Background: Body mass index (BMI) does not accurately reflect body composition, particularly among cancer survivors. Sarcopenia (low skeletal muscle mass) and low muscle radio-density (MD; suggesting fat infiltration into muscle, compromising function) increase risk of surgical complications and chemotherapy toxicity and are associated with worse survival in advanced cancer. Little is known about the prevalence or predictors of sarcopenia and low MD in early-stage breast cancer. Methods: We studied 2,914 Kaiser Permanente members diagnosed with Stage I-III breast cancer from 2005-2013. Using computed tomography (CT) scans of the third lumbar vertebra from clinical care, we determined sarcopenia (skeletal muscle index < 41 muscle [cm²]/height [m²]) and low MD ( < 25-Hounsfield Units for non-obese; < 33 for obese) using published cut points. We assessed associations with characteristics including age, race, BMI, age, stage, lifestyle and co-morbidities with logistic regression. We also examined moderate/vigorous physical activity among a subset of 672 women with activity questionnaires. Results: At diagnosis, mean age was 56 years and time to CT was 1 month. Both sarcopenia and low MD were common among early-stage breast cancer survivors (40% and 38% respectively). In multivariate analyses, the odds of sarcopenia and low MD increased with age (per 5 years, Odds Ratio [OR] 95% Confidence Interval [CI] of OR = 1.33; 95%CI: 1.27, 1.39 and OR = 1.41; 95%CI: 1.35, 1.47 respectively). The odds of sarcopenia decreased with greater BMI (OR = 0.80; 95%CI: 0.78, 0.82 per kg/m²), while the odds of low MD increased (OR = 1.03; 95%CI: 1.01, 1.04 per kg/m²). Black race was associated with lower odds of sarcopenia and low MD, while physical activity levels were associated with lower odds of sarcopenia and more favorable MD. Conclusions: Sarcopenia and low MD are highly prevalent among breast cancer survivors. While older age is strongly associated with these conditions, they occur across ages and stages. Differences in body composition by race and age may underlie differences in the association of BMI with cancer outcomes; understanding these may help guide clinical interventions.