Mild cognitive impairment (MCI) patients are at risk of cognitive decline, while elevated serum homocysteine is also associated with cognitive impairment. Thus, older people with MCI and hyperhomocysteinemia may be under greater risk of cognitive decline. We therefore performed a randomized trial of homocysteine-lowering by B vitamins supplementation to prevent cognitive decline in older MCI patients with elevated serum homocysteine. 279 MCI outpatients aged ≥65 years with serum homocysteine ≥10.0 μmol/L were randomly assigned to take either methylcobalamin 500 μg and folic acid 400 μg once daily, or two placebo tablets for 24 months. All subjects were followed up at 12 monthly intervals. The primary outcome was cognitive decline as defined by an increase in clinical dementia rating scale (CDR) sum of boxes (CDR_SOB). The secondary outcomes were global CDR, memory Z score, executive function Z score and Hamilton depression rating scale (HDRS) score. The clinical characteristics between two groups were well matched, except that the supplement group had better executive function. The supplement effectively lowered serum homocysteine (mean 13.9 ± sd 3.5 μmol at baseline to 9.3 ± 2.4 μmol/L at month 24). At month 24, there was no significant group difference in CDR_SOB or any secondary outcomes (mean changes in CDR_SOB 0.36 versus 0.22 in supplement and placebo groups respectively). At month 12, the supplement group significantly improved in executive function and had lower HDRS score (P = 0.004 and 0.012 respectively). Group difference was significant for HDRS, but borderline significant for executive function. (P = 0.01; 0.06 respectively) These effects were not significant at month 24. Subgroup analysis showed that aspirin use had significant interaction with B supplements in CDR_SOB at month 24 (Beta 0.189, P = 0.005). Vitamin B12 and folic acid supplementation did not reduce cognitive decline in older people with MCI and elevated serum homocysteine, though the cognitive decline over two years in placebo group was small. The supplement led to a significant reduction in depressive symptoms at month 12, though this effect was not sustained. Aspirin use had a negative interaction effect on cognitive functioning with B supplements. Centre for Clinical Research and Biostatistics (CCRB) Clinical Trials Registry: CUHK_CCT00373.

Background: The polygenic nature of Alzheimer’s disease (AD) suggests that multiple variants jointly contribute to disease susceptibility. As an individual’s genetic variants are constant throughout life, evaluating the combined effects of multiple disease-associated genetic risks enables reliable AD risk prediction. Because of the complexity of genomic data, current statistical analyses cannot comprehensively capture the polygenic risk of AD, resulting in unsatisfactory disease risk prediction. However, deep learning methods, which capture nonlinearity within high-dimensional genomic data, may enable more accurate disease risk prediction and improve our understanding of AD etiology. Accordingly, we developed deep learning neural network models for modeling AD polygenic risk. Methods: We constructed neural network models to model AD polygenic risk and compared them with the widely used weighted polygenic risk score and lasso models. We conducted robust linear regression analysis to investigate the relationship between the AD polygenic risk derived from deep learning methods and AD endophenotypes (i.e., plasma biomarkers and individual cognitive performance). We stratified individuals by applying unsupervised clustering to the outputs from the hidden layers of the neural network model. Results: The deep learning models outperform other statistical models for modeling AD risk. Moreover, the polygenic risk derived from the deep learning models enables the identification of disease-associated biological pathways and the stratification of individuals according to distinct pathological mechanisms. Conclusion: Our results suggest that deep learning methods are effective for modeling the genetic risks of AD and other diseases, classifying disease risks, and uncovering disease mechanisms.