Abstract
INTRODUCTION
We investigated how the associations between tau and cognitive measures differ by sex in the preclinical Alzheimer's disease (AD) stage.
METHODS
A total of 343 cognitively unimpaired, amyloid‐positive individuals (205 women, 138 men) who self‐identified as non‐Hispanic White from the Anti‐Amyloid Treatment in Asymptomatic Alzheimer's (A4) Study were included. We assessed sex‐stratified associations between 18F‐flortaucipir positron emission tomography (PET) standardized uptake value ratio (SUVR) in the meta‐temporal region and Preclinical Alzheimer's Cognitive Composite (PACC) and Computerized Cognitive Composite (C3) components.
RESULTS
We observed that higher tau level was significantly associated with worse cognitive performance only in women: PACC and its components except for Mini‐Mental State Examination (MMSE) and C3 components: First Letter Name Recall (FNLT) and One‐Card Learning Reaction Time (OCL RT). These associations except for FNLT were apolipoprotein E (APOE) ε4 independent.
DISCUSSION
Women show stronger associations between tau PET and cognitive outcomes in preclinical AD. These findings have important implications for sex‐specific tau‐targeted preventive AD clinical trials.
Highlights
The tau positron emission tomography (PET) signal in the meta‐temporal region was associated with poor cognitive performance in preclinical Alzheimer's disease (AD).
After sex stratification, the associations between regional tau PET and cognitive outcomes were observed only in women.
The associations between tau PET and some cognitive outcomes were independent of apolipoprotein E (APOE) ε4.
Keywords: Alzheimer's disease, APOE ε4, C3, cognition, PACC, sex differences, tau PET
1. BACKGROUND
Sex differences in Alzheimer's disease (AD) are widely recognized. Women have a higher prevalence of AD compared to men, even when differences in longevity are considered. 1 In addition there are differences in the cognitive trajectories Likely due in part to the female advantage in verbal memory, women are more likely to experience the delayed diagnosis of amnestic mild cognitive impairment (MCI), a precursor condition to AD dementia, but show more accelerated cognitive decline thereafter. 2 , 3 , 4 Cognitively normal older men have also been reported to outperform women on some visuospatial tasks and maintain this advantage after AD diagnosis, although findings have not been consistent across studies, and appear to be task specific. 5 , 6 , 7 These measures have been shown on standard paper‐and‐pencil tests often given in clinics or clinical trials.
Accumulation of abnormal tau deposition is a hallmark pathological feature of AD, which precedes cognitive decline and is a potential target for AD clinical trials. Tau pathology measured using positron emission tomography (PET) better predicts cognitive decline compared to brain volume and amyloid PET and might be a prognostic marker in preclinical and prodromal AD. 8 , 9 , 10 , 11 Greater regional tau deposition is associated with poor cognitive performance and faster cognitive decline in the context of normal aging, especially in amyloid‐positive older adults. 12 , 13 , 14 Sex differences in tau pathology among older individuals with normal cognition or MCI have also been reported, 3 , 15 , 16 , 17 and women with high amyloid burden exhibit higher regional tau PET burden than age‐matched men. Previous studies demonstrated sex differences in the relationships between regional tau burden and cognition, with women showing a stronger relationship between regional tau and verbal memory in the MCI stage 3 and faster cognitive decline in the combined cognitively normal (CN) and MCI group. 18 However, less is known about how the associations between tau PET and cognitive measures differ by sex in preclinical AD.
The Anti‐Amyloid Treatment in Asymptomatic Alzheimer's (A4) Study is an ongoing prevention trial in preclinical AD endeavoring to slow cognitive decline among cognitively normal older participants who are at risk of AD. 19 Baseline, pre‐randomization data is currently available for analysis. Preclinical Alzheimer's Cognitive Composite (PACC), a multidomain composite paper and pencil measure believed to detect early signs of cognitive impairment, 20 is the primary outcome of the A4 Study. A previous study with clinically normal, amyloid‐positive adults reported that women outperformed men on PACC and individual components. 21 In addition to standard paper‐and‐pencil tasks, A4 also includes the Computerized Cognitive Composite (C3), self‐administered, tablet‐based cognitive measures developed to reduce data administration, score mistakes and site burden. A recent study has shown that C3 was moderately correlated with PACC and had the potential to identify subtle cognitive decline in preclinical AD. 22 Women and men have also been reported to show different advantages in C3 components. 22
This study aimed to examine sex differences in the associations between regional tau PET and cognitive outcomes in preclinical AD. We included cognitively normal amyloid‐positive individuals who self‐identified as non‐Hispanic White. First, we examined whether there are sex differences in the PACC and C3 components. Then we assessed the associations between tau PET and cognitive outcomes and how the associations differ by sex. We expected that women and men would show advantages in different cognitive performances: Women might outperform men on verbal memory and men performed better on processing speed. We also hypothesized that women would show stronger relationships between tau PET and some cognitive outcomes in memory and that these associations would be independent of the strongest genetic risk factor for sporadic AD, the apolipoprotein E (APOE) ε4 allele.
2. METHODS
2.1. Participants
The A4 Study is an ongoing multicenter clinical trial conducted in four countries—Australia, Canada, Japan, and the United States—among cognitively normal participants, from 65 to 85 years of age, funded by the National Institute on Aging/National Institutes of Health (NIH), Eli Lilly and Company, and several philanthropic organizations. 22 We included 343 amyloid‐positive individuals (205 women, 138 men) from the A4 study who had available 18F‐flortaucipir (FTP) PET data and completed the PACC and C3 tests. Given the disparities of cognition by race and ethnicity 23 , 24 and valid race and ethnicity cognitive norms being unavailable in the A4 Study, we included only participants who self‐identified as non‐Hispanic White, since they were the majority of the sample. Participants completed the PACC and C3 at their first visit and then underwent 18F‐florbetapir PET scans for the classification of amyloid status before enrollment at their second visit. Potential participants completed alternative C3 tests on their third visit and were informed whether they were qualified to participate in the clinical trials for amyloid positive or negative. The time interval between C3 data collection at the first visit and the third visit is about 3 months. The tau PET scans were acquired on their fourth visit.
2.2. Image processing
The preprocessed FTP PET images were downloaded from the Laboratory of NeuroImaging website (https://loni.usc.edu). FTP PET data were acquired at 80 to 110 minutes (6 × 5 min frames) after tracer injection. Preprocessed FTP PET images were realigned, summed, and co‐registered to the corresponding bias‐corrected T1 scans created by FreeSurfer (version 5.30). Standardized uptake value ratio (SUVR) images were created using the inferior cerebellum gray matter as the reference region. A composite meta‐temporal region (entorhinal, amygdala, fusiform, inferior temporal, and middle temporal region defined by FreeSurfer) was the region of interest in this study. Amyloid PET data were downloaded from the A4 website. Amyloid SUVR was generated using the whole cerebellum grey matter as the reference region. Determination of amyloid eligibility has been described elsewhere. 25 Briefly, it was made using an algorithm incorporating both quantitative (amyloid SUVR >1.15) and qualitative (visual read) reviews.
2.3. Cognitive tests
All cognitive scores were downloaded directly from the A4 website. In this study, primary analyses were performed on the PACC and C3 at Visit 1. C3 at Visit 3 was also analyzed, and the results are summarized in the Supplemental Materials.
2.3.1. PACC
PACC is a composite of multi‐domain neuropsychological tests and is the primary outcome of the A4 Study. PACC includes four tests: Mini‐Mental State Examination (MMSE, range 0–30), Logical Memory Delayed Recall (LMDR, range 0–25), Digit‐Symbol Coding Test (DSC, range 0–93), and the Free and Cued Selective Reminding Test–Free + Total Recall (FCSRT96, range 0–96). The total score of PACC is calculated as the sum of the normalized z scores of these four measures. 26 In this study, we also evaluated Free Recall, a portion of FCSRT96, which was reported to be sensitive to amyloid‐related cognitive change. 19
RESEARCH IN CONTEXT
Systematic review: We performed the literature review using Google Scholar and PubMed. Sex differences exist in Alzheimer's disease (AD). In preclinical AD, tau positron emission tomography (PET) is associated with poor cognitive performance. However, less is known about how these associations differ by sex in preclinical AD.
Interpretation: Among cognitively normal participants with amyloid‐positive status, we found that a higher tau PET level in the meta‐temporal region was associated with worse cognitive performance only in women. These associations were independent of apolipoprotein E (APOE) ε4 status.
Future directions: Our work emphasized the stronger relationships between tau PET and cognition in women. To expand the findings, future studies should assess the associations between tau PET and cognitive decline when longitudinal cognitive data are available and replicate these findings in other cohorts.
2.3.2. C3
C3 includes measures from the Behavioral Pattern Separation task‐Object (BPS‐O), which evaluates pattern separation performance, an aspect of episodic memory, the Face Name Associative Memory Exam (FNAME), which tests associative memory and has been shown to be sensitive to early AD‐related changes, and the Cogstate Brief Battery (CBB), which uses playing‐card stimuli to assess the features of memory and attention. 22 , 27 , 28 Detailed descriptions of these three measures have been described elsewhere. 22 In BPS‐O, the lure discrimination index (LDI) is the primary outcome with a higher value indicating better pattern separation performance. FNAME includes three measures of memory: face recognition (FSBT), first letter name recall (FNLT), and face‐name matching (FNMT). The accuracy is the outcome for each measure. In CBB, there are tests of Detection (DET), Identification (IDN), One‐Back Test (ONB), and One‐Card Learning (OCL). Reaction Time (RT) of DET, RT and accuracy of IDN, and ONB and OCL were the outcomes of CBB. For RT measures, a log10 transformation is used. For accuracy measures, an arcsin sqrt transformation is used. Completion and performance checks were applied for C3 components. 22
Table S1 summarizes the cognitive outcomes and specific domains they test. Higher values represent better performance in these cognitive outcomes except DET, IDN RT, ONB RT, and OCL RT. ,
TABLE 1.
Characteristics of participants.
| All (N = 343) | Men (N = 138) | Women (N = 205) | P | |
|---|---|---|---|---|
| Age, mean (SD) | 71.9 (4.7) | 72.9 (4.8) | 71.3 (4.6) | 0.002** |
| Education, mean (SD) | 16.2 (2.9) | 16.8 (3.0) | 15.7 (2.7) | 0.001** |
| APOE ε4 genotype, N (%) | N = 338 | N = 134 | N = 204 | 0.864 |
| 139 (41.1) | 60 (44.8) | 79 (38.7) | ||
| Amyloid_SUVR mean (SD) | 1.325 (0.149) | 1.328 (0.150) | 1.323 (0.149) | 0.579 |
| Meta_Temporal_SUVR, mean (SD) | 1.225 (0.120) | 1.217 (0.107) | 1.230 (0.128) | 0.196 |
*P < 0.05.
**P < 0.005.
TABLE 2.
Cognitive outcomes difference between men and women.
|
Cognitive outcomes (Women N vs. Men N) |
Women | Men | P | Cohen's d | |
|---|---|---|---|---|---|
| PACC and components | |||||
| PACC (205 vs 138) | −0.003 (2.595) | −1.474 (2.824) | 0.000*** | 0.580 | |
| MMSE (205 vs 138) | 28.78 (1.157) | 28.304 (1.507) | 0.001** | 0.442 | |
| LMDR (205 vs 138) | 11.459 (3.209) | 11.587 (3.702) | 0.719 | 0.024 | |
| DSC (205 vs 138) | 44.249 (9.374) | 38.986 (8.226) | 0.000*** | 0.511 | |
| FCSRT96 (205 vs 138) | 76.585 (6.411) | 73.442 (5.648) | 0.000*** | 0.452 | |
| Free recall (205 vs 138) | 29.244 (5.922) | 26.341 (5.203) | 0.000*** | 0.462 | |
| C3 components | |||||
| BPS‐O | LDI (193 vs 126) | 0.398 (0.186) | 0.367 (0.209) | 0.230 | 0.134 |
| FNAME | FNLT (192 vs 117) | 0.626 (0.195) | 0.541 (0.191) | 0.000*** | 0.428 |
| FNMT (190 vs 124) | 1.007 (0.178) | 0.948 (0.198) | 0.006* | 0.308 | |
| FSBT (196 vs 126) | 1.326 (0.227) | 1.201 (0.256) | 0.000*** | 0.470 | |
| CBB | DET (195 vs 128) | 2.613 (0.101) | 2.583 (0.099) | 0.007* | 0.301 |
| IDN_RT (194 vs 127) | 2.774 (0.074) | 2.782 (0.081) | 0.561 | −0.065 | |
| IDN_Acc (194 vs 127) | 1.424 (0.149) | 1.402 (0.152) | 0.216 | 0.138 | |
| ONB_Acc_ (195 vs 127) | 1.362 (0.163) | 1.328 (0.163) | 0.061 | 0.208 | |
| ONB_RT (195 vs 127) | 2.959 (0.089) | 2.979 (0.097) | 0.129 | −0.169 | |
| OCL_Acc (196 vs 128) | 0.953 (0.117) | 0.946 (0.121) | 0.580 | 0.061 | |
| OCL_RT (196 vs 128) | 3.076 (0.102) | 3.081 (0.092) | 0.692 | −0.044 | |
Note: Comparisons were adjusting for age and education.
Abbreviations: Acc, accuracy; BPS‐O, behavioral pattern separation task‐object; CBB, Cogstate brief battery; DET: detection; DSC, digit‐symbol coding test; FCSRT96, free and cued selective reminding test–free + total recall; FNAME, face name associative memory exam; FNLT, first letter name recall; FNMT, face‐name matching; FSBT, face recognition; IDN, identification; LDI, lure discrimination index; LMDR, logical memory delayed recall; MMSE, Mini‐Mental State Examination; OCL: one‐card learning; ONB: one‐back test; PACC, Preclinical Alzheimer's Cognitive Composite; RT, reaction time.
*P < 0.007.
**P < 0.001.
2.4. Statistics
Age and education differences between men and women were compared using independent t‐tests. Pearson's chi‐square tests were used to detect the sex differences in APOE ε4 carriership. Sex differences in amyloid SUVR, tau SUVR in the meta‐temporal region, and cognitive outcomes (PACC and C3 components) were assessed using linear regression models, adjusting for age and education. The effect sizes for sex differences in cognitive outcomes were also calculated by Cohen's d, adjusting for age and education. To detect the associations between tau SUVR in the meta‐temporal region and cognitive outcomes in the whole sample, standardized linear regression models were applied adjusting for sex, age, and education. We used a series of linear regression models with interaction terms to assess the interaction effects of sex and tau SUVR in the meta‐temporal region on the cognitive outcomes, adjusting for age and education. To interpret the significant interaction effects, standardized linear regression was performed to assess the associations between tau SUVR in the meta‐temporal region and cognitive outcomes in men and women separately, adjusting for age and education. Similar analyses with APOE ε4 carriership as additional covariates were also performed after excluding participants with unavailable APOE data. A significant threshold α < 0.007 (0.05/7) was used for correcting multiple comparisons. All the analyses in this study were performed in R (version 4.0.4).
3. RESULTS
3.1. Participants
Characteristics of participants were presented in Table 1. Women were on average younger and less educated than men. There were no significant sex differences in APOE ε4 carriership, amyloid SUVR, and tau SUVR in the meta‐temporal region.
3.2. Sex differences in the PACC and C3 components
Women outperformed men on PACC (P < 0.001) and its components: MMSE (P = 0.001), DSC (P < 0.001), and FCSRT96 (P < 0.001) (Table 2). Women also performed better than men on the Free Recall (P < 0.001). For C3 components, women outperformed men on FNAME components: FNLT (P < 0.001), FNMT (P = 0.006), and FSBT (P < 0.001) (Table 2). However, men outperformed women on DET (P = 0.0068). After controlling for APOE ε4, FNMT and DET did not survive after multiple comparison corrections (Table S2). For C3 components at Visit 3, women only outperformed men on FSBT (Table S3). Sex differences in the PACC and the DSC displayed the largest effect size (Table 2).
TABLE 3.
Association between tau PET (meta temporal) and cognitive outcomes.
| Cognitive outcomes (N) | Standardized β | SE | P | |
|---|---|---|---|---|
| PACC and components | ||||
| PACC (N = 343) | −0.222 | 0.047 | 0.000*** | |
| MMSE (N = 343) | −0.116 | 0.052 | 0.026 | |
| LMDR (N = 343) | −0.180 | 0.053 | 0.001** | |
| DSC (N = 343) | −0.113 | 0.048 | 0.020 | |
| FCSRT96 (N = 343) | −0.172 | 0.050 | 0.001** | |
| Free recall (N = 343) | −0.156 | 0.050 | 0.002** | |
| C3 components | ||||
| BPS‐O | LDI (N = 319) | −0.116 | 0.056 | 0.037 |
| FNAME | FNLT (N = 309) | −0.151 | 0.054 | 0.006* |
| FNMT (N = 314) | −0.010 | 0.056 | 0.855 | |
| FSBT (N = 322) | −0.105 | 0.053 | 0.049 | |
| CBB | DET (N = 323) | 0.013 | 0.056 | 0.812 |
| IDN_RT (N = 321) | −0.034 | 0.056 | 0.543 | |
| IDN_Acc (N = 321) | −0.109 | 0.055 | 0.048 | |
| ONB_Acc (N = 322) | −0.141 | 0.054 | 0.010 | |
| ONB_RT (N = 322) | 0.048 | 0.055 | 0.385 | |
| OCL_Acc (N = 324) | −0.118 | 0.054 | 0.029 | |
| OCL_RT (N = 324) | 0.124 | 0.055 | 0.025 | |
Note: Standardized regression coefficients β, SE and P‐values in the association between tau PET (meta temporal) and cognitive components, adjusting for age, sex and education.
Abbreviations: Acc, accuracy; BPS‐O, behavioral pattern separation task‐object; CBB, Cogstate brief battery; DET: detection; DSC, digit‐symbol coding test; FCSRT96, free and cued selective reminding test–free + total recall; FNAME, face name associative memory exam; FNLT, first letter name recall; FNMT, face‐name matching; FSBT, face recognition; IDN: identification; LDI, lure discrimination index; LMDR, logical memory delayed recall; MMSE, mini‐mental state examination; OCL: one‐card learning; ONB: one‐back test; PACC, preclinical Alzheimer cognitive composite; RT, reaction time.
*P < 0.007.
**P < 0.001.
3.3. Associations between tau SUVR and PACC and C3 components
Higher tau SUVR in the meta‐temporal region was significantly associated with poorer cognitive performance: PACC (standardized β = −0.222, P < 0.001) and its components LMDR (standardized β = −0.180, P = 0.001), and FCSRT96 (standardized β = −0.172, P = 0.001), as well as Free Recall (standardized β = −0.156, P = 0.002) (Table 3). For C3 components, only FNLT (standardized β = −0.151 P = 0.006) was associated with tau SUVR in the meta‐temporal region (Table 3). While adjusting for APOE ε4, significant associations still exist in PACC and its components LMDR and FCSRT96 (Table S4). For C3 components at Visit 3, tau SUVR in the meta‐temporal region was significantly associated with LDI (standardized β = −0.170, P = 0.002) and FNMT (standardized β = −0.165, P = 0.002) (Table S5).
3.4. Association between tau SUVR and cognitive outcomes stratified by sex
We assessed the interaction effects of sex by tau SUVR in the meta‐temporal region on the cognitive outcomes and found that the interaction P values < 0.05 were in the PACC component of DSC (interaction β = −17.690, P = 0.028) and C3 components of DET (interaction β = 0.202, P = 0.046) and OCL RT (interaction β = 0.207, P = 0.033) (Table 4). Similar results were also found for OCL RT at Visit 3 (interaction β = 0.183, P = 0.045) (Table S6). After adjusting for APOE ε4 carriership, we still observed the interaction effects of sex by tau in DSC and OCL RT (Table S7). Interaction effects did not survive correction for multiple comparisons.
TABLE 4.
Interaction effects of tau PET (meta temporal) and sex on cognitive outcomes.
| Cognitive Outcomes (N) | Interaction β | SE | P | |
|---|---|---|---|---|
| PACC and components | ||||
| PACC (N = 343) | −3.542 | 2.341 | 0.131 | |
| MMSE (N = 343) | 0.514 | 1.235 | 0.678 | |
| LMDR (N = 343) | −2.986 | 3.271 | 0.362 | |
| DSC (N = 343) | −17.690 | 8.021 | 0.028 . | |
| FCSRT96 (N = 343) | −6.299 | 5.658 | 0.266 | |
| Free recall (N = 343) | −7.588 | 5.263 | 0.150 | |
| C3 components | ||||
| BPS‐O | LDI (N = 319) | 0.016 | 0.199 | 0.935 |
| FNAME | FNLT (N = 309) | −0.139 | 0.198 | 0.483 |
| FNMT (N = 314) | −0.150 | 0.191 | 0.432 | |
| FSBT (N = 322) | −0.211 | 0.235 | 0.370 | |
| CBB | DET (N = 323) | 0.202 | 0.101 | 0.046 . |
| IDN_RT (N = 321) | 0.005 | 0.078 | 0.950 | |
| IDN_Acc (N = 321) | −0.157 | 0.148 | 0.290 | |
| ONB_Acc (N = 322) | 0.067 | 0.159 | 0.673 | |
| ONB_RT (N = 322) | 0.091 | 0.091 | 0.318 | |
| OCL_Acc (N = 324) | −0.060 | 0.115 | 0.600 | |
| OCL_RT (N = 324) | 0.207 | 0.097 | 0.033 . | |
Note: Coefficients of tau SUVR in the meta temporal region and sex interaction on PACC and components from the multiple linear regression models, adjusting for age and education.
Abbreviations: Acc: accuracy; BPS‐O, behavioral pattern separation task‐object; CBB, Cogstate brief battery; DET, detection; DSC, digit‐symbol coding test; FCSRT96, free and cued selective reminding test–free + total recall; FNAME, face name associative memory exam; FNLT, first letter name recall; FNMT, face‐name matching; FSBT, face recognition; IDN: identification; LDI, lure discrimination index; LMDR, logical memory delayed recall; MMSE, mini‐mental state examination; OCL: one‐card learning; ONB: one‐back test; PACC, preclinical Alzheimer cognitive composite; RT, reaction time; SE, standard error.
. P < 0.05.
After sex stratification, we observed that higher tau SUVR in the meta‐temporal region was significantly associated with DSC (standardized β = −0.204, P = 0.002) and OCL RT (standardized β = −0.218, P = 0.002) in women, not in men (Figure 1D,Q). Besides, in women, higher levels of tau in the meta‐temporal region were also associated with poorer cognitive performance: PACC (standardized β = −0.305, P < 0.001), LMDR (standardized β = −0.239, P = 0.001), FCSRT96 (standardized β = −0.213, P = 0.002), Free Recall (standardized β = −0.208, P = 0.002), and FNLT (standardized β = −0.192, P = 0.007). No significant associations were found in men. FNLT was not significant after controlling for APOE ε4 in women (Table S8). For C3 components at Visit 3, tau SUVR was also significantly associated with LDI (standardized β = −0.193, P = 0.006) and FNMT (standardized β = −0.199, P = 0.004) (Figure S1).
FIGURE 1.
Associations between of tau positron emission tomography (PET; meta temporal) and cognitive outcomes in men and women separately. Scatter plots showing the associations between tau standardized uptake volume ratio (SUVR) in meta temporal region and cognitive outcomes in men and women separately. Red, women; blue, men. PACC: Preclinical Alzheimer's Cognitive Composite; MMSE, Mini‐Mental State Examination; LMDR, logical memory delayed recall; DSC, digit‐symbol coding test; FCSRT96, free and cued selective reminding test–free + total recall. BPS‐O, behavioral pattern separation task‐object; LDI, lure discrimination index; FNAME, face name associative memory exam; FSBT, face recognition; FNLT, first letter name recall; FNMT, face‐name matching; CBB, Cogstate brief battery; DET, detection; IDN, identification; ONB, one‐back test; OCL, one‐card learning; RT, reaction time; Acc, accuracy. Standardized regression coefficients β, SE, and P values are labeled in the upper right of plots (red: women, blue: men). *P < 0.007. **P < 0.001.
4. DISCUSSION
In this study with cognitively normal, amyloid‐positive older non‐Hispanic White participants, we found sex differences in PACC and C3 components. Women outperformed men on PACC and the components of MMSE, DSC, and FCSRT96, APOE ε4, which was consistent with a previous study with a larger sample size not limited to non‐Hispanic White participants. 21 Sex differences in C3 components have also been reported among cognitively normal individuals: women outperformed men on all FNAME components including FNLT, FNMT, FSBT, IDN accuracy, and ONB accuracy, whereas men outperformed women on DET. Consistent with this eariler finding, in amyloid‐positive participants in the current study, women outperformed men on all FNAME cognitive outcomes that measure episodic memory and men performed better in DET that measures psychomotor function. The sex differences in FNAME were APOE ε4 independent. However, we did not detect differences in IDN accuracy and ONB accuracy between men and women with or without controlling for APOE ε4. Amyloid‐positive individuals performed worse on IDN accuracy and ONB accuracy than amyloid‐negative participants, 22 so the sex differences in IDN and ONB accuracy might be amyloid dependent.
Tau PET has been shown to be a strong predictor of cognitive performance along the full AD spectrum including preclinical and prodromal AD. 13 , 29 , 30 Our results also support the link between greater tau pathology and poorer cognitive performance in preclinical AD. We further extended these findings by indicating that these relationships were found only in women. With weak sex by tau interactive effects, women showed significant associations between regional tau PET signal and cognitive outcomes on DSC and OCL RT, which mainly measured psychomotor speed. Because we did not detect sex differences in the extent of tau burden in the meta‐temporal region, our findings suggest that the same level of tau accumulation in this region might contribute more to cognitive difficulties in psychomotor function in women than in men with preclinical AD. Kotler's group has reported that women have an advantage in DSC compared to men (Cohen's d = 0.63, controlling for education) 31 in a population at a younger age (age range: 20–44 years). In this study with older participants in preclinical AD, women outperformed men on DSC with a smaller effect size (Cohen's d = 0.51, controlling for age and education) compared to Kotler's study. The reduced effect size of sex differences in DSC in our study might be explained by the women‐specific relationship between regional tau pathology and cognitive outcome. When longitudinal cognitive data for the A4 Study is available, we also plan to assess whether sex could modify the association between tau PET and longitudinal cognitive decline in PACC and C3. Besides, more extensive work will be undertaken to explore additional regions. In addition, sex‐stratified analyses also found significant associations between greater tau burden and memory deficits of PACC and C3 components in women. Without interaction effects, it is hard to conclude that these associations were sex specific. Future studies with larger sample sizes are needed to confirm these findings. Furthermore, APOE ε4 has been reported to interact with tau PET to influence memory independently of amyloid PET in older adults. 32 In this study, the associations between tau pathology and PACC components and OCL RT were APOE ε4 independent.
A recent study exploring the associations between hippocampal volume and PACC in the A4 study reported that significant associations were also present only in women. 21 In this study, after sex stratification, the significant associations between tau SUVR in the meta‐temporal region and cognitive outcomes were also found only in women at the preclinical AD stage. Women and men in CN and MCI have been reported to show different tau topographic distributions. 3 , 18 It would be interesting to explore if tau pathology or brain atrophy in other regions, or other pathologies such as white matter hyperintensity, are associated with worse cognitive performance in men at the preclinical AD stage in the future. Furthermore, in this study, although most of the results for C3 components at Visit 1 and Visit 3 are similar, there are a few discrepancies for the C3 at two visits, which might be due to the increased familiarity with the tablet and task demands at re‐testing. 22
This study has some other limitations. A4 Study is from multi‐site cohorts, which might introduce variability to neuroimaging measures. Imaging data harmonization and site‐effect adjustment should be considered in future analyses. In addition, we included amyloid‐positive CN participants, who were not perfectly representative of preclinical AD, since some amyloid‐positive individuals could remain tau negative status for at least 5 years and would not develop AD. 33 Furthermore, the participants of this study were restricted to non‐Hispanic White, which does not generalize to the overall population. Future studies should test whether these findings apply to other populations. Furthermore, tau PET data in this study were not corrected for partial volume effects, which might decrease the sensitivity of association analysis.
5. CONCLUSION
Taken together, our results suggest that there are sex differences in the associations between tau pathology and cognition in early AD. Women with higher amounts of tau in the meta‐temporal region showed poorer cognitive performance; however, these relationships were not observed in men. These findings have important implications for sex‐specific tau‐targeted preventive AD clinical trials or trials with tau or cognition as outcomes.
CONFLICT OF INTEREST STATEMENT
Xin Wang has no relevant disclosures. Dr. Sundermann has no relevant disclosures. She discloses the research funding outside this paper from the National Institutes of Health/National Institute on Aging (NIH/NIA; R01AG074221). Dr. Buckley has no relevant disclosures. She is supported by the NIH/NIA (R00AG061238, DP2AG082342) and the Alzheimer's Association (AARF‐20‐675646). Dr. Banks has no relevant disclosures. Author disclosures are available in the Supporting Information.
CONSENT STATEMENT
All participants in this study signed informed consent before participating in this study.
Supporting information
Supporting Information
Supporting Information
ACKNOWLEDGMENTS
The A4 Study is a secondary prevention trial in preclinical Alzheimer's disease, aiming to slow cognitive decline associated with brain amyloid accumulation in clinically normal (CN) older individuals. The A4 Study is funded by a public‐private‐philanthropic partnership, including funding from the National Institutes of Health/National Institute on Aging, Eli Lilly and Company, Alzheimer's Association, Accelerating Medicines Partnership, GHR Foundation, an anonymous foundation and additional private donors, with in‐kind support from Avid and Cogstate. The companion observational Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) Study is funded by the Alzheimer's Association and GHR Foundation. The A4 and LEARN Studies are led by Dr. Reisa Sperling at Brigham and Women's Hospital, Harvard Medical School and Dr. Paul Aisen at the Alzheimer's Therapeutic Research Institute (ATRI), University of Southern California. The A4 and LEARN Studies are coordinated by ATRI at the University of Southern California, and the data are made available through the Laboratory for NeuroImaging at the University of Southern California. The participants screening for the A4 Study provided permission to share their de‐identified data in order to advance the quest to find a successful treatment for Alzheimer's disease. We would like to acknowledge the dedication of all the participants, the site personnel, and all of the partnership team members, who continue to make the A4 and LEARN Studies possible. The complete A4 Study Team list is available on: a4study.org/a4‐study‐team. This work was supported by the National Institutes of Health (grant numbers 1R01AG066088).
STUDY TEAM OF A4
1.
Wang X, Sundermann EE, Buckley RF, Banks SJ. Sex differences in the association between tau PET and cognitive performance in a non‐Hispanic White cohort with preclinical AD. Alzheimer's Dement. 2024;20:25–33. 10.1002/alz.13432
Full listing of A4 Study team and site personnel available at A4STUDY.org (also see Appendix).
REFERENCES
- 1. Nebel RA, Aggarwal NT, Barnes LL, et al. Understanding the impact of sex and gender in Alzheimer's disease: a call to action. Alzheimers Dement. 2018;14:1171‐1183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sundermann EE, Maki PM, Rubin LH, et al. Female advantage in verbal memory: evidence of sex‐specific cognitive reserve. Neurology. 2016;87:1916‐1924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Banks SJ, Andrews MJ, Digma L, et al. Sex differences in Alzheimer's disease: do differences in tau explain the verbal memory gap? Neurobiol Aging. 2021;107:70‐77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lin KA, Choudhury KR, Rathakrishnan BG, et al. Marked gender differences in progression of mild cognitive impairment over 8 years. Alzheimers Dement (N Y). 2015;1:103‐110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Laws KR, Irvine K, Gale TM. Sex differences in cognitive impairment in Alzheimer's disease. World J Psychiatry. 2016;6:54‐65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. McCarrey AC, An Y, Kitner‐Triolo MH, Ferrucci L, Resnick SM. Sex differences in cognitive trajectories in clinically normal older adults. Psychol Aging. 2016;31:166‐175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Beinhoff U, Tumani H, Brettschneider J, Bittner D, Riepe MW. Gender‐specificities in Alzheimer's disease and mild cognitive impairment. J Neurol. 2008;255:117‐122. [DOI] [PubMed] [Google Scholar]
- 8. Ossenkoppele R, Smith R, Mattsson‐Carlgren N, et al. Accuracy of Tau positron emission tomography as a prognostic marker in preclinical and prodromal Alzheimer disease: a head‐to‐head comparison against amyloid positron emission tomography and magnetic resonance imaging. JAMA Neurology. 2021;78:961‐971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Biel D, Brendel M, Rubinski A, et al. Tau‐PET and in vivo Braak‐staging as prognostic markers of future cognitive decline in cognitively normal to demented individuals. Alzheimers Res Ther. 2021;13:137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Teng E, Manser PT, Sanabria Bohorquez S, et al. 18)F]GTP1 tau PET imaging is associated with subsequent cognitive decline in Alzheimer's disease. Alzheimers Res Ther. 2021;13:196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Digma LA, Madsen JR, Reas ET, et al. Tau and atrophy: domain‐specific relationships with cognition. Alzheimers Res Ther. 2019;11:65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Scholl M, Lockhart SN, Schonhaut DR, et al. PET imaging of tau deposition in the aging human brain. Neuron. 2016;89:971‐982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Maass A, Lockhart SN, Harrison TM, et al. Entorhinal tau pathology, episodic memory decline, and neurodegeneration in aging. J Neurosci. 2018;38:530‐543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Lowe VJ, Bruinsma TJ, Wiste HJ, et al. Cross‐sectional associations of tau‐PET signal with cognition in cognitively unimpaired adults. Neurology. 2019;93:e29‐e39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Buckley RF, O'Donnell A, McGrath ER, et al. Menopause status moderates sex differences in tau burden: a framingham PET study. Ann Neurol. 2022;92:11‐22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Wisch JK, Meeker KL, Gordon BA, et al. Sex‐related differences in tau positron emission tomography (PET) and the effects of hormone therapy (HT). Alzheimer Dis Assoc Disord. 2021;35:164‐168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Buckley RF, Mormino EC, Rabin JS, et al. Sex differences in the association of global amyloid and regional tau deposition measured by positron emission tomography in clinically normal older adults. JAMA Neurol. 2019;76:542‐551. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Buckley RF, Scott MR, Jacobs HIL, et al. Sex mediates relationships between regional tau pathology and cognitive decline. Ann Neurol. 2020;88:921‐932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Insel PS, Donohue MC, Sperling R, Hansson O, Mattsson‐Carlgren N. The A4 study: beta‐amyloid and cognition in 4432 cognitively unimpaired adults. Ann Clin Transl Neurol. 2020;7:776‐785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Ohman F, Hassenstab J, Berron D, Scholl M, Papp KV. Current advances in digital cognitive assessment for preclinical Alzheimer's disease. Alzheimers Dement (Amst). 2021;13:e12217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Petersen KK, Grober E, Lipton RB, et al. Impact of sex and APOE epsilon4 on the association of cognition and hippocampal volume in clinically normal, amyloid positive adults. Alzheimers Dement (Amst). 2022;14:e12271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Papp KV, Rentz DM, Maruff P, et al. The computerized cognitive composite (C3) in an Alzheimer's disease secondary prevention trial. J Prev Alzheimers Dis. 2021;8:59‐67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Avila JF, Renteria MA, Jones RN, et al. Education differentially contributes to cognitive reserve across racial/ethnic groups. Alzheimers Dement. 2021;17:70‐80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Zahodne LB, Manly JJ, Azar M, Brickman AM, Glymour MM. Racial disparities in cognitive performance in mid‐ and late adulthood: analyses of two cohort studies. J Am Geriatr Soc. 2016;64:959‐964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Sperling RA, Donohue MC, Raman R, et al. Association of factors with elevated amyloid burden in clinically normal older individuals. JAMA Neurol. 2020;77:735‐745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Donohue MC, Sperling RA, Salmon DP, et al. The preclinical Alzheimer cognitive composite: measuring amyloid‐related decline. JAMA Neurol. 2014;71:961‐970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Amariglio RE, Frishe K, Olson LE, et al. Validation of the face name associative memory exam in cognitively normal older individuals. J Clin Exp Neuropsychol. 2012;34:580‐587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Stark SM, Yassa MA, Lacy JW, Stark CE. A task to assess behavioral pattern separation (BPS) in humans: data from healthy aging and mild cognitive impairment. Neuropsychologia. 2013;51:2442‐2449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Fukunaga T, Ueno Y, Imabayashi T, Mizoi Y, Adachi J, Nakagawa K. Effect of aldehyde dehydrogenase deficiency on ethanol elimination after peroral or intravenous administrations. Acta Med Leg Soc (Liege). 1989;39:499‐503. [PubMed] [Google Scholar]
- 30. Brier MR, Gordon B, Friedrichsen K, et al. Tau and Abeta imaging, CSF measures, and cognition in Alzheimer's disease. Sci Transl Med. 2016;8:338ra66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kertzman S, Ben‐Nahum Z, Gotzlav I, Grinspan H, Birger M, Kotler M. Digit symbol substitution test performance: sex differences in a Hebrew‐readers' health population. Percept Mot Skills. 2006;103:121‐130. [DOI] [PubMed] [Google Scholar]
- 32. Weigand AJ, Thomas KR, Bangen KJ, et al. APOE interacts with tau PET to influence memory independently of amyloid PET in older adults without dementia. Alzheimers Dement. 2021;17:61‐69. [DOI] [PubMed] [Google Scholar]
- 33. Josephs KA, Weigand SD, Whitwell JL. Characterizing amyloid‐positive individuals with normal tau PET levels after 5 years: an ADNI study. Neurology. 2022;98:e2282. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Supporting Information