With increasing survival rates for many cancers, evaluating long-term complications of treatment has become critical. Growing data indicate that chemotherapy and hormonal therapy may be associated with cognitive impairment, a strong determinant of quality of life and daily function. However, existing studies have largely followed patients up to just one year post-treatment. Extensive data indicate that adverse exposures at mid-life may be strong risk factors for cognitive decline in later life, and initial data suggest tat effects of chemotherapy on brain health may emerge years after treatment. Finally, epidemiologic studies have focused largely on breast cancer, although chemotherapeutic agents for other cancers (e.g., colorectal) demonstrate neurotoxicity. We propose here to take advantage of existing data from 20,000 women in the Cognitive Sub cohort of the Nurses' Health Study, in which four repeated measures of cognitive function were collected via a battery of 6 cognitive tests. Initial cognitive testing was conducted in 1995-2000 in all women age 70 years and older at that time. This included 1239 with a history of invasive breast cancer, 210 with DCIS, 317 with colorectal cancer, and 15,000 with no cancer history. These data will allow the first larger-scale, prospective investigation of systemic cancer treatment and longer-term risk of cognitive decline. We will focus on key questions identified by the recent report from the International Cognition and Cancer Task Force, and our Aims will address: (1) if women given systemic therapy (chemotherapy, hormonal therapy) for breast cancer, or for colorectal cancer, have greater decline in global cognitive function (an average of all cognitive tests in our battery versus control patients not given systemic therapy, or versus controls with no cancer; and (2) if systemic therapy is related to specific cognitive domains (episodic memory, executive function) identified by the Task Force as requiring further research, due to initial findings in the literatue. In particular, we will focus on treatment regimens including: (i) 5-FU, as this agent crosses the blood brain barrier more readily than many others; and (ii) cyclophosphamide, the backbone of breast cancer treatment during the period our breast cancer survivors were diagnosed (with separate exploration of the two most common regimens at the time, cyclophosphamide/methotrexate/5-FU and Adriamycin/ cyclophosphamide) Depending on findings, results from this application will support future research: to initiate cognitive testingin more recent patients to address newer treatments; and to evaluate biologic and lifestyle risk factors for cognitive decline in patients given systemic therapy. Thus, our research will eventually inform survivors and clinicians of any need to screen for cognitive decline even years after treatment, and, enable identification of interventions to maintain long-term cognition after treatment.