If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Chemotherapy for breast cancer has been associated with cognitive problems; however, the impact of adjuvant hormone therapy is less clear. No studies have explored provider discussions about cognitive concerns or factors associated with neurocognitive treatment. This study examined cognitive problems, factors associated with having a provider discussion, and receipt of neurocognitive treatment.
Methods
Female breast cancer survivors (N=2,537) from the Sister Study and the Two Sister Study who were at least 1 year post-treatment were surveyed in 2012 about their cancer therapies (confirmed by medical records); cognitive concerns; related provider discussions; and neurocognitive treatment. A total of 2,296 women were included in the current 2014 analysis. Extensive covariate information was also ascertained for predictive multivariate models.
Results
The prevalence of self-reported cognitive problems after treatment was 60%. Of those reporting cognitive problems, only 37% had discussed those concerns with a provider and 15% had been treated for cognitive symptoms. The odds of reported cognitive concerns that started during and after treatment were elevated for those who received only hormone therapy and no chemotherapy (OR=1.64, 95% CI=1.15, 2.33); chemotherapy and no hormone therapy (OR=5.63, 95% CI=3.52, 9.00); or both (OR=6.33, 95% CI=4.21, 9.54) compared with those reporting neither treatment.
Conclusions
The high prevalence of cognitive concerns underscores the importance of monitoring breast cancer survivors for potential neurocognitive effects of hormone and chemotherapy, discussions with survivors about those concerns, and treatment referrals. Monitoring changes over time can help to evaluate both psychosocial and neurocognitive care provided for survivors.
Introduction
Although advances in chemotherapy and other adjuvant treatments have contributed to improved outcomes for women diagnosed with breast cancer, these therapies can have persistent side effects.
have found that 17%–75% of breast cancer survivors subjectively experience cognitive impairment after chemotherapy. Cognitive impairments, including problems with learning, language, and concentration, can profoundly affect a survivor’s daily functioning.
A large proportion of patients treated with chemotherapy (CTX) also receive hormonal adjuvant therapies (HTX) such as the estrogen receptor antagonist tamoxifen or aromatase inhibitors. Research findings examining the influence of CTX and HTX on reported neurocognitive functioning among breast cancer survivors are varied, and there is little research examining cognitive effects of HTX use among premenopausal survivors. Some analyses focused on postmenopausal survivors (compared with healthy controls) found use of HTX (e.g., tamoxifen and exemestane) to be associated with lower cognitive functioning (e.g., verbal memory, executive functioning),
Neuropsychological functioning in postmenopausal breast cancer patients treated with tamoxifen or exemestane after AC-chemotherapy: cross-sectional findings from the neuropsychological TEAM-side study.
Schilder CM, Seynaeve C, Linn SC, et al. Self-reported cognitive functioning in postmenopausal breast cancer patients before and during endocrine treatment: findings from the neuropsychological TEAM side-study. Psychooncology. 2012;21(5):479–487. http://dx.doi.org/10.1002/pon.1928.
Schilder CM, Seynaeve C, Beex LV, et al. Effects of tamoxifen and exemestane on cognitive functioning of postmenopausal patients with breast cancer: results from the neuropsychological side study of the Tamoxifen and Exemestane Adjuvant Multinational trial. J Clin Oncol. 2010;28(8):1294–1300. 10.1200/JCO.2008.21.3553.
Patient and provider discussions regarding treatment-induced cognitive concerns are underexplored, and studies often do not distinguish discussions specific to neurocognitive dysfunction from those related to emotional/behavioral concerns. One recent estimate is that 31% of cancer survivors had a provider discussion about psychosocial concerns,
but it is unclear if these discussions included neurocognitive symptoms.
There is also limited information on receipt of neurocognitive treatments and interventions (e.g., neurocognitive testing, behavioral therapy, psychostimulants, and alternative treatments). Only 14% of survivors participated in counseling in one population-based study.
Though breast cancer survivors and survivors diagnosed fewer than 5 years previously are more likely to receive counseling and supportive services (compared with survivors with other cancers and survivors diagnosed more than 5 years previously),
Hewitt M, Breen N, Devesa S. Cancer prevalence and survivorship issue: analyses of the 1992 National Health Interview Survey. J Natl Cancer Inst. 1991;91:1480–1486. http://dx.doi.org/10.1093/jnci/91.17.1480.
it is unclear if these services are used to decrease cognitive effects related to cancer treatment.
The current study explored characteristics of survivors by treatment type, predictors of self-reported cognitive impairment, patient–provider discussions, and receipt of neurocognitive treatment. Although some relationships explored were descriptive and hypothesis generating, the authors predicted that receipt of CTX, HTX, or both (compared with no treatment) would be associated with cognitive impairment and that daily living concerns would be associated with having provider discussions and receipt of neurocognitive treatment.
Methods
In 2012, CDC and the National Institute of Environmental Health Sciences carried out the CDC Sisters Study Survivorship Survey (Survivorship Survey), to examine survivorship concerns among women with a prior diagnosis of breast cancer. The Survivorship Survey included women diagnosed with breast cancer who were enrolled in either the Sister Study, a cohort of initially breast cancer–free women whose sister had been diagnosed with breast cancer, or the Two Sister Study, an ancillary study of women with young-onset (diagnosed age <50 years) breast cancer who had a sister in the Sister Study cohort. Study design and methodology for the Sister Study and the Two Sister Study are described on the study website (www.sisterstudy.org) and in previous publications.
The 2,832 women eligible for the Survivorship Survey were aged 34–82 years, lived in the U.S. or Puerto Rico, spoke English, and met all other Sister or Two Sister Study inclusion criteria (www.sisterstudy.org). Sister Study participants were eligible if their first breast cancer (ductal carcinoma in situ or invasive breast cancer) was diagnosed by October 9, 2012, and they completed study follow-up activities. All Two Sister Study participants were eligible for the Survivorship Survey. Invited participants were offered a $20 Visa gift card. The survey took no more than 1.5 hours to complete and was approved by relevant IRBs (OMB Control #0925-0522). A total of 2,537 women completed the Survivorship Survey via mailed, paper questionnaire (n=2,378) or computer-assisted telephone interview (n=159), for a response rate of 90.3%. After implementing exclusion criteria for the study, 2,296 women were included in the analytic sample, and 241 were excluded.
Measures
The Survivorship Survey included questions on medical history, health behaviors, employment, quality of life, and other aspects of the survivor experience. Survey questions used in the current analysis were taken from existing surveys,
were newly developed, or modified based on existing validated measures (Appendix, available online).
Years since diagnosis, age at diagnosis, breast cancer type, race/ethnicity, marital status, education, U.S. region of residence, and employment were recoded as categorical variables (Table 1).
Table 1Characteristics of Breast Cancer Survivors by Adjuvant Treatment Type, Sister Study Survivorship Survey, 2014
Survey and medical records data were combined to categorize participant’s treatment for first breast cancer (source agreement was >90%). Using an algorithm (Figure 1), respondents were classified into four mutually exclusive treatment groups: CTX alone, HTX alone, CTX and HTX, and neither CTX nor HTX. Respondents were excluded if they had Stage 4 cancer and radiation, or Stage 3 cancer and reported their breast cancer had metastasized (n=59) because of the possibility of radiation to the central nervous system and associated neurocognitive deficits.
Figure 1Selection criteria for CTX and HTX treatment history groups.
Problems with thinking, memory, and attention and trouble with concentration, short-term memory, recall memory, organizing daily functions, and multitasking were assessed (Appendix, available online). Respondents who experienced symptoms sometimes, often, or always were categorized as having “neurocognitive concerns.” Additional questions asked included when respondents first noticed neurocognitive symptoms, if they experienced any improvement, and if they spoke to their doctor (yes or no) or received treatments for neurocognitive symptoms. A variable for daily living concerns was created based on how many daily living tasks the respondent’s neurocognitive symptoms interfered with, categorized as zero to one, two to three, four to five, or six to eight (Appendix, available online).
Statistical Analysis
Statistical analyses were performed in 2014 with SAS, version 9.3. Descriptive statistics for study variables were reported separately by cancer treatment group (Table 1).
Multivariable logistic regression models were used to assess factors associated with three outcomes: reported neurocognitive concerns (Table 2); provider discussions; and treatment receipt for neurocognitive symptoms (Table 3). Unadjusted bivariate analyses were first conducted to examine the association between scientifically relevant individual characteristics and the three outcomes. Correlation coefficients and chi-square statistics were used to assess the association between variables. Final adjusted models to determine predictors of each outcome included all potential predictors that had been examined in unadjusted bivariate analyses as well as age at survey, which was assessed in 10-year categories using dummy variables. Women who reported that they had been premenopausal and aged ≥60 years at diagnosis were excluded from analyses, as were those who reported neurocognitive concerns prior to treatment.
Table 2Predictors of Reporting Neurocognitive Symptoms After Treatment Among Breast Cancer Survivors, Sister Study Survivorship Survey, 2014
Women who reported being premenopausal and aged >60 years at dx were excluded from the sample. After this exclusion, there were no women who were aged >65 years at the time of survey in the sample who reported being premenopausal prior to treatment.
Women who reported being premenopausal and aged >60 years at dx were excluded from the sample. After this exclusion, there were no women who were aged >65 years at the time of survey in the sample who reported being premenopausal prior to treatment.
Women who reported being premenopausal and aged >60 years at dx were excluded from the sample. After this exclusion, there were no women who were aged >65 years at the time of survey in the sample who reported being premenopausal prior to treatment.
Women who reported being premenopausal and aged >60 years at dx were excluded from the sample. After this exclusion, there were no women who were aged >65 years at the time of survey in the sample who reported being premenopausal prior to treatment.
Menopausal status (ref: postmenopausal prior to tx)
CTX, chemotherapy; dx, diagnosis; HAT, hormonal adjuvant therapy for breast cancer; HS, high school; tx, treatment.
a Restricted to those who reported neurocognitive concerns during or after treatment but not prior to treatment.
b Premenopausal refers to women who were premenopausal prior to and remained premenopausal after treatment.
c Adjusted for all predictors including age at survey (10 year categories using dummy variable).
d Women who reported being premenopausal and aged >60 years at dx were excluded from the sample. After this exclusion, there were no women who were aged >65 years at the time of survey in the sample who reported being premenopausal prior to treatment.
e Not included in the model (no variability in the response).
Of 2,296 breast cancer survivors included in the study, 79% were married, 90% were white and non-Hispanic, 62% were employed for wages, 79% had invasive breast cancer, and 54% were >5 years post-treatment when surveyed. More than half of respondents had a college degree or higher and, because of inclusion criteria for the Two Sister Study, many were aged 40–49 years at diagnosis (45%); aged 45–54 years at the time of survey (40%); and half were premenopausal prior to treatment.
Thirteen percent of all survivors were treated with CTX alone, 36% with HTX alone, 37% with both, and 14% did not receive either treatment (Table 1). Demographic characteristics of survivors treated with CTX alone or both CTX and HTX were similar (Table 1). Those treated with CTX (with or without HTX) overwhelmingly had invasive disease (99%), whereas 72% of those treated with HTX alone had invasive disease. Survivors not treated with either CTX or HTX had predominantly in situ disease (70%; Table 1).
Sixty percent of survivors reported problems with thinking, memory, and attention (Table 1). Of those, 60% first noticed the problems within 6 months after starting their primary breast cancer treatment and 76% reported that their symptoms had either stayed the same or gotten worse during the past 12 months (data not shown). The majority of women who received CTX, either alone or with HTX, reported cognitive problems after treatment (74% and 78%, respectively). Forty-five percent of women who received HTX alone also reported post-treatment cognitive problems, as did 38% of those who did not receive either treatment.
Most women with cognitive concerns, regardless of treatment type, reported problems with concentration, short-term memory, and recall memory. Thirty-seven percent of women in all treatment groups reported problems with organization. Difficulties with multitasking were reported by 43% of women who received HTX alone, 47% who received CTX alone or both treatments, and 52% of those who received neither treatment. The majority of women who received CTX, HTX, or both first noticed cognitive concerns <6 months after treatment and reported that their symptoms had stayed the same or worsened in the past 12 months.
In unadjusted models with all survivors, treatment history, breast cancer type, years since diagnosis, education, current age, and menopausal status at diagnosis were associated with reported neurocognitive concerns. After simultaneous adjustment for the other variables in Table 2, only treatment history and age at survey remained significant (p=0.00). The odds for incident neurocognitive concerns following treatment with CTX, HTX, or both (compared with those with neither treatment) remained elevated, though slightly attenuated. Similarly, odds of neurocognitive concerns for those who were aged ≥65 years (vs 34–44 years) remained below the null after adjustment, though slightly attenuated.
Because unadjusted models showed that being premenopausal prior to treatment was a predictor of neurocognitive concerns, especially among those with treatment-induced menopause (OR=2.94, 95% CI=2.37, 3.65), additional analyses of premenopausal women were conducted (Table 2). Unadjusted models for survivors who were premenopausal at treatment showed that treatment history, breast cancer type, years since diagnosis, and (to a lesser extent) current age were predictors of reported neurocognitive concerns. In the fully adjusted model, for survivors who were premenopausal at treatment, similar to results for all survivors, odds of reporting post-treatment neurocognitive problems remained elevated among those who had a history of CTX; HTX; or both treatments (compared with those with neither treatment) with ORs of 6.46 (95% CI=3.19, 13.09); 1.82 (95% CI=1.05, 3.17); and 7.26 (95% CI=3.92, 13.45), respectively.
Of those with cognitive concerns, only 37% reported discussing their cognitive symptoms with a medical provider. Of those, 30% reported receiving treatments to address cognitive symptoms (data not shown). After adjustment for all predictors, modeling showed that the relative odds of having a provider discussion were significantly elevated for those who had a master’s degree or above (compared with high school or lower); those who first noticed their cognitive symptoms within 6 months of treatment (vs >6 months later); and those who had two or more daily living concerns (compared with none or one) (Table 3). Specifically, the odds of having a provider discussion were increased twofold for survivors reporting four to five daily living concerns and increased almost fivefold for survivors reporting six to eight daily living concerns.
Among all respondents with cognitive concerns, 3% received psychological testing, 4% received behavioral therapy, 8% took prescription medications, and 8% used alternative treatments to address cognitive difficulties (data not shown). After adjustment, the relative odds of receiving testing or treatment (behavioral and pharmacologic) for cognitive concerns were significantly elevated for those with a master’s degree or above (compared with high school or lower) and those with two or more daily living concerns (compared with those with none or one) (Table 3). Specifically, the odds of reporting neurocognitive treatment were four times higher for survivors reporting four to five daily living concerns and almost ten times higher for those reporting six to eight daily living concerns.
Discussion
Treatment and demographic factors were associated with reported cognitive concerns. The odds for reporting cognitive concerns were five to six times higher for survivors who received CTX or both CTX and HTX (compared with neither treatment), consistent with the literature showing increased cognitive impairment among those treated with CTX (compared with no CTX and individuals without cancer).
The odds of reported cognitive concerns were also greater for those treated with HTX alone (compared with those with neither treatment), which suggests that HTX has a role in inducing cognitive impairment in breast cancer survivors.
Odds for reporting cognitive symptoms following treatment were similarly elevated in analyses restricted to survivors who were premenopausal prior to being treated. Some studies have shown that postmenopausal women report fewer cognitive concerns and may even receive a cognitive benefit from HTX.
Among all survivors, those aged >55 years were less likely to report incident cognitive problems. There is a literature showing a positive association between cognitive declines and increasing age.
Thus, it is possible that older survivors may attribute their cognitive concerns to normal aging, rather than to their cancer experience.
Of the 60% of survivors with reported cognitive concerns, only 37% reported having a related provider discussion. Though there is a dearth of literature regarding provider discussions about cognitive decline, these findings are similar to estimates of psychosocial provider discussions reported from the National Health Interview Survey.
Low rates of treatment receipt for cognitive symptoms are also consistent with previously reported estimates of survivors’ engagement in supportive or treatment services to address “psychosocial” concerns.
In this study, the odds of having a provider discussion about cognitive concerns were greater among those who first noticed cognitive symptoms within 6 months of treatment, had a master’s degree or above, and reported two or more daily living concerns. Survivors who first notice neurocognitive concerns within 6 months of treatment may be more aware of cognitive decline and able to associate declines with treatment when the problems first emerge during or soon after treatment. They may also have more frequent contact with providers and subsequent opportunities to initiate discussions.
Higher education levels and reporting two or more daily living concerns were associated with increased odds of having a provider discussion and also with receipt of treatment for neurocognitive concerns. Higher education has been associated with positive attitudes toward and increased willingness to seek help for mental health concerns and corresponding treatment.
It is possible that high education level among survivors might also be associated with willingness to talk to providers about neurocognitive concerns and seek out treatment. The findings regarding education might also be indicative of an increased level of situational awareness and metacognition (e.g., knowledge about one’s cognition and factors affecting cognition).
Teaching thinking: a cognitive-behavioral perspective.
in: Chipman S.F. Segal J.W. Glaser R. Thinking and Learning Skills, Vol. 2: Research and Open Questions. Lawrence Erlbaum Associates,
Hillsdale, NJ1985
Doctors might also perceive survivors with a high education level as having increased ability to comprehend neurocognitive discussions. Reporting two or more daily living concerns was also predictive of having a provider discussion about neurocognitive concerns. Survivors reporting problems with six to eight activities of daily living were two to three times more likely to talk to their provider and seek out or accept treatment than those with only two to three concerns of daily living. Increased disruption in daily living activities may create a heightened level of urgency and further motivate survivors to speak with providers and accept treatment for cognitive symptoms in order to improve their overall quality of life.
Overall, the present findings further validate other population estimates showing low rates of provider discussion and receipt of treatment for mental health concerns, and illustrate gaps in achieving IOM-recommended levels of psychosocial care.
Findings are based on data from a large sample of U.S. adult female breast cancer survivors and contribute to the existing literature by examining differences in reported cognitive impairment as related to both CTX and HTX treatment, as well as impairment related to menopausal status at the time of treatment. Findings regarding the use of neurocognitive treatments among breast cancer survivors also add to the existing literature. Utilizing the Sister and Two Sister Study cohorts provided a unique opportunity to examine concerns in a cohort of breast cancer survivors with a large number who were aged <50 years at diagnosis. Additionally, few studies have examined self-reported cognitive symptoms separately from emotional/behavioral concerns when looking at treatment and provider communication.
Study limitations include possible recall and reporting biases and the absence of objective pre-and post-measures of cognitive function. Although the sample of surveyed breast cancer survivors was large, small numbers precluded interpretation of estimates for some questions. Small numbers in some racial/ethnic categories was another limitation. The sample was also a volunteer cohort of women with sisters, which might limit generalizability of study results. Lastly, given the large number of statistical comparisons, some findings may be spurious.
Conclusions
Research should continue to examine treatment-induced neurocognitive concerns. Improved screening of neurocognitive effects is needed for survivors who received CTX and HTX. Conducting baseline and follow-up cognitive assessments could help monitor progression of cognitive concerns and determine if they improve on their own or as a result of neurocognitive treatment. Validation of self-reported data on neurocognitive concerns with medical records and neuropsychological testing results is also needed. In practice, better provider and patient education is needed regarding screening for cognitive late effects. Additionally, oncology nurses may be helpful in assessing signs of cognitive decline and educating patients about management strategies.
Few neurocognitive treatments are evidence-based, and further exploration is needed to determine their efficacy in alleviating cancer treatment–induced cognitive impairment over time.
Further investigation is needed of optimal timing for neurocognitive interventions so as not to cause or exacerbate cognitive deficits.
Additionally, barriers and facilitators to provider discussions about cognitive late effects could be explored. Although survivors often undergo cancer treatments despite the risk of side effects,
some providers may be reluctant to offer information about cognitive late effects out of concern that patients could reject treatments that may impact cognitive functioning. The essential components of having an effective discussion about cognitive decline should also be better defined. Additional practice standards of care could further incorporate cognitive late effects information into care planning. Lastly, the impact of billing and reimbursement policies on provider discussion could be evaluated further.
Acknowledgments
Publication of this article was supported by the Centers for Disease Control and Prevention, Division of Cancer Prevention and Control.
This research was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (Z01 ES044005 and Z01 ES0102245); Susan G. Komen for the Cure FAS 0703856 (the Two Sister Study); and by CDC, Division of Cancer Prevention and Control (Survivorship Survey).
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of CDC or the National Institute of Environmental Health Sciences.
No financial disclosures were reported by the authors of this paper.
Neuropsychological functioning in postmenopausal breast cancer patients treated with tamoxifen or exemestane after AC-chemotherapy: cross-sectional findings from the neuropsychological TEAM-side study.
Schilder CM, Seynaeve C, Linn SC, et al. Self-reported cognitive functioning in postmenopausal breast cancer patients before and during endocrine treatment: findings from the neuropsychological TEAM side-study. Psychooncology. 2012;21(5):479–487. http://dx.doi.org/10.1002/pon.1928.
Schilder CM, Seynaeve C, Beex LV, et al. Effects of tamoxifen and exemestane on cognitive functioning of postmenopausal patients with breast cancer: results from the neuropsychological side study of the Tamoxifen and Exemestane Adjuvant Multinational trial. J Clin Oncol. 2010;28(8):1294–1300. 10.1200/JCO.2008.21.3553.
Hewitt M, Breen N, Devesa S. Cancer prevalence and survivorship issue: analyses of the 1992 National Health Interview Survey. J Natl Cancer Inst. 1991;91:1480–1486. http://dx.doi.org/10.1093/jnci/91.17.1480.
Teaching thinking: a cognitive-behavioral perspective.
in: Chipman S.F. Segal J.W. Glaser R. Thinking and Learning Skills, Vol. 2: Research and Open Questions. Lawrence Erlbaum Associates,
Hillsdale, NJ1985
00487-0&id=gr1.jpg){kind=link}