The purpose of this longitudinal study was to examine potential predictors of cancer-related fatigue (CRF) before, during, and after adjuvant therapy in women with breast cancer. A convenience sample of 44 women postsurgery (M = 18) aged 38 to 77 years (M = 52) were recruited from a Southern breast clinic. Based on Piper's Integrated Fatigue Model, the women (1) completed questionnaires assessing innate host factors (age, income, and education level), disease and treatment patterns (disease stage, surgery type, and adjuvant therapy), psychological patterns (perceived stress, mood disturbance, and optimism), social patterns (type and satisfaction with social support) and (2) provided a blood sample to examine regulation patterns (morning cortisol levels, interleukin-1β, tumor necrosis factor alpha, and natural killer cell activity) before adjuvant therapy. The Piper Fatigue Scale-Revised was completed at all 3 time points. Mood disturbance was the most significant predictor of CRF at all time points. Interleukin-1β predicted CRF levels before adjuvant therapy and morning cortisol before adjuvant therapy predicted CRF during and after adjuvant therapy. These findings suggest that interventions to reduce mood disturbances might be effective in decreasing CRF. Further research regarding the physiological mechanisms underlying the relationships between CRF, mood disturbance, interleukin-1β, and cortisol is needed.