Objective: The aim of this study was to identify the comprehensive risk factors for lymphedema, thereby enabling a more informed multidisciplinary treatment decision-making. Summary Background Data: Lymphedema is a serious long-term complication in breast cancer patients post-surgery; however, the influence of multimodal therapy on its occurrence remains unclear. Methods: We retrospectively collected treatment-related data from 5549 breast cancer patients who underwent surgery between 2007 and 2015 at our institution. Individual radiotherapy plans were reviewed for regional nodal irradiation (RNI) field design and fractionation type. We identified lymphedema risk factors and used them to construct nomograms to predict individual risk of lymphedema. Nomograms were validated internally using 100 bootstrap samples and externally using 2 separate datasets of 1877 Asian and 191 Western patients. Results: Six hundred thirty-nine patients developed lymphedema during a median follow-up of 60 months. The 3-year lymphedema incidence was 10.5%; this rate increased with larger irradiation volumes (no RNI vs RNI excluding axilla I–II vs RNI including axilla I–II: 5.7% vs 16.8% vs 24.1%) and when using conventional fractionation instead of hypofractionation (13.5% vs 6.8%). On multivariate analysis, higher body mass index, larger number of dissected nodes, taxane-based regimen, total mastectomy, larger irradiation field, and conventional fractionation were strongly associated with lymphedema (all P < 0.001). Nomograms constructed based on these variables showed good calibration and discrimination internally (concordance index: 0.774) and externally (0.832 for Asian and 0.820 for Western patients). Conclusions: Trimodality breast cancer treatment factors interact to promote lymphedema. Lymphedema risk can be decreased by deintensifying node dissection, chemotherapy regimen, and field and dose of radiotherapy. Deescalation strategies on a multidisciplinary basis might minimize lymphedema risk.

Background: In breast cancer patients receiving radiotherapy (RT), accurate target delineation and reduction of radiation doses to the nearby normal organs is important. However, manual clinical target volume (CTV) and organs-at-risk (OARs) segmentation for treatment planning increases physicians’ workload and inter-physician variability considerably. In this study, we evaluated the potential benefits of deep learning-based auto-segmented contours by comparing them to manually delineated contours for breast cancer patients. Methods: CTVs for bilateral breasts, regional lymph nodes, and OARs (including the heart, lungs, esophagus, spinal cord, and thyroid) were manually delineated on planning computed tomography scans of 111 breast cancer patients who received breast-conserving surgery. Subsequently, a two-stage convolutional neural network algorithm was used. Quantitative metrics, including the Dice similarity coefficient (DSC) and 95% Hausdorff distance, and qualitative scoring by two panels from 10 institutions were used for analysis. Inter-observer variability and delineation time were assessed; furthermore, dose-volume histograms and dosimetric parameters were also analyzed using another set of patient data. Results: The correlation between the auto-segmented and manual contours was acceptable for OARs, with a mean DSC higher than 0.80 for all OARs. In addition, the CTVs showed favorable results, with mean DSCs higher than 0.70 for all breast and regional lymph node CTVs. Furthermore, qualitative subjective scoring showed that the results were acceptable for all CTVs and OARs, with a median score of at least 8 (possible range: 0–10) for (1) the differences between manual and auto-segmented contours and (2) the extent to which auto-segmentation would assist physicians in clinical practice. The differences in dosimetric parameters between the auto-segmented and manual contours were minimal. Conclusions: The feasibility of deep learning-based auto-segmentation in breast RT planning was demonstrated. Although deep learning-based auto-segmentation cannot be a substitute for radiation oncologists, it is a useful tool with excellent potential in assisting radiation oncologists in the future. Trial registration Retrospectively registered.

Physical activity is known to significantly impact cardiometabolic health. Accelerometer data, as a measure of physical activity, can be used to objectively identify a disparity in movement (movement discordance) between healthy and unhealthy adults. The purpose of this study was to examine the Movement Discordance between healthy and unhealthy adults in a large US population sample. Demographic, health and accelerometer data from the National Health and Nutrition Examination Study (NHANES) 2003–2004 and 2005–2006 cohorts were used for this study. Participants were classified as either having a “normal” or “abnormal” value for each cardiometabolic health parameter examined, based on published criteria. Linear regression analyses were performed to determine significance of each abnormal health parameter (risk factor) in its unique effect on the accelerometer counts, controlling for age and gender. Average accelerometer counts per minute (cpm) by gender and age categories were estimated separately for the groups of normal and abnormal cardiometabolic risk. Average cpm for those with healthy levels of each individual cardiometabolic health parameter range from 296 cpm (for C reactive protein) to 337 cpm (for waist circumference), while average cpm for those with abnormal levels of each individual cardiometabolic health parameter range from 216 cpm (for insulin) to 291 cpm (for LDL-cholesterol). After controlling for age and gender, waist circumference, HbA1c, Insulin, Homocysteine, and HDL-Cholesterol were the cardiometabolic health parameters that showed significant, unique and independent effects on cpm. Overall, individuals who have abnormal values for all significant cardiometabolic health parameters (“unhealthy”) averaged 267 cpm (SE = 15 cpm), while the healthy sample of this study averaged 428 cpm (SE = 10 cpm). The difference in cpm between the unhealthy and healthy groups is similar between males and females. Further, for both males and females, the cpm gap between unhealthy and healthy is largest in the 30s (males: 183 cpm; females 144 cpm) and lessens as age increases, with the lowest gap seen in those 80+ years (males, 81 cpm; females, 85 cpm). This Movement Discordance between healthy and unhealthy adults represents a gap in movement that needs to be closed to improve the health of individuals with, or at risk for cardiometabolic disease.