Examining Sex Differences in Markers of Cognition and Neurodegeneration in Autosomal Dominant Alzheimer’s Disease: Preliminary Findings from the Colombian Alzheimer’s Prevention Initiative Biomarker Study

INTRODUCTION

Epidemiological studies suggest that females are disproportionately affected by Alzheimer’s disease (AD) [1]. Although AD incidence increases with age [2] and more females survive to late-life than males [3], there may also be a sex-specific biological risk for AD. Growing evidence suggests that females have greater AD pathology burden, as evidenced by postmortem studies [4, 5], in vivo cerebrospinal fluid [6], and neuroimaging studies [7, 8], and that they are more susceptible to AD-related pathology. As levels of AD-related pathology increase, females show greater tau accumulation [7], faster hippocampal volume loss [8], and faster cognitive decline [9] and progression [5] than males. Possible mechanisms underlying sex differences in biological risk have been suggested, including endocrinological [10], cardiovascular [1, 11], or heightened inflammatory responses [12].

However, previous studies investigating sex differences in AD-related pathology and cognitive performance have yielded mixed findings. For instance, Caldwell and colleagues [13] did not find sex differences in hippocampal volume among cognitively normal older adults, while other studies showed that cognitively normal females had greater hippocampal volumes than their male counterparts [14, 15]. Notably, Sundermann and colleagues [15, 16] found that among individuals with mild cognitive impairment, females had better verbal memory than men despite similar levels of medial and inferior temporal glucose metabolism, and hippocampal volume, suggesting that females may have a sex-specific form of cognitive reserve. Thus, sex differences in clinical and pathological manifestations of AD are not well understood.

To our knowledge, no study to date has investigated sex differences in individuals with autosomal dominant AD. Our group follows individuals from the world’s largest autosomal dominant AD kindred due to a single mutation (E280A) in the Presenilin1 gene (PSEN1). Importantly, this mutation is fully penetrant and not sex-linked, meaning that males and females are equally affected. PSEN1 mutation carriers are genetically determined to develop early-onset dementia, and although the pathogenesis of autosomal dominant AD is different from late-onset AD, these conditions have similar clinical, cognitive, and physiological profiles [17]. PSEN1 mutation carriers have a well-characterized clinical profile, with mild cognitive impairment symptoms emerging at a median age of 44 years old (95% C.I. = 43, 45 years) and dementia at 49 years old (95% C.I. = 49, 50 years) [18, 19]. AD-pathological changes have also been characterized in PSEN1 mutation carriers, with elevated cortical amyloid accumulation 15 years before clinical symptom onset (i.e., mild cognitive impairment) and elevated levels of tau in medial temporal lobe regions (i.e., entorhinal cortex and inferior temporal lobe) 6 years before clinical symptom onset [17, 20, 21].

Thus, this extraordinary cohort allows us to investigate sex differences along the AD trajectory from preclinical to clinical stages. Moreover, PSEN1 carriers are virtually destined to develop early-onset dementia and thus, have few age-related factors that confer risk for AD and are known to vary by sex such as cardiovascular disease [1 , 22], or hormonal changes [10, 23]. Lastly, examining sex differences in autosomal dominant AD limits common methodological challenges that may confound research into sex differences in sporadic AD, such as survival bias due to differences in mortality among males and females [24].

For the purpose of this study, we leveraged data from the Alzheimer’s Prevention Initiative Biomarker study [20] to retrospectively examine sex differences in global cognition, verbal memory (i.e., learning, delayed recall, and recognition in a list-learning task), and hippocampal volume ratio in PSEN1 mutation carriers. We hypothesized that cognitively intact female and male mutation carriers would not differ in hippocampal volume or global cognition, while cognitively intact female mutation carriers would exhibit better verbal memory. When examining sex differences across disease stages by also including symptomatic carriers, we further hypothesized that female mutation carriers would exhibit worse global cognition, verbal memory delayed recall, and lower hippocampal volume than male mutation carriers.

METHODS

RESULTS

DISCUSSION

Growing evidence suggests that females may be more susceptible to AD-related pathology [5, 7–9 , 35]. However, sex differences in clinical and pathological manifestations of AD remain poorly understood. To our knowledge, this is the first study to investigate sex differences in markers of cognition and neurodegeneration in carriers of an autosomal dominant Alzheimer’s disease mutation who are destined to develop early-onset AD by midlife [18, 19]. We leveraged data from the world’s largest kindred with a single mutation (E280A) in the Presenilin1 gene (PSEN1). Importantly, PSEN1 carriers develop mild cognitive impairment at a median age of 44 years and dementia at a median age of 49 years [18], and thus have few age-related risk factors for AD at symptom onset that are known to vary by sex (i.e., life expectancy, cardiovascular disease, hormonal changes).

OPEN IN VIEWER

Fig. 1

Relations among age, cognition, and neurodegeneration in PSEN1 male and female carriers. A) Verbal memory learning as a function of age. B) Verbal memory delayed recall as a function of age. C) Verbal memory recognition as a function of age. D) CERAD Total score as a function of age. E) Hippocampal volume ratio as a function of age. Green dots represent male PSEN1 mutation carriers and orange triangles represent female PSEN1 mutation carriers.

OPEN IN VIEWER

Fig. 2

Relations among neurodegeneration and cognition in PSEN1 male and female carriers. A) Verbal memory learning as a function of hippocampal volume ratio. B) Verbal memory delayed recall as a function of hippocampal volume ratio. C) Verbal memory recognition as a function of hippocampal volume ratio. D) CERAD Total scores as a function of hippocampal volume ratio. Green dots represent male PSEN1 mutation carriers and orange triangles represent female PSEN1 mutation carriers.

In autosomal dominant AD, age is commonly used as a proxy of disease progression, because as carriers age, they get closer to their age of estimated symptom onset. Therefore, we examined the relationship between age and markers of cognition and neurodegeneration. We found that among cognitively unimpaired male and female carriers, age was not associated with markers of cognition or neurodegeneration, perhaps due to limited age range in this group. When examining this relationship across all carriers, older age was associated with lower hippocampal volume in both male and female carriers, while older age was associated with worse global cognition and verbal memory in female carriers only. Of note, older age was also associated with worse cognition and verbal memory in male carriers but these associations did not reach significance. Nonetheless, the magnitude of these associations was not statistically different among male and female carriers. This finding may be driven by the larger number of symptomatic female carriers resulting in a greater range of cognitive performance.

We examined sex differences in markers of cognition and preliminary findings show that cognitively unimpaired female carriers showed better global cognition and verbal memory learning than cognitively unimpaired male carriers. This finding is consistent with previous extensive literature showing a verbal memory advantage for females [36]. However, we did not find a difference in word list delayed recall or recognition variables, perhaps due to restricted range in performance. With regard to the sex difference on global cognition, it is important to note that global cognition was measured using the CERAD total score, an aggregate of six CERAD subtests that also includes word list learning, thus likely contributing to the effect found. Non-carrier females also showed slightly better cognitive performance than non-carrier males, although differences were not statistically significant. Again, restricted range of performance in the CERAD may have limited our ability to detect differences, particularly among cognitively unimpaired individuals. In contrast, when examining all carriers by also including symptomatic carriers, cognitive performances did not differ among males and females. This may suggest that this verbal memory advantage may only be present in earlier stages of the disease, as previously shown [37], although further examination in samples with larger number of symptomatic carriers is required.

We also examined sex differences in markers of neurodegeneration and found that males and females had similar levels of hippocampal volume ratio among cognitively unimpaired and symptomatic carriers. This finding does not support our hypothesis, as we posited that as disease progresses and pathology accumulates, female carriers would exhibit worse levels of neurodegeneration. However, it is consistent with previous reports that found no sex-specific differences in hippocampal volume preservation among cognitively normal individuals [13]. Nonetheless, it is important to highlight that previous studies show great variability in findings wherein some studies found that females showed greater brain resilience to tau deposition [38], while others found that females showed lower brain resilience [35]. Further research is needed to clarify whether females show greater resilience or susceptibility to AD-related pathology than males, and whether this effect varies across disease stages.

With regard to the effect of sex moderating the relationship between neurodegeneration and cognitive performance, we found that there was no interaction between sex and hippocampal volume ratio in predicting cognitive performance among cognitively unimpaired carriers. Similarly, across all carriers, sex by hippocampal volume interaction did not predict verbal memory learning, delayed recall, or global cognition. However, there was a trend toward significance, wherein female carriers with lower hippocampal volume ratio had worse verbal memory recognition than male carriers. Overall, contrary to our hypotheses, our findings show no evidence that females with this autosomal dominant AD mutation have greater cognitive susceptibility to AD-related neurodegeneration compared to male carriers. Our ability to detect such an effect may have been limited by the relatively low number of symptomatic carriers, particularly with mild dementia. Alternatively, the sex differences previously found in sporadic AD may be linked to factors that are not relevant in this unique cohort of autosomal dominant AD, such as older age [2, 19], cardiovascular disease [1 , 22], menopause [10, 23], or survival bias (i.e., mortality differences in males and females) [24].

This study has limitations and thus, findings should be interpreted with caution. First, this is a retrospective, cross-sectional study with a small sample size and relatively few symptomatic carriers, which may have resulted in limited power to detect sex differences, particularly in later stages of the disease. Additionally, there were a larger number of females in the sample and a greater number of symptomatic female carriers. The generalizability of these findings to others at risk for autosomal dominant AD or late-onset sporadic AD requires further investigation. However, despite the small sample size/power limitations, which are generally inherent in autosomal dominant AD research, this cohort offers a unique opportunity to examine sex differences using neuroimaging methods with fewer age- and sex-related confounds than in research with sporadic AD (i.e., mortality/life expectancy, cardiovascular disease, or menopause).