SAN FRANCISCO—A study begun in 1969 examining cardiovascular disease and risk factors in adult male twins is yielding valuable insights into the relationship of sex hormones to brain structure and cognition in the same cohort of participants 35 years later.

Although there is a general consensus in the literature that higher concentrations of sex hormones are associated with better cognitive performance in older women, studies of older men are inconclusive. Furthermore, there are no data on the association of sex hormone concentrations and brain structure in men.

The results of this latest study on a subset of 349 adult male twins, who were between ages 68 and 80 when last examined (1995 to 1997), include several important findings. Of the four sex hormones studied, only one—estrone—was significantly associated with a measure of executive function and with brain volume. This is a departure from the published literature, in which the relationship of either estradiol or testosterone with cognitive function is reported much more frequently than is the relationship of estrone with cognitive function. In the present study, higher concentrations of estrone were significantly associated with smaller total brain volume, left hemisphere volume, and smaller left and right occipital volume, while higher concentrations of testosterone were associated with greater left occipital volume and left and right parietal volume.

Sex hormone concentrations were more often associated with measures of brain volume than with measures of cognitive task performance —results that have larger implications for current clinical practice in neurology, said Christina Lessov, PhD. “The battery of cognitive tests used [today] may not be sensitive enough to detect early stages of decrease in brain volume, particularly in otherwise healthy elderly men.” Dr. Lessov is a genetic epidemiologist at the Center for Health Sciences at SRI International, a not-for-profit research institute in Menlo Park, California. She presented the study at the 56th Annual Meeting of the American Academy of Neurology.

A “COMPLICATED BUNDLE” OF DATA

These latest findings may help to fill in some important blanks in an area of neuroscience in which much is already known about sex hormone concentrations and cognitive function, but where little is known about their relationship to brain morphology, especially in men. Extensive research has established that men and women differ in how well they excel in certain cognitive domains such as verbal performance and spatial reasoning. Because of the different composition of sex hormones between women and men, one active hypothesis is that differences in sex hormone concentrations may help account for the differences seen in cognitive performance. Furthermore, strong evidence now exists that hormone replacement therapy in postmenopausal women improves cognitive function and may delay cognitive decline—itself a powerful predictor of such neurodegenerative disorders as Alzheimer’s disease, Dr. Lessov pointed out.

“Thus, the general background [to the study] is a complicated bundle of a motivation to understand sex differences in behavior and cognition on the one hand and the role of sex hormones in age-related cognitive decline on the other,” she said.

TWIN TEST

Because of the availability of prospective data in a large sample of healthy aging men measuring repeat assessment of cognitive task performance and brain magnetic resonance imaging (MRI) at midlife and later life and sex hormone concentrations at midlife, Dr. Lessov and her colleagues believed they were in a good position to examine the relationship of sex hormone concentrations and cognitive performance. Their objective was to help clear up some of the inconsistencies in the literature on this relationship in men, while shedding some light on the unexplored relationship between sex hormone concentrations and brain structure.

The source of their data was ripe for this area of analysis. The National Heart, Lung, and Blood Institute (NHLBI) Twin Study is a collaborative, longitudinal study of a cohort of white, male twins born between 1917 and 1927, all of whom were veterans of World War II. The cohort was first examined between 1969 and 1972, when the twins were ages 42 to 55, and the men were subsequently examined between 1981 and 1982, 1986 and 1987, 1995 and 1997, and 1999 and 2001. Each time, the subjects completed questionnaires on their medical history and general health and underwent laboratory examinations. In the third data collection period, participants were assayed for serum blood hormone concentrations and underwent a battery of neuropsychological examinations. Additional cognitive tests of attention and memory were added, along with brain MRI, in the data collection that took place during the 1995 to 1997 period.

Although the researchers corrected for the presence of twins in the present study’s statistical analysis (data from pairs of twins violate the assumption that each data point is independent of every other data point), the twin design of the study did not factor significantly in the interpretation of the main results. However, Dr. Lessov and her colleagues—led by Dorit Carmelli, PhD, principal investigator of the NHLBI Twin Study, and Gary Swan, PhD, Director of the Center for Health Sciences at SRI International—plan to examine the data to see whether the relationship between sex hormone concentration and brain structure (the significant associations found in this study) is genetically influenced.

COMPLEX ASSOCIATIONS

In the analysis of the present study’s results, only one cognitive function test—measuring executive function—proved to be significant in establishing the complex association among cognitive function, sex hormone concentration, brain volume, and aging. “Cognitive function is a multidimensional construct,” Dr. Lessov explained. “It is assessed via cognitive performance tasks, which are assumed to reflect brain activity.

“Because sex hormones are so ubiquitous in their maintenance of normal physiologic function and because their concentration also decreases with aging,” Dr. Lessov continued, “if it could be reliably shown that declining sex hormone concentration is associated with decline in cognitive function and decreasing brain volume, this would provide a very logical prevention strategy—hormone supplementation.

“Cognitive decline and decreased brain volume are risk factors and precursors of neurodegenerative disorders such as Alzheimer’s disease. Executive function, compared to other cognitive domains, is particularly susceptible to aging effects.”

CONNECTING THE DOTS

Dr. Lessov then explained the findings related to executive function by reiterating the difference between positive and negative in terms of association between two variables. In the study, higher concentration of estrone was associated with longer time in completing the Trails B executive function test, thus constituting a negative association. Because higher estrone concentration was also associated with smaller total brain volume, left hemisphere volume, and left and right occipital volume, the “dots” in this case can be connected, according to Dr. Lessov.

“The estrone story makes some sense: Higher concentration of estrone is associated with worse executive function performance (but only on a single task) and smaller total and regional brain volume,” she said. “One would expect that smaller brain volume would be associated with worse cognitive task performance.”

OCCIPITAL OBSERVATIONS AND THE ESTRONE ENIGMA

What is the significance of the occipital lobes? The most prominent function of the occipital cortex is vision, Dr. Lessov pointed out. “The Trails B task involves drawing a continuous line between consecutive and alternating number and letter sequences—ie, 1-A-2-B, etc—which are placed randomly on a piece of paper. So there is a visual component of scanning the piece of paper to find the appropriate numbers and letters.

“The reason why this is an executive task is because there is processing of sequential information and an inhibition of the overlearned response of connecting numbers in consecutive order separately from letters. Executive function is primarily attributed to the frontal lobes, and we see no association of estrone with frontal lobe volume. So, though the task is considered an executive function task, there are other brain mechanisms that are recruited and necessary for task performance, such as vision,” she said.

As to why estrone emerged as the sex hormone of greatest significance in this study, the explanation is unclear, and Dr. Lessov and her colleagues are now pursuing this line of inquiry; a manuscript with updated and more thorough analysis is near completion. What is clear is that the prospect of hormone supplementation for men is not likely to become a practical reality any time soon, she said.

NR

—Fred Balzac

Return to table of contents