Comment
With a sample size of 693 women, KEEPS-Cog is the largest randomized, controlled trial of the effects of hormone therapy (HT) on cognition and mood in the early postmenopausal period. One strength of the study was the focus on cyclic HT regimens in which 200 mg m-P with either 0.45 mg oral CEE or 50 µg E2 was used on the first 12 days of each month.
Previous large-scale, randomized trials of CEE plus medroxyprogesterone acetate (CEE/MPA) have shown cognitive risks with MPA in older women. Specifically, continuous combined CEE/MPA doubled the risk of all-cause dementia and decreased verbal memory in older women (aged ≥65 y) and led to either near-significant or significant decreases in verbal memory in younger postmenopausal women.
Findings from the Women's Health Initiative Memory Study of Younger Women (women aged 50-55 y at time of randomization), however, showed neutral effects of CEE/MPA and CEE-alone on cognitive function assessed several years after stopping HT, suggesting that any negative effects of MPA on cognitive function in younger postmenopausal women are not enduring.
KEEPS-Cog showed neutral cognitive effects with either cyclic regimen while women were receiving treatment. Thus, these data suggest that these cyclic HT regimens are safe for cognitive function when used in the early postmenopausal period. Moreover, beneficial mood effects were found with oral CEE on some subscales in KEEPS-Cog, and E2 was associated with neutral mood and affective outcomes.
Many questions remain after publication of KEEPS-Cog data. Perhaps the most critically important unanswered question is whether HT improves cognition in women with moderate to severe vasomotor symptoms (VMS).
Hormone therapy is not indicated for the treatment or prevention of cognitive complaints or problems but is the gold-standard treatment for VMS. About 40% of women in KEEPS experienced moderate to severe VMS at baseline. Both HT treatment groups showed similarly large reductions in symptoms, and the magnitude of reduction was greater than in the placebo group at 12 months. Results indicated that group differences in VMS lessened over time because VMS decreased spontaneously for women in the placebo group. In addition to exerting direct effects on neural systems subserving cognition, HT could improve cognition in women with VMS indirectly by improving VMS and sleep.
Another remaining unanswered question is whether there are immediate or long-term benefits of HT for cognition and brain function that are not evident on cognitive tests. Three observational studies provide support for the view that timing of HT initiation is a significant determinant of risk for Alzheimer disease (AD) and that early initiation confers protection against AD. It is not feasible to do a randomized clinical trial of early use of HT and AD risk, but neuroimaging methods can be applied to evaluate AD biomarkers such as tau and beta-amyloid.
To date, there are no large-scale randomized trials of the effects of estrogen alone on cognitive function in younger postmenopausal women. Large-scale clinical trials in older women show neutral cognitive effects with estrogen regimens. Nonhuman primate studies suggest that neuroprotection occurs with cyclic E2 but not with E2 plus cyclic progesterone, continuous estrogen plus progesterone (either cyclic or continuous), or unopposed continuous estrogen. The effects of estrogen therapy when combined with a selective estrogen receptor modulator (SERM) such as bazedoxifene are also unknown but are of interest, given evidence from randomized clinical trials that the SERM raloxifene enhances verbal memory.
Last, the cyclic HT regimen using CEE improved scores on a depression subscale but not on a widely used depression inventory. That regimen also improved scores on an anxiety subscale, which is important given the link between VMS and anxiety. The basis and clinical significance of the more favorable mood effects with oral CEE than with E2 warrants further study.