gastounioti, Christopher Scott, Aaron Norman, Walter C. Mankowski, Lauren Pantalone, Matthew Jensen, Eric A. Cohen, Hannah Horng, Kathleen Brandt, Emily F. Conant, Karla Kerlikowske, Despina Kontos, Celine Vachon. Radiomic phenotypes of breast texture and association with breast cancer risk and masking [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 GS4-06.

Abstract Introduction: Mammographic breast density is among the strongest risk factors for breast cancer. However, breast density is typically assessed subjectively by the radiologist according to the Breast Imaging Reporting and Data System (BI-RADS) based on 2 dimensional (2D) digital mammography (DM) images. Digital breast tomosynthesis (DBT) is quickly replacing DM and allows more detailed volumetric imaging of the breast. Advances in radiomics, the high-throughput extraction of radiologic features, has enabled characterization of breast parenchymal complexity beyond breast density alone. The purpose of this study was to compare the performance of volumetric parenchymal pattern analysis from DBT and DM with conventional breast density measurement with respect to breast cancer risk estimation. Methods: We performed a case control study among women with concurrent DM and DBT screening (Selenia Dimensions, Hologic Inc.) at our institution between 3/2011-12/2014. Cases were diagnosed with breast cancer within 1 year of screening; controls were confirmed negative or benign at 1 year follow-up, matched on race (Black, White, other/unknown) and age (5-year bins). After exclusions for imaging artifacts, craniocaudal (CC) and mediolateral oblique (MLO) views for 187 cases and 737 controls, in six image formats were assessed: 1) raw (“FOR PROCESSING”) DM; 2) processed (“FOR PRESENTATION”) DM; 3) raw DBT central projection; 4) processed DBT central projection; 5) DBT central reconstructed slice; and 6) DBT reconstructed stack. For cases, we analyzed the breast contralateral to cancer diagnosis; for controls the same breast as the matched case. We extracted radiomic features using a lattice-based approach with the publicly available CaPTk software, averaging features for each breast over CC and MLO views. We examined 3 lattice window sizes (6.4, 12.8, and 25.6 mm) and 23 resolutions for image resampling (0.075 - 2mm). We performed PCA on the resulting 487 features for each combination of window size and resolution and built conditional logistic regression models to assess the association of the first 7 principal components with breast cancer, with models including age, BMI, and BI-RADS density. For each image type we calculated the model C-statistic at all window sizes and resolutions, for a total of 2304 experimental conditions. Results: Features from reconstructed DBT scans had on average higher C-statistics across all experimental conditions. A model using only age, BMI, and BI-RADS density had a C-statistic of 0.61. Models using radiomic features plus age, BMI, and BI-RADS density had mean C-statistic of 0.68 (IQR 0.68, 0.69) for reconstructed DBT scans; for all other image types, the mean C-statistic ranged from 0.64 to 0.66. Conclusions: Incorporating volumetric breast parenchymal patterns from DBT improves breast cancer risk estimation beyond markers derived from DM and beyond conventional BI-RADS density. Citation Format: Eric A. Cohen, Omid Haji Maghsoudi, Raymond Acciavatti, Lauren Pantalone, Walter Mankowski, Alex A Nguyen, Christopher G. Scott, Stacey Winham, Andrew D. Maidment, Anne Marie McCarthy, Celine M Vachon, Emily F Conant, Despina Kontos. Volumetric parenchymal pattern analysis for breast cancer risk estimation. [abstract]. In: Proceedings of the AACR Special Conference: Precision Prevention, Early Detection, and Interception of Cancer; 2022 Nov 17-19; Austin, TX. Philadelphia (PA): AACR; Can Prev Res 2023;16(1 Suppl): Abstract nr P070.

Abstract Background: It has been widely established that breast density is an independent breast cancer risk factor. With the increasing utilization of digital breast tomosynthesis (DBT) in breast cancer screening, there is an opportunity to estimate volumetric breast density (VBD) routinely. However, current available methods extrapolate VBD from 2D images acquired with DBT and/or depend on the existence of raw DBT data which is rarely archived by clinical centers due to cost and storage constraints. This study aims to harness deep learning to develop a computational tool for VBD assessment based solely on 3D reconstructed, “for presentation” DBT images. Methods: We retrospectively analyzed 1,080 negative DBT screening exams (09/20/2011 - 11/25/2016) from the Hospital of the University of Pennsylvania (mean age ± SD, 57 ± 11 years; mean BMI ± SD, 28.7 ± 7.1 kg/m2; racial makeup, 41.2% White, 54.2% Black, 4.6% Other), for which both 3D reconstructed and 2D raw DBT images (Selenia Dimensions, Hologic Inc) were available. All available standard views (left and right mediolateral-oblique and cranio-caudal views) were included for each exam, leading to 7,850 DBT views. Corresponding 3D reference-standard tissue segmentations were generated from a previously validated software that uses both 3D reconstructed slices and raw 2D DBT data to provide VBD metrics, shown to be strongly correlated with VBD measures from MRI image volumes. We based our deep learning algorithm on the U-Net architecture within the open-source Generally Nuanced Deep Learning Framework (GaNDLF) and created a 3-label image segmentation task (background, dense tissue, and fatty tissue). Our dataset was randomly split into training (70%), validation (15%) and test (15%) sets, while ensuring that all views of the same DBT exam were assigned to the same set. The performance of our deep learning algorithm against the corresponding reference-standard segmentations was measured in terms of Dice scores (DSC), with 0 signifying no overlap and 1 signifying perfect overlap, overall, as well as separately for each label. Results: After training was complete, our deep learning algorithm achieved a DSC of 0.78 on the validation, as well as on the test set. Our method accurately segmented background from breast tissue (DSC = 0.94) and demonstrated moderate to high performance in segmenting dense and fatty tissue, respectively (DSC = 0.49 and 0.89). Conclusion: Our preliminary analysis suggests that deep learning shows promise in the estimation of VBD using 3D DBT reconstructed, “for presentation” images. Future work involving transfer learning based on ground truth masks by clinical radiologists could further enhance this method’s performance. In view of rapid clinical conversion to DBT screening, such a tool has the potential to enable large retrospective epidemiologic and personalized risk assessment studies of breast density with DBT. Citation Format: Vinayak S. Ahluwalia, Walter Mankowski, Sarthak Pati, Spyridon Bakas, Ari Brooks, Celine M. Vachon, Emily F. Conant, Aimilia Gastounioti, Despina Kontos. Deep-learning-enabled volumetric breast density estimation with digital breast tomosynthesis [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 1929.