Living beyond cancer: The long-term impact of breast cancer diagnosis on cognitive function.

1633 Background: While some clinical studies report greater cognitive difficulties in middle-aged women diagnosed with and treated for breast cancer over the short-term, observational studies of older persons, with longer follow-up, found that a history of cancer was associated with lower Alzheimer’s disease risk. We estimated the relation of breast cancer and treatment history with cognitive status and rate of decline among older women. We further divided breast cancer survivors by: (1) time since cancer diagnosis (assessing recency), (2) age of cancer onset (reflecting body aging at diagnosis), and (3) stage (capturing disease aggressiveness). To evaluate any overlap in shared or opposing genetic risk, we estimated cognitive status by breast cancer polygenic risk score (PRS). Methods: A cognitive sub-study was initiated in 1995-2001 in the Nurses’ Health Study, including 1,378 breast cancer survivors and 14,196 cancer-free women. Breast cancer diagnoses were self-reported and confirmed by medical records; treatment was self-reported. Cognitive function was assessed up to 4 times (over a mean of 6.6 years) and combined into 4 outcomes: global composite score, Telephone Interview for Cognitive Status (TICS), verbal memory, and working memory. We used linear models to assess breast cancer history and treatment with cognitive status (averaged across follow-ups). We used mixed-effects models to assess breast cancer history and treatment with rate of cognitive decline. We computed a breast cancer PRS and evaluated cognitive function across quartiles of PRS. Results: The mean age of breast cancer diagnosis was 65.4 years, and the mean time between cancer diagnosis and baseline cognitive assessment was 8.6 years. We observed similar distributions of key risk factors for AD between women with and without history of breast cancer, including age, education, depression, and physical activity. No significant differences in global cognitive status were noted comparing women with a history of breast cancer with those who were cancer-free. Associations did not differ by age of cancer onset ( < 65 vs. ≥65 years), time since diagnosis ( < 5 vs. ≥5 years), or stage. Genetically predicted breast cancer risk was not associated with the global cognitive status. Women with breast cancer and treated with hormone therapy, chemotherapy, and/or radiation therapy had similar global cognitive status compared to cancer-free women. No significant differences by breast cancer history were observed for TICS, verbal memory, or working memory. Over the modest follow-up time, we observed no significant differences in cognitive decline between women with a history of breast cancer and cancer-free women. Conclusions: We observed no association of history of breast cancer or breast cancer treatment with cognitive function status or rate of decline, suggesting there is neither harm nor benefit of breast cancer diagnosis on long-term cognition.