Exploring the Impact of the Hormonal Milieu on Cognitive Functioning

Abstract

An area receiving increasing empirical attention in the past several years is cognitive functioning among cancer survivors exposed to adjuvant systemic therapies. This increased focus can be attributed to the recognition that a sizeable minority of those patients undergoing adjuvant chemotherapy report and/or demonstrate some degree of cognitive compromise—colloquially called chemobrain or chemofog. However, to date, there has been considerably less research on the possible cognitive effects of endocrine therapies for cancer treatment or risk reduction/prevention. Medications such as tamoxifen and raloxifene or the aromatase inhibitors—drugs that modify the hormonal milieu in those women with hormone receptor– positive tumors—may independently, or with chemotherapy, be associated with cognitive complaints. Given the importance of estrogen, the main target for these drugs, for normal brain function and its relationship to several neurological processes, it makes sense to better understand the potential cognitive sequelae of these medications. In this issue of the Journal of Clinical Oncology, Legault and colleagues report their findings comparing the impact of two selective estrogen receptor modulators (SERM)—tamoxifen and raloxifene—on the cognitive function of postmenopausal women at increased risk for breast cancer. The Cognitive Aging Study of Tamoxifen and Raloxifene (Co-STAR) study enrolled nearly 1,500 healthy women older than 65 years (excluding any with suspected dementia) who had been randomly assigned to one of two groups, either oral tamoxifen 20 mg/d or oral raloxifene 60 mg/d, as part of the STAR trial. The majority of women were enrolled in Co-STAR after having initiated their STAR trial medication, although a small number of women (n 273) had cognitive functioning assessment before starting SERM therapy. In this article, they report results from those women observed for at least 3 years who completed a focused, brief, neurocognitive battery at three assessment points roughly 12 months apart. Although the investigators had hypothesized that raloxifene would confer cognitive benefits compared with tamoxifen, they found little difference in cognitive performance between the tamoxifen and raloxifene groups and suggested that these two SERMs have a similar impact on cognitive functioning in this particular cohort of older women deemed at elevated risk for breast cancer. There are several important strengths of the Co-STAR study conducted by Legault and colleagues. For example, they undertook the challenging task of assessing cognitive function in a large scale, multisite, longitudinal study, paying close attention to the important issue of practice effects (ie, that repeated exposure to the same or similar tests often leads to higher performance after the initial assessment). They also attempted to measure meaningful domains of cognition that have been previously implicated in cognitive aging and also with chemobrain sufferers—learning and memory and executive functioning (ie, working memory and verbal fluency). A large sample size, a sound analytic strategy, and the investigators’ careful attention to potential confounds such as depression and affective style all lend weight to their conclusion that those women who received tamoxifen did not differ in neuropsychological performance from those that received raloxifene. Scores on the majority of neuropsychological tasks either stayed stable or improved over time (as would be expected based on practice effects). The one task on which a clinically significant decline in performance was noted was on a verbal learning and memory task that had been significantly modified including what the authors describe as the introduction of a more difficult form. However, the applicability of these findings beyond this population, for example to patients with breast cancer, is unclear. As the authors note, there may be group differences in lifetime and current estrogen exposure that may leave some groups primed to benefit from SERM administration (eg, those with osteoporosis) or, conversely, groups that are more likely to experience cognitive compromise. In terms of groups more vulnerable to cognitive compromise, this article does not address (and wasn’t intended to) an issue of considerable importance to the Journal of Clinical Oncology readership, which is whether SERM exposure during adjuvant systemic chemotherapy might be more likely to cause cognitive changes in women with breast cancer. Certainly the limited data available in the breast cancer studies that have examined adjuvant tamoxifen therapy does not suggest a neuroprotective effect of SERM exposure. Our group at University of California, Los Angeles has reported, in a cross-sectional study of long-term breast cancer survivors, that patients exposed to both tamoxifen and chemotherapy were more likely to experience cognitive compromise several years after treatment than were those who received no adjuvant therapy or who received chemotherapy alone and that this group is more likely to show altered neurophysiology, as assessed by brain positron emission tomography. A recent study of premenopausal breast cancer patients who did not receive any chemotherapy, but who had been taking tamoxifen for an average of 2.3 JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L