Genetic effects on longitudinal cognitive decline during the early stages of Alzheimer's disease

Disentangling genetic risks for development and progression of Alzheimer's disease

This scientific commentary refers to ‘Towards cascading genetic risk in Alzheimer’s disease’ by Altmann et al. (https://doi.org/10.1093/brain/awae176).

Apolipoprotein E in Alzheimer's disease trajectories and the next-generation clinical care pathway

Alzheimer's disease (AD) is a complex, progressive primary neurodegenerative disease. Since pivotal genetic studies in 1993, the ε4 allele of the apolipoprotein E gene (APOE ε4) has remained the strongest single genome-wide associated risk variant in AD. Scientific advances in APOE biology, AD pathophysiology and ApoE-targeted therapies have brought APOE to the forefront of research, with potential translation into routine AD clinical care. This contemporary Review will merge APOE research with the emerging AD clinical care pathway and discuss APOE genetic risk as a conduit to genomic-based precision medicine in AD, including ApoE's influence in the ATX(N) biomarker framework of AD. We summarize the evidence for APOE as an important modifier of AD clinical-biological trajectories. We then illustrate the utility of APOE testing and the future of ApoE-targeted therapies in the next-generation AD clinical-diagnostic pathway. With the emergence of new AD therapies, understanding how APOE modulates AD pathophysiology will become critical for personalized AD patient care.

Polygenic burden and its association with baseline cognitive function and postoperative cognitive outcome in temporal lobe epilepsy

OBJECTIVE

Demographic and disease factors are associated with cognitive deficits and postoperative cognitive declines in adults with pharmacoresistant temporal lobe epilepsy (TLE), but the role of genetic factors in cognition in TLE is not well understood. Polygenic scores (PGS) for neurological and neuropsychiatric disorders and IQ have been associated with cognition in patient and healthy populations. In this exploratory study, we examined the relationship between PGS for Alzheimer's disease (AD), depression, and IQ and cognitive outcomes in adults with TLE.

METHODS

202 adults with pharmacoresistant TLE had genotyping and completed neuropsychological evaluations as part of a presurgical work-up. A subset (n = 116) underwent temporal lobe resection and returned for postoperative cognitive testing. Logistic regression was used to determine if PGS for AD, depression, and IQ predicted baseline domain-specific cognitive function and cognitive phenotypes as well as postoperative language and memory decline.

RESULTS

No significant findings survived correction for multiple comparisons. Prior to correction, higher PGS for AD and depression (i.e., increased genetic risk for the disorder), but lower PGS for IQ (i.e., decreased genetic likelihood of high IQ) appeared possibly associated with baseline cognitive impairment in TLE. In comparison, higher PGS for AD and IQ appeared as possible risk factors for cognitive decline following temporal lobectomy, while the possible relationship between PGS for depression and post-operative cognitive outcome was mixed.

SIGNIFICANCE

We did not observe any relationships of large effect between PGS and cognitive function or postsurgical outcome; however, results highlight several promising trends in the data that warrant future investigation in larger samples better powered to detect small genetic effects.

Association of retinal optical coherence tomography metrics and polygenic risk scores with cognitive function and future cognitive decline

PURPOSE

To evaluate the potential of retinal optical coherence tomography (OCT) measurements and polygenic risk scores (PRS) to identify people at risk of cognitive impairment.

METHODS

Using OCT images from 50 342 UK Biobank participants, we examined associations between retinal layer thickness and genetic risk for neurodegenerative disease and combined these metrics with PRS to predict baseline cognitive function and future cognitive deterioration. Multivariate Cox proportional hazard models were used to predict cognitive performance. P values for retinal thickness analyses are false-discovery-rate-adjusted.

RESULTS

Higher Alzheimer's disease PRS was associated with a thicker inner nuclear layer (INL), chorio-scleral interface (CSI) and inner plexiform layer (IPL) (all p<0.05). Higher Parkinson's disease PRS was associated with thinner outer plexiform layer (p<0.001). Worse baseline cognitive performance was associated with thinner retinal nerve fibre layer (RNFL) (aOR=1.038, 95% CI (1.029 to 1.047), p<0.001) and photoreceptor (PR) segment (aOR=1.035, 95% CI (1.019 to 1.051), p<0.001), ganglion cell complex (aOR=1.007, 95% CI (1.002 to 1.013), p=0.004) and thicker ganglion cell layer (aOR=0.981, 95% CI (0.967 to 0.995), p=0.009), IPL (aOR=0.976, 95% CI (0.961 to 0.992), p=0.003), INL (aOR=0.923, 95% CI (0.905 to 0.941), p<0.001) and CSI (aOR=0.998, 95% CI (0.997 to 0.999), p<0.001). Worse future cognitive performance was associated with thicker IPL (aOR=0.945, 95% CI (0.915 to 0.999), p=0.045) and CSI (aOR=0.996, 95% CI (0.993 to 0.999) 95% CI, p=0.014). Prediction of cognitive decline was significantly improved with the addition of PRS and retinal measurements.

CONCLUSIONS AND RELEVANCE

Retinal OCT measurements are significantly associated with genetic risk of neurodegenerative disease and may serve as biomarkers predictive of future cognitive impairment.

Comparing a pre-defined versus deep learning approach for extracting brain atrophy patterns to predict cognitive decline due to Alzheimer's disease in patients with mild cognitive symptoms

BACKGROUND

Predicting future Alzheimer's disease (AD)-related cognitive decline among individuals with subjective cognitive decline (SCD) or mild cognitive impairment (MCI) is an important task for healthcare. Structural brain imaging as measured by magnetic resonance imaging (MRI) could potentially contribute when making such predictions. It is unclear if the predictive performance of MRI can be improved using entire brain images in deep learning (DL) models compared to using pre-defined brain regions.

METHODS

A cohort of 332 individuals with SCD/MCI were included from the Swedish BioFINDER-1 study. The goal was to predict longitudinal SCD/MCI-to-AD dementia progression and change in Mini-Mental State Examination (MMSE) over four years. Four models were evaluated using different predictors: (1) clinical data only, including demographics, cognitive tests and APOE ε4 status, (2) clinical data plus hippocampal volume, (3) clinical data plus all regional MRI gray matter volumes (N = 68) extracted using FreeSurfer software, (4) a DL model trained using multi-task learning with MRI images, Jacobian determinant images and baseline cognition as input. A double cross-validation scheme, with five test folds and for each of those ten validation folds, was used. External evaluation was performed on part of the ADNI dataset, including 108 patients. Mann-Whitney U-test was used to determine statistically significant differences in performance, with p-values less than 0.05 considered significant.

RESULTS

In the BioFINDER cohort, 109 patients (33%) progressed to AD dementia. The performance of the clinical data model for prediction of progression to AD dementia was area under the curve (AUC) = 0.85 and four-year cognitive decline was R2 = 0.14. The performance was improved for both outcomes when adding hippocampal volume (AUC = 0.86, R2 = 0.16). Adding FreeSurfer brain regions improved prediction of four-year cognitive decline but not progression to AD (AUC = 0.83, R2 = 0.17), while the DL model worsened the performance for both outcomes (AUC = 0.84, R2 = 0.08). A sensitivity analysis showed that the Jacobian determinant image was more informative than the MRI image, but that performance was maximized when both were included. In the external evaluation cohort from ADNI, 23 patients (21%) progressed to AD dementia. The results for predicted progression to AD dementia were similar to the results for the BioFINDER test data, while the performance for the cognitive decline was deteriorated.

CONCLUSIONS

The DL model did not significantly improve the prediction of clinical disease progression in AD, compared to regression models with a single pre-defined brain region.

Comparing a pre-defined versus deep learning approach for extracting brain atrophy patterns to predict cognitive decline due to Alzheimer's disease in patients with mild cognitive symptoms

Background

Predicting future Alzheimer's disease (AD)-related cognitive decline among individuals with subjective cognitive decline (SCD) or mild cognitive impairment (MCI) is an important task for healthcare. Structural brain imaging as measured by magnetic resonance imaging (MRI) could potentially contribute when making such predictions. It is unclear if the predictive performance of MRI can be improved using entire brain images in deep learning (DL) models compared to using pre-defined brain regions.

Methods

A cohort of 332 individuals with SCD/MCI were included from the Swedish BioFINDER-1 study. The goal was to predict longitudinal SCD/MCI-to-AD dementia progression and change in Mini-Mental State Examination (MMSE) over four years. Four models were evaluated using different predictors: 1) clinical data only, including demographics, cognitive tests and APOE e4 status, 2) clinical data plus hippocampal volume, 3) clinical data plus all regional MRI gray matter volumes (N=68) extracted using FreeSurfer software, 4) a DL model trained using multi-task learning with MRI images, Jacobian determinant images and baseline cognition as input. Models were developed on 80% of subjects (N=267) and tested on the remaining 20% (N=65). Mann-Whitney U-test was used to determine statistically significant differences in performance, with p-values less than 0.05 considered significant.

Results

In the test set, 21 patients (32.3%) progressed to AD dementia. The performance of the clinical data model for prediction of progression to AD dementia was area under the curve (AUC)=0.87 and four-year cognitive decline was R2=0.17. The performance was significantly improved for both outcomes when adding hippocampal volume (AUC=0.91, R2=0.26, p-values <0.05) or FreeSurfer brain regions (AUC=0.90, R2=0.27, p-values <0.05). Conversely, the DL model did not show any significant difference from the clinical data model (AUC=0.86, R2=0.13). A sensitivity analysis showed that the Jacobian determinant image was more informative than the MRI image, but that performance was maximized when both were included.

Conclusions

The DL model did not significantly improve the prediction of clinical disease progression in AD, compared to regression models with a single pre-defined brain region.

Cognitive trajectories diverge by genetic risk in a preclinical longitudinal cohort

INTRODUCTION

We sought to characterize the timing of changes in cognitive trajectories related to genetic risk using the apolipoprotein E (APOE) score, a continuous measure of Alzheimer's disease (AD) risk. We also aimed to determine whether that timing was different when genetic risk was measured using an AD polygenic risk score (PRS) that contains APOE.

METHODS

We analyzed trajectories (N ≈1135) for four neuropsychological composite scores using mixed effects regression for longitudinal change across APOE scores and PRS of participants in the Wisconsin Registry for Alzheimer's Prevention, a longitudinal study of adults aged 40 to 70 at baseline, with a median participant follow-up time of 7.8 years.

RESULTS

We found a significant non-linear age-by-APOE score interaction in predicting cognitive decline. Cognitive trajectories diverged by APOE score at approximately 65 years of age. A 0.5 standard deviation difference in cognition between extreme percentiles of the PRS was predicted to occur 1 to 2 years before that of the APOE score.

DISCUSSION

Cognitive decline differs across time and APOE score. Estimates did not substantially shift with the AD PRS.

HIGHLIGHTS

The apolipoprotein E (APOE) score, a continuous measure, accounts for non-linear genetic risk of Alzheimer's disease. Non-linear age interacts with the APOE score to affect cognition. Cognitive decline starts to differ by APOE score levels at approximately age 65. Cognitive decline timing by polygenic risk (including APOE) is similar to APOE alone.

Alzheimer's disease genetic risk and cognitive reserve in relationship to long-term cognitive trajectories among cognitively normal individuals

BACKGROUND

Both Alzheimer's disease (AD) genetic risk factors and indices of cognitive reserve (CR) influence risk of cognitive decline, but it remains unclear whether they interact. This study examined whether a CR index score modifies the relationship between AD genetic risk factors and long-term cognitive trajectories in a large sample of individuals with normal cognition.

METHODS

Analyses used data from the Preclinical AD Consortium, including harmonized data from 5 longitudinal cohort studies. Participants were cognitively normal at baseline (M baseline age = 64 years, 59% female) and underwent 10 years of follow-up, on average. AD genetic risk was measured by (i) apolipoprotein-E (APOE) genetic status (APOE-ε2 and APOE-ε4 vs. APOE-ε3; N = 1819) and (ii) AD polygenic risk scores (AD-PRS; N = 1175). A CR index was calculated by combining years of education and literacy scores. Longitudinal cognitive performance was measured by harmonized factor scores for global cognition, episodic memory, and executive function.

RESULTS

In mixed-effects models, higher CR index scores were associated with better baseline cognitive performance for all cognitive outcomes. APOE-ε4 genotype and AD-PRS that included the APOE region (AD-PRSAPOE) were associated with declines in all cognitive domains, whereas AD-PRS that excluded the APOE region (AD-PRSw/oAPOE) was associated with declines in executive function and global cognition, but not memory. There were significant 3-way CR index score × APOE-ε4 × time interactions for the global (p = 0.04, effect size = 0.16) and memory scores (p = 0.01, effect size = 0.22), indicating the negative effect of APOE-ε4 genotype on global and episodic memory score change was attenuated among individuals with higher CR index scores. In contrast, levels of CR did not attenuate APOE-ε4-related declines in executive function or declines associated with higher AD-PRS. APOE-ε2 genotype was unrelated to cognition.

CONCLUSIONS

These results suggest that APOE-ε4 and non-APOE-ε4 AD polygenic risk are independently associated with global cognitive and executive function declines among individuals with normal cognition at baseline, but only APOE-ε4 is associated with declines in episodic memory. Importantly, higher levels of CR may mitigate APOE-ε4-related declines in some cognitive domains. Future research is needed to address study limitations, including generalizability due to cohort demographic characteristics.