Genetic and Clinical Correlates of AI-Based Brain Aging Patterns in Cognitively Unimpaired Individuals

Importance

Brain aging elicits complex neuroanatomical changes influenced by multiple age-related pathologies. Understanding the heterogeneity of structural brain changes in aging may provide insights into preclinical stages of neurodegenerative diseases.

Objective

To derive subgroups with common patterns of variation in participants without diagnosed cognitive impairment (WODCI) in a data-driven manner and relate them to genetics, biomedical measures, and cognitive decline trajectories.

Design, Setting, and Participants

Data acquisition for this cohort study was performed from 1999 to 2020. Data consolidation and harmonization were conducted from July 2017 to July 2021. Age-specific subgroups of structural brain measures were modeled in 4 decade-long intervals spanning ages 45 to 85 years using a deep learning, semisupervised clustering method leveraging generative adversarial networks. Data were analyzed from July 2021 to February 2023 and were drawn from the Imaging-Based Coordinate System for Aging and Neurodegenerative Diseases (iSTAGING) international consortium. Individuals WODCI at baseline spanning ages 45 to 85 years were included, with greater than 50 000 data time points.

Exposures

Individuals WODCI at baseline scan.

Main Outcomes and Measures

Three subgroups, consistent across decades, were identified within the WODCI population. Associations with genetics, cardiovascular risk factors (CVRFs), amyloid β (Aβ), and future cognitive decline were assessed.

Results

In a sample of 27 402 individuals (mean [SD] age, 63.0 [8.3] years; 15 146 female [55%]) WODCI, 3 subgroups were identified in contrast with the reference group: a typical aging subgroup, A1, with a specific pattern of modest atrophy and white matter hyperintensity (WMH) load, and 2 accelerated aging subgroups, A2 and A3, with characteristics that were more distinct at age 65 years and older. A2 was associated with hypertension, WMH, and vascular disease-related genetic variants and was enriched for Aβ positivity (ages ≥65 years) and apolipoprotein E (APOE) ε4 carriers. A3 showed severe, widespread atrophy, moderate presence of CVRFs, and greater cognitive decline. Genetic variants associated with A1 were protective for WMH (rs7209235: mean [SD] B = -0.07 [0.01]; P value = 2.31 × 10-9) and Alzheimer disease (rs72932727: mean [SD] B = 0.1 [0.02]; P value = 6.49 × 10-9), whereas the converse was observed for A2 (rs7209235: mean [SD] B = 0.1 [0.01]; P value = 1.73 × 10-15 and rs72932727: mean [SD] B = -0.09 [0.02]; P value = 4.05 × 10-7, respectively); variants in A3 were associated with regional atrophy (rs167684: mean [SD] B = 0.08 [0.01]; P value = 7.22 × 10-12) and white matter integrity measures (rs1636250: mean [SD] B = 0.06 [0.01]; P value = 4.90 × 10-7).

Conclusions and Relevance

The 3 subgroups showed distinct associations with CVRFs, genetics, and subsequent cognitive decline. These subgroups likely reflect multiple underlying neuropathologic processes and affect susceptibility to Alzheimer disease, paving pathways toward patient stratification at early asymptomatic stages and promoting precision medicine in clinical trials and health care.

Gene-SGAN: discovering disease subtypes with imaging and genetic signatures via multi-view weakly-supervised deep clustering

Disease heterogeneity has been a critical challenge for precision diagnosis and treatment, especially in neurologic and neuropsychiatric diseases. Many diseases can display multiple distinct brain phenotypes across individuals, potentially reflecting disease subtypes that can be captured using MRI and machine learning methods. However, biological interpretability and treatment relevance are limited if the derived subtypes are not associated with genetic drivers or susceptibility factors. Herein, we describe Gene-SGAN - a multi-view, weakly-supervised deep clustering method - which dissects disease heterogeneity by jointly considering phenotypic and genetic data, thereby conferring genetic correlations to the disease subtypes and associated endophenotypic signatures. We first validate the generalizability, interpretability, and robustness of Gene-SGAN in semi-synthetic experiments. We then demonstrate its application to real multi-site datasets from 28,858 individuals, deriving subtypes of Alzheimer's disease and brain endophenotypes associated with hypertension, from MRI and single nucleotide polymorphism data. Derived brain phenotypes displayed significant differences in neuroanatomical patterns, genetic determinants, biological and clinical biomarkers, indicating potentially distinct underlying neuropathologic processes, genetic drivers, and susceptibility factors. Overall, Gene-SGAN is broadly applicable to disease subtyping and endophenotype discovery, and is herein tested on disease-related, genetically-associated neuroimaging phenotypes.

Five dominant dimensions of brain aging are identified via deep learning: associations with clinical, lifestyle, and genetic measures

Brain aging is a complex process influenced by various lifestyle, environmental, and genetic factors, as well as by age-related and often co-existing pathologies. MRI and, more recently, AI methods have been instrumental in understanding the neuroanatomical changes that occur during aging in large and diverse populations. However, the multiplicity and mutual overlap of both pathologic processes and affected brain regions make it difficult to precisely characterize the underlying neurodegenerative profile of an individual from an MRI scan. Herein, we leverage a state-of-the art deep representation learning method, Surreal-GAN, and present both methodological advances and extensive experimental results that allow us to elucidate the heterogeneity of brain aging in a large and diverse cohort of 49,482 individuals from 11 studies. Five dominant patterns of neurodegeneration were identified and quantified for each individual by their respective (herein referred to as) R-indices. Significant associations between R-indices and distinct biomedical, lifestyle, and genetic factors provide insights into the etiology of observed variances. Furthermore, baseline R-indices showed predictive value for disease progression and mortality. These five R-indices contribute to MRI-based precision diagnostics, prognostication, and may inform stratification into clinical trials.

Gene-SGAN: a method for discovering disease subtypes with imaging and genetic signatures via multi-view weakly-supervised deep clustering

Disease heterogeneity has been a critical challenge for precision diagnosis and treatment, especially in neurologic and neuropsychiatric diseases. Many diseases can display multiple distinct brain phenotypes across individuals, potentially reflecting disease subtypes that can be captured using MRI and machine learning methods. However, biological interpretability and treatment relevance are limited if the derived subtypes are not associated with genetic drivers or susceptibility factors. Herein, we describe Gene-SGAN - a multi-view, weakly-supervised deep clustering method - which dissects disease heterogeneity by jointly considering phenotypic and genetic data, thereby conferring genetic correlations to the disease subtypes and associated endophenotypic signatures. We first validate the generalizability, interpretability, and robustness of Gene-SGAN in semi-synthetic experiments. We then demonstrate its application to real multi-site datasets from 28,858 individuals, deriving subtypes of Alzheimer's disease and brain endophenotypes associated with hypertension, from MRI and SNP data. Derived brain phenotypes displayed significant differences in neuroanatomical patterns, genetic determinants, biological and clinical biomarkers, indicating potentially distinct underlying neuropathologic processes, genetic drivers, and susceptibility factors. Overall, Gene-SGAN is broadly applicable to disease subtyping and endophenotype discovery, and is herein tested on disease-related, genetically-driven neuroimaging phenotypes.

Quantitative analysis of Op18 phosphorylation in childhood acute leukemia

Oncoprotein 18 (Op18) is a major cystolic phosphoprotein constituent of leukemia cells. There is cumulative evidence that suggests a role for Op18 in integrating signals from diverse pathways involved in cell growth. Op18 phosphorylation is induced with proliferation in a variety of cell types, and is essential for cell cycle progression. In this study we analyzed the level of unphosphorylated Op18 and of its major phosphorylated forms, Op18a and Op18b, in a series of 177 childhood acute leukemias by means of quantitative two-dimensional polyacrylamide gel electrophoresis (2-D PAGE). Op18 phosphorylation was significantly correlated with the white blood count at the time of diganosis, and with a high percentage of cells in the S phase. Our findings suggest that strategies to inhibit Op18 expression or phosphorylation may be effective in inhibiting leukemia cell proliferation.

Proliferation-related expression of p19/nm23 nucleoside diphosphate kinase

High level expression of the nm23-H1 gene, which encodes for a nucleoside diphosphate kinase, has been found to correlate with diminished metastasis in some tumors but not in others. We have previously identified the protein product of the nm23-H1 gene in two-dimensional electrophoretic gels and have designated it p19/nm23. In neuroblastoma, higher levels of p19/nm23, which are associated with amplification of the N-myc oncogene, large tumor mass, and metastasis, were observed in advanced stage tumors compared with limited stage disease. Because of the variable expression of nm23-H1 in different tumors, we have investigated the relationship between amounts of the protein and cell proliferation. The levels of p19/nm23 were compared between resting and mitotically stimulated normal human PBLs and in leukemia cells. The amount of p19/nm23 increased in normal lymphocytes in response to mitotic stimulation and paralleled the increase in DNA synthesis. In leukemia cells obtained from patients with different subtypes of acute leukemia, p19/nm23 levels were also increased relative to resting normal lymphocytes. Treatment of mitotically stimulated lymphocytes with cyclosporin, which inhibits proliferation, blocked the increase in p19/nm23; treatment of the leukemia cell line HL-60 with dimethylsulfoxide, which induces terminal differentiation, resulted in diminished levels of p19/nm23. Our data therefore provide evidence that nm23-H1 expression is related to cell proliferative activity.

Highly resolving two-dimensional gels for protein sequencing

Two-dimensional (2D) PAGE, using carrier ampholytes for the first-dimension separation, has provided a tool for the simultaneous analysis of cellular proteins. To extend the utility of 2D PAGE to the preparative level, we have investigated the use of immobilized pH gradients (IPG) for the first-dimension separation. The results we have obtained indicate that as much as 1 mg of cellular protein can be loaded onto a single IPG gel without loss of resolution. Mutant polypeptides previously detected in carrier ampholyte-based 2D gels were equally detectable in IPG-based 2D gels. With IPG gels several hundred cellular polypeptides can be isolated, from as few as 10 gels, in sufficient amount for sequencing with current sequencing technology. We therefore conclude that IPG greatly enhances the prospects for the large-scale sequencing of cellular proteins for the development of 2D gel-related protein data bases and for the identification of new polypeptide gene products, with the attendant implications for a genome sequencing effort.

N-myc gene amplification in neuroblastoma is associated with altered phosphorylation of a proliferation related polypeptide (Op18)

We have recently identified and cloned the gene for a cytosolic polypeptide, designated oncoprotein 18 (Op18), which is expressed in acute lymphocytic leukemia and some solid tumors including neuroblastoma. Op18 is phosphorylated upon treatment of lymphoid cells with phorbol myristate acetate. We have proposed that unphosphorylated Op18 plays a role in cellular proliferation, and that its phosphorylated forms, namely Op18a and Op18b, are associated with diminished cell proliferation. In this study, we report that in neuroblastoma tumors, the phosphorylation of Op18 was substantially diminished with increasing N-myc gene copy number. Treatment of the neuroblastoma cell line SMS-KCNR, which contains 75 copies of the N-myc gene, with retinoic acid for ten days resulted in an increase in Op18 phosphorylation. Our findings provide evidence for distinct patterns of Op18 phosphorylation in neuroblastoma tumors with and without N-myc gene amplification.

Molecular cloning of a novel human leukemia-associated gene. Evidence of conservation in animal species

We have recently described an 18-kilodalton polypeptide (p18) that is present in much greater abundance in acute leukemic blast cells (myeloid and lymphoid) than in resting or proliferating nonleukemic lymphoid cells or chronic lymphoid and myeloid leukemic cells. In this report we describe the cloning of two different sized full-length cDNAs that code for p18. The two cDNAs differ in their 3'-noncoding regions as a result of alternative polyadenylation. Analysis of the complete nucleotide sequence and the corresponding amino acid sequence did not reveal significant homology to any previously described sequences. We show evidence that this gene is highly conserved in several animal species and low stringency hybridization studies suggest that the p18 gene may be a member of a family of partially homologous genes in the human genome.

Unilateral pulmonary edema as a complication of contralateral bronchial obstruction

A unique case of acute unilateral bronchial obstruction complicated by contralateral pulmonary edema is reported. Postulated pathogenetic mechanisms are hypoxia and hyperfusion of one lung with resultant pulmonary capillary endothelial damage and lymphatic insufficiency.

Stimulation of electron transport from photosystem II to photosystem I in spinach chloroplasts

Electron transport from Photosystem II to Photosystem I of spinach chloroplasts can be stimulated by bicarbonate and various carbonyl or carboxyl compounds. Monovalent or divalent cations, which have hitherto been implicated in the energy distribution between the two photosystems, i.e., "spillover" phenomena at low light intensities, show a similar effect under high light conditions employed in this study. A mechanism for this stimulation of forward electron transport from Photosystem II to Photosystem I could involve inhibition of two types of Photosystem II partial reactions, which may involve cycling of electrons around Photosystem II. One of these is the DCMU-insensitive silicomolybdate reduction, and the other is ferricyanide reduction by Photosystem II at pH 8 in the presence of dibromothymoquinone. Greater stimulation of forward electron transport reactions is observed when both types of PHotosystem II cyclic reactions are inhibited by bicarbonate, carbonyl and carboxyl-type compounds, or by certain mono- or divalent cations.

Bilateral pulmonary varicosities associated with coarctation of the aorta

Varicosities of the pulmonary vein are rare. Twenty-eight cases confirmed anatomically or by angiography have been reported. In this paper the authors report a rather unique patient who had bilateral varices of the lungs in conjunction with coarctation of the aorta. The plain film and angiographic features of pulmonary varices are discussed. A brief review of the cases reported previously is also presented.