Breast cancer and neurocognitive outcomes

Institution: Beckman Research Institute of the City of Hope
Investigator(s): Sunita Patel, Ph.D. -
Award Cycle: 2013 (Cycle 19) Grant #: 19IB-0151 Award: $267,092
Award Type: IDEA
Research Priorities
Community Impact of Breast Cancer>Sociocultural, Behavioral, and Psychological Issues: the human side



Initial Award Abstract (2013)

Non-technical overview and relevance to breast cancer: Cancer patients report cognitive difficulties following diagnosis and are understandably concerned about the potential impact in their daily lives, both for the short term and the future. Patients frequently refer to these symptoms as “chemo brain” because they are widely thought to be a result of chemotherapy effects on the brain. Following research, patients and researchers are now realizing that other factors probably also contribute to the reports of considerable cognitive difficulties expressed by a subgroup of survivors. In particular, patients with breast cancer express concern about the potential cognitive side effects associated with endocrine medications (for example, tamoxifen) that are prescribed to reduce the risk of cancer relapse in hormone-sensitive tumors, and which patients may take for several years.

Recent research suggests that combing chemotherapy with endocrine therapy is associated with greater cognitive side effects than chemotherapy alone. However, this research has been mostly done in the two years following diagnosis, and so little is known about the long-term effects. It is also not known whether non-cancer factors that are considered as “risk” or “protective” of cognitive decline in the general population, influence cognitive and QOL outcomes in women treated breast cancer. We hypothesize that as breast cancer survivors grow older, and have also taken endocrine medications for longer, cognitive impact may become more apparent.

We propose to examine the long-term impact of breast cancer and its treatment among a previously understudied group, that of older women, in order to increase the understanding of patients and providers so they know what to expect and to identify factors that may potentially minimize the adverse effects. Specifically, we will assess the long-term impact of endocrine therapy (type, duration, etc) on cognitive outcomes in women treated with and without chemotherapy. This objective is specific to breast cancer because most breast tumors are highly sensitive to circulating hormones, requiring many women to take endocrine therapy for years to reduce the risk of tumor relapse.

The question(s) of this research are:
• Do long-term survivors of breast cancer treated with both chemotherapy and endocrine therapy have worse cognitive functioning when they reach older age than older, age-matched women without a history of cancer?
• Does having other health conditions, such as diabetes or hypertension, increase the risk for worse cognitive functioning in long-term, older survivors of breast cancer?
• Are lifestyle factors, such as higher physical activity, higher social connection, and/or higher engagement in cognitive activities, associated with better cognitive functioning in long-term survivors of breast cancer?

General methods: We will conduct neurocognitive testing in two groups of older women (= 60years) from the ongoing California Teachers Study and evaluate them for differences in their current functioning level. One group will contain women who are at least five years post breast cancer diagnosis, and the group for comparison will contain age-matched women without a history of cancer. In addition to the new data collection (i.e., neurocognitive testing and self-report questionnaires), we will also use data that already exists that was collected in 1995, 2000, and 2005 on the women in both our groups to help answer our research questions.

Innovative elements of the project: The proposed study is innovative because it will focus on understanding the long-term effects of combined chemotherapy and endocrine therapy in older, long-term survivors, a breast cancer subgroup that has been understudied for the neurocognitive impact of treatments. In addition, this study will use data collected from study participants at a time when the women in the breast cancer group did not yet have the diagnosis and this allows cost-effective longitudinal analyses. Finally, this study will conduct neurocognitive assessment using two approaches: traditional in-person testing, and web-based, self-administered, computerized testing (telephone assistance if needed). This approach is an innovative way to obtain neurocognitive data from a large number of participants across the state, who would otherwise not be able to participate.




Final Report (2016)

Overview: The study goals are to better understand how cancer-related treatment factors, such as chemotherapy and endocrine therapies, impact neurocognitive functioning in long-term survivors of breast cancer, particularly in older women who have a higher risk of functional decline. The study will also examine if non-treatment related factors, such as non-cancer chronic health conditions or lifestyle activities, are associated with neurocognitive outcomes in long-term survivors.

The question(s) of this research are:


• Do long-term survivors of breast cancer treated with both chemotherapy and endocrine therapy have worse cognitive functioning when they reach older age than older, age-matched women without a history of cancer?
• Does having other health conditions, such as diabetes or hypertension, increase the risk for worse cognitive functioning in long-term, older survivors of breast cancer?
• Are lifestyle factors, such as higher physical activity, higher social connection, and/or higher engagement in cognitive activities, associated with better cognitive functioning in long-term survivors of breast cancer?

Methods: All participants were recruited from the California Teachers Study cohort. Women with a confirmed breast cancer diagnosis who were at least five years post treatment, not receiving primary cancer treatment for relapse or other cancer, and with an active email account on file, were invited for participation via email. Age-matched controls without cancer were recruited at approximately 2:1 ratio. Along with mailed instructions on paper, a web-based link was emailed to participants for computerized assessment. Telephone assistance was provided as needed. A subgroup of participants who lived within 50 miles of our center was also invited for in-person, traditional neurocogntiive assessment. Accrual: We enrolled 214 breast cancer survivors who were approximately 9 years post diagnosis, and 122 healthy controls age matched to survivors within a 5 year span (n=336). Mean age was 68 years averaging 16 years of education, with 95% being Caucasian. Total accrual was slightly higher than the original target sample size in an effort to have a 1:1 age matching for the in-person group. All participants completed web-based cognitive testing on their home computer (i.e. remote assessment). Of these, 63 cases and 57 controls were also seen for in-person traditional cognitive testing.

Within the breast cancer group, 90 women had received both chemotherapy and endocrine therapy as part of their cancer treatments. Only 28 women were treated with chemotherapy without endocrine therapy. We did not reach the target sample size in this subgroup as very few women who get chemotherapy do not take endocrine therapy since it is given to reduce risk of relapse. The final 4 study subgroups were different in mean age with women who were treated with endocrine therapy but no chemotherapy being approximately 4 years older than the other comparison groups. The mean age for the total patient group was similar to the control group.

Results: In our data analyses so far, we did not observe a pattern where women treated with both chemotherapy and endocrine therapy consistently performed worse than age-matched healthy controls across measures of cognitive functioning. We did, however, observe that the overall total group of breast cancer patients had statistically significantly reduced functioning in working memory skills compared to women without a history of cancer, as assessed by in-person testing. We observed this reduced cognitive functioning in survivors only on selected domains rather than across all cognitive functions assessed. Furthermore, the performance of both groups was within the normal range relative to population means. In our sample, we did not find that older survivors of breast cancer (over age 70) had worse cognitive functioning compared to similar aged women without a history of cancer.

Our results found that non-treatment related factors such as comorbid chronic health conditions (e.g. high blood pressure) appear to influence cognitive functioning among our sample of participants, such that those with high blood pressure tended to have lower cognitive scores. Our preliminary analyses suggested that social factors such as marital status, past friends network, and mood are also associated with cognitive functioning. Findings related to the influence of other lifestyle factors such as physical activity and leisure activities on cognitive functioning appear mixed -data analyses is ongoing.

Plan: We will continue to analyze our data and are working towards manuscripts to disseminate our results so that they may inform the work of other researchers and clinicians who care for survivors. In general, our results suggest that long-term survivors of breast cancer who have not experienced a relapse in the past 5 years tend to be similar to women without a cancer history on many quality of life outcomes, but do show trends towards subtle cognitive differences in select domains. These differences appear small in magnitude and it is not yet clear if, and how, they may impact daily functioning.




Conference Abstract (2016)

Web-based Assessment of Neurocognitive Outcomes in Long-term Survivors of Breast Cancer from the California Teacher’s Study.

Authors: Nicole Delgado, Eunjung Lee, Nathanial Hernandez, Rich Pinder, Leslie Bernstein, Dennis Deapen & Sunita Patel
Corresponding author: spatel@coh.org
CBCRP Award 19IB-0151 (Patel)

Background: Cancer patients report cognitive difficulties following diagnosis and treatment and are concerned about the potential impact in their daily lives. Treatment-related cognitive effects of breast cancer have been widely studied in recent years. The majority of these have focused on cognition across the two years following treatment, with less attention to outcomes into long-term survivorship. In particular, the role of lifestyle health behaviors and other non-treatment factors following cancer therapies in understanding survivors’ cognitive long-term outcomes has yet to be delineated.

A major challenge to conducting research that can address these and other similar questions is the high cost of conducting objective neurocognitive assessments with large groups of participants when administered in the traditional, in-person, interactive manner. We proposed to investigate treatment and non-treatment contributors to cognitive functioning in long-term survivors using an innovative web-based assessment approach. For this report, we investigated cognitive functioning in long-term survivors of breast cancer compared to matched women without cancer from the California Teachers Study (CTS), using both web-based cognitive testing and the traditional, in-person testing procedures. We investigated whether statistically significant differences between breast cancer survivors and their controls on gold-standard traditional tests would also be found on the newer, less-investigated, computerized tests.

Methods: All participants were recruited from the CTS cohort. Women with a confirmed breast cancer diagnosis who were at least five years post treatment, not receiving primary cancer treatment for relapse or other cancer, and with an active email account on file, were invited for participation via email. Age-matched controls without cancer were recruited at a 2:1 ratio. Along with mailed instructions on paper, a web-based link was emailed to participants for computerized assessment. Telephone assistance was provided as needed. A subgroup of participants who lived within 50 miles of our center was also invited for in-person, traditional neurocogntive assessment.

Results: 214 breast cancer survivors and 122 controls completed web-based testing on their home computer (n=336). Mean age was 68 years averaging 16 years of education. Of these, 63 cases and 57 controls were also seen for in-person traditional testing. Among this subgroup, breast cancer patients had significantly reduced performance on traditional tests of working memory skills compared to controls (p=.02), controlled for age, education, mood and BMI effects. Differences in selected cognitive functioning on web-based computerized tests reached statistical significance for the total sample (p=.05), but not when analyses was restricted to the smaller subgroup sample. Further analyses showed that the patients had faster reaction time but lower accuracy scores on web-based tests compared to controls. Reduced cognitive functioning in survivors was observed only on selected domains rather than across all cognitive functions assessed, and was within the normal range.

Conclusion: Mild differences on selected measures of cognitive functioning were evident in long-term survivors of breast cancer compared to controls without cancer. Of note, results provide preliminary support for the sensitivity of remotely-administered computerized assessment in detecting cognitive differences in adequately powered studies. Web-based cognitive assessment using validated tests holds promise in assessing participants that otherwise could not participate.