BackgroundDisruption of the endocannabinoid system has been implicated in both migraine and depression, which are frequently comorbid in patients. Nevertheless, supporting evidence for this hypothesis is limited. Hence, we investigated whether the endocannabinoid system is disrupted in interictal migraine patients and whether confounding factors were of influence.MethodsTo this end, we measured the levels of endocannabinoids N-arachidonoylethanolamine (AEA) and 2-arachidonoylglycerol (2-AG), and the endocannabinoid analogue docosahexaenoylethanolamine (DHEA) in the cerebrospinal fluid (CSF), obtained from 194 individuals with episodic migraine between attacks (97 migraine with aura and 97 migraine without aura individuals) and from 94 healthy volunteers, using micro-liquid chromatography - tandem mass spectrometry (micro-LC-MS/MS). Groups were age- and sex-matched. Timing and processing of sampling was protocolized. Multivariate linear regression analyses for log10-transformed endocannabinoid concentrations were performed to compare groups. Covariates that were included were sex, age, BMI, substance use and depressive symptoms.ResultsNo differences were detected between individuals with migraine with or without aura and healthy controls in the CSF for AEA (p = 0.46, p = 0.08), 2-AG (p = 0.46, p = 0.90) and DHEA (p = 0.58, p = 0.79), respectively. For 2-AG, levels were, however, positively correlated with depressive symptoms (β = 0.053; 95% CI: 0.014 to 0.093, p = 0.008).ConclusionCSF endocannabinoids (AEA, 2-AG, and DHEA) may not serve as effective biomarkers for distinguishing interictal episodic migraine patients from controls. Instead, CSF levels of 2-AG are linked to depressive symptoms, shedding further light on the relevance of endocannabinoids in patients with migraine and depression. This study highlights the intricate nature of the endocannabinoid system concerning episodic migraine and confounding factors.

Many central nervous system disorders (CNS), including chronic pain and migraine, involve metabolic changes in the brain. These changes are best detected and monitored in cerebrospinal fluid (CSF), which requires lumbar puncture. Blood-based measurements may offer an alternative, if they reflect CSF changes. To assess this, we measured and correlated the concentrations of 39 amino acids, biogenic amines, and other amines in blood and CSF of 95 healthy volunteers and, in addition, correlated the ratios of 741 amines. Amines were measured using a validated UPLC-MS platform. In healthy volunteers, only 4/39 (10.3%) analyzed amine metabolite concentrations had a correlation coefficient ≥ 0.70. Correlations of metabolite ratios were significantly better for 308/741 (41.5%) combinations. Specifically, ratios of amino acids showed high correlations. In addition, amines were investigated in 197 participants with migraine. Six amine metabolite ratios were different in migraineurs versus healthy volunteers. Most blood amine concentrations do not reflect those in CSF, but many of the ratios did correlate between CSF and plasma, showing diagnostic potential. This study improves our understanding of blood-CSF relationships, and our data suggest that ratios of amines may be of relevance to CNS disorders, as we showed for migraine.

Alternating hemiplegia of childhood (AHC) is a rare neurodevelopment disorder that is typically characterized by debilitating episodic attacks of hemiplegia, seizures, and intellectual disability. Over 85% of individuals with AHC have a de novo missense variant in ATP1A3 encoding the catalytic α3 subunit of neuronal Na +/ K + ATPases. The remainder of the patients are genetically unexplained. Here, we used next-generation sequencing to search for the genetic cause of 26 ATP1A3 -negative index patients with a clinical presentation of AHC or an AHC-like phenotype. Three patients had affected siblings. Using targeted sequencing of exonic, intronic, and flanking regions of ATP1A3 in 22 of the 26 index patients, we found no ultra-rare variants. Using exome sequencing, we identified the likely genetic diagnosis in 9 probands (35%) in five genes, including RHOBTB2 ( n = 3), ATP1A2 ( n = 3), ANK3 ( n = 1), SCN2A ( n = 1), and CHD2 ( n = 1). In follow-up investigations, two additional ATP1A3 -negative individuals were found to have rare missense SCN2A variants, including one de novo likely pathogenic variant and one likely pathogenic variant for which inheritance could not be determined. Functional evaluation of the variants identified in SCN2A and ATP1A2 supports the pathogenicity of the identified variants. Our data show that genetic variants in various neurodevelopmental genes, including SCN2A , lead to AHC or AHC-like presentation. Still, the majority of ATP1A3 -negative AHC or AHC-like patients remain unexplained, suggesting that other mutational mechanisms may account for the phenotype or that cases may be explained by oligo- or polygenic risk factors.

Hemiplegic migraine (HM) is a rare subtype of migraine with aura. Given that causal missense mutations in the voltage-gated calcium channel α1A subunit gene CACNA1A have been identified in a subset of HM patients, we investigated whether HM patients without a mutation have an increased burden of such variants in the “CACNA1x gene family”. Whole exome sequencing data of an Australian cohort of unrelated HM patients (n = 184), along with public data from gnomAD, as controls, was used to assess the burden of missense variants in CACNA1x genes. We performed both a variant and a subject burden test. We found a significant burden for the number of variants in CACNA1E (p = 1.3 × 10−4), CACNA1H (p < 2.2 × 10−16) and CACNA1I (p < 2.2 × 10−16). There was also a significant burden of subjects with missense variants in CACNA1E (p = 6.2 × 10−3), CACNA1H (p < 2.2 × 10−16) and CACNA1I (p < 2.2 × 10−16). Both the number of variants and number of subjects were replicated for CACNA1H (p = 3.5 × 10−8; p = 0.012) and CACNA1I (p = 0.019, p = 0.044), respectively, in a Dutch clinical HM cohort (n = 32), albeit that CACNA1I did not remain significant after multiple testing correction. Our data suggest that HM, in the absence of a single causal mutation, is a complex trait, in which an increased burden of missense variants in CACNA1H and CACNA1I may contribute to the risk of disease.

Migraine is a disabling episodic brain disorder with an increased familial relative risk, an increased concordance in monozygotic twins, and an estimated heritability of approximately 50%. Various genetic approaches have been applied to identify genetic factors conferring migraine risk. Initially, candidate gene associations studies (CGAS) have been performed that test DNA variants in genes prioritized based on presumed a priori knowledge of migraine pathophysiology. More recently, genome-wide association studies (GWAS) are applied that test genetic variants, single-nucleotide polymorphisms (SNPs), in a hypothesis-free manner. To date, GWAS have identified ~40 genetic loci associated with migraine. New GWAS data, which are expected to come out soon, will reveal over 100 loci. Also, large-scale GWAS, which have appeared for many traits over the last decade, have enabled studying the overlap in genetic architecture between migraine and its comorbid disorders. Importantly, other genetic factors that cannot be identified by a GWAS approach also confer risk for migraine. First steps have been taken to determine the contribution of these mechanisms by investigating mitochondrial DNA and epigenetic mechanisms. In addition to typical epigenetic mechanisms, that is, DNA methylation and histone modifications, also RNA-based mechanisms regulating gene silencing and activation have recently gotten attention. Regardless, until now, most relevant genetic discoveries related to migraine still come from investigating monogenetic syndromes with migraine as a prominent part of the phenotype. Experimental studies on these syndromes have expanded our knowledge on the mechanisms underlying migraine pathophysiology. It can be envisaged that when all (epi)genetic and phenotypic data on the common and rare forms of migraine will be integrated, this will help to unravel the biological mechanisms for migraine, which will likely guide decision-making in clinical practice in the future.

ObjectiveTo assess whether migraine may be genetically and/or causally associated with inflammatory bowel disease (IBD) or celiac disease. BackgroundMigraine has been linked to IBD and celiac disease in observational studies, but whether this link may be explained by a shared genetic basis or could be causal has not been established. The presence of a causal association could be clinically relevant, as treating one of these medical conditions might mitigate the symptoms of a causally linked condition. MethodsLinkage disequilibrium score regression and two-sample bidirectional Mendelian randomization analyses were performed using summary statistics from cohort-based genome-wide association studies of migraine (59,674 cases; 316,078 controls), IBD (25,042 cases; 34,915 controls) and celiac disease (11,812 or 4533 cases; 11,837 or 10,750 controls). Migraine with and without aura were analyzed separately, as were the two IBD subtypes Crohn's disease and ulcerative colitis. Positive control analyses and conventional Mendelian randomization sensitivity analyses were performed. ResultsMigraine was not genetically correlated with IBD or celiac disease. No evidence was observed for IBD (odds ratio [OR] 1.00, 95% confidence interval [CI] 0.99-1.02, p = 0.703) or celiac disease (OR 1.00, 95% CI 0.99-1.02, p = 0.912) causing migraine or migraine causing either IBD (OR 1.08, 95% CI 0.96-1.22, p = 0.181) or celiac disease (OR 1.08, 95% CI 0.79-1.48, p = 0.614) when all participants with migraine were analyzed jointly. There was some indication of a causal association between celiac disease and migraine with aura (OR 1.04, 95% CI 1.00-1.08, p = 0.045), between celiac disease and migraine without aura (OR 0.95, 95% CI 0.92-0.99, p = 0.006), as well as between migraine without aura and ulcerative colitis (OR 1.15, 95% CI 1.02-1.29, p = 0.025). However, the results were not significant after multiple testing correction. ConclusionsWe found no evidence of a shared genetic basis or of a causal association between migraine and either IBD or celiac disease, although we obtained some indications of causal associations with migraine subtypes.

Objective The objective of this study was to aggregate data for the first genomewide association study meta‐analysis of cluster headache, to identify genetic risk variants, and gain biological insights. Methods A total of 4,777 cases (3,348 men and 1,429 women) with clinically diagnosed cluster headache were recruited from 10 European and 1 East Asian cohorts. We first performed an inverse‐variance genomewide association meta‐analysis of 4,043 cases and 21,729 controls of European ancestry. In a secondary trans‐ancestry meta‐analysis, we included 734 cases and 9,846 controls of East Asian ancestry. Candidate causal genes were prioritized by 5 complementary methods: expression quantitative trait loci, transcriptome‐wide association, fine‐mapping of causal gene sets, genetically driven DNA methylation, and effects on protein structure. Gene set and tissue enrichment analyses, genetic correlation, genetic risk score analysis, and Mendelian randomization were part of the downstream analyses. Results The estimated single nucleotide polymorphism (SNP)‐based heritability of cluster headache was 14.5%. We identified 9 independent signals in 7 genomewide significant loci in the primary meta‐analysis, and one additional locus in the trans‐ethnic meta‐analysis. Five of the loci were previously known. The 20 genes prioritized as potentially causal for cluster headache showed enrichment to artery and brain tissue. Cluster headache was genetically correlated with cigarette smoking, risk‐taking behavior, attention deficit hyperactivity disorder (ADHD), depression, and musculoskeletal pain. Mendelian randomization analysis indicated a causal effect of cigarette smoking intensity on cluster headache. Three of the identified loci were shared with migraine. Interpretation This first genomewide association study meta‐analysis gives clues to the biological basis of cluster headache and indicates that smoking is a causal risk factor. ANN NEUROL 2023;94:713–726

BackgroundMigraine is considered a multifactorial genetic disorder. Different platforms and methods are used to unravel the genetic basis of migraine. Initially, linkage analysis in multigenerational families followed by Sanger sequencing of protein-coding parts (exons) of genes in the genomic region shared by affected family members identified high-effect risk DNA mutations for rare Mendelian forms of migraine, foremost hemiplegic migraine. More recently, genome-wide association studies testing millions of DNA variants in large groups of patients and controls have proven successful in identifying many dozens of low-effect risk DNA variants for the more common forms of migraine with the number of associated DNA variants increasing steadily with larger sample sizes. Currently, next-generation sequencing, utilising whole exome and whole genome sequence data, and other omics data are being used to facilitate their functional interpretation and the discovery of additional risk factors. Various methods and analysis tools, such as genetic correlation and causality analysis, are used to further characterise genetic risk factors. FindingsWe describe recent findings in genome-wide association studies and next-generation sequencing analysis in migraine. We show that the combined results of the two most recent and most powerful migraine genome-wide association studies have identified a total of 178 LD-independent (r(2) < 0.1) genome-wide significant single nucleotide polymorphisms (SNPs), of which 99 were unique to Hautakangas et al., 11 were unique to Choquet et al., and 68 were identified by both studies. When considering that Choquet et al. also identified three SNPs in a female-specific genome-wide association studies then these two recent studies identified 181 independent SNPs robustly associated with migraine. Cross-trait and causal analyses are beginning to identify and characterise specific biological factors that contribute to migraine risk and its comorbid conditions. ConclusionThis review provides a timely update and overview of recent genetic findings in migraine.

The endocannabinoid system (ECS) is implicated in various brain disorders. Changes in the composition of the cerebrospinal fluid (CSF) may be associated with ECS-related pathologies. Endocannabinoids (eCBs) and their analogues are present at low concentrations in human CSF, which hampered the investigation of the ECS in this body fluid. In this study, we developed a highly sensitive and selective micro-flow liquid chromatography-tandem mass spectrometry (micro-LC-MS/MS) method for the analysis of eCBs and eCB analogues in human CSF. The developed method allowed for the quantitative analysis of 16 eCBs and their analogues in human CSF. Micro-LC-MS/MS analyses were performed at a flow-rate of 4 μL min-1 with a 0.3-mm inner diameter column. A minor modification of a novel spray needle was carried out to improve the robustness of our method. By using an injection volume of 3 μL, our method reached limits of detection in the range from 0.6 to 1293.4 pM and limits of quantification in range from 2.0 to 4311.3 pM while intra- and interday precisions were below 13.7%. The developed workflow was successfully used for the determination of eCBs in 288 human CSF samples. It is anticipated that the proposed approach will contribute to a deeper understanding of the role of ECS in various brain disorders.

Migraine affects over a billion individuals worldwide but its genetic underpinning remains largely unknown. Here, we performed a genome-wide association study of 102,084 migraine cases and 771,257 controls and identified 123 loci, of which 86 are previously unknown. These loci provide an opportunity to evaluate shared and distinct genetic components in the two main migraine subtypes: migraine with aura and migraine without aura. Stratification of the risk loci using 29,679 cases with subtype information indicated three risk variants that seem specific for migraine with aura (in HMOX2, CACNA1A and MPPED2), two that seem specific for migraine without aura (near SPINK2 and near FECH) and nine that increase susceptibility for migraine regardless of subtype. The new risk loci include genes encoding recent migraine-specific drug targets, namely calcitonin gene-related peptide (CALCA/CALCB) and serotonin 1F receptor (HTR1F). Overall, genomic annotations among migraine-associated variants were enriched in both vascular and central nervous system tissue/cell types, supporting unequivocally that neurovascular mechanisms underlie migraine pathophysiology.

Administration of glyceryl trinitrate (GTN), a donor of nitric oxide, can induce migraine-like attacks in subjects with migraine. Provocation with GTN typically follows a biphasic pattern; it induces immediate headache in subjects with migraine, as well as in healthy controls, whereafter only subjects with migraine may develop a migraine-like headache several hours later. Interestingly, intravenous infusion with prostaglandin-E2 (PGE2) can also provoke a migraine-like headache, but seems to have a more rapid onset compared to GTN. The aim of the study was to shed light on the mechanistic aspect PGE2 has in migraine attack development. Therefore, PGE2 plasma levels were measured towards the (pre)ictal state of an attack, which we provoked with GTN. Blood samples from women with migraine (n = 37) and age-matched female controls (n = 25) were obtained before and- 140 min and- 320 min after GTN infusion. PGE2 levels were measured using liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. Data was analyzed using a generalized linear mixed-effect model. Immediate headache after GTN infusion occurred in 85 % of migraine participants and in 75 % of controls. A delayed onset migraine-like attack was observed in 82 % of migraine subjects and in none of the controls. PGE2 levels were not different between the interictal and preictal state (P = 0.527) nor between interictal and ictal state (defined as having migraine-like headache) (P = 0.141). Hence, no evidence was found that a rise in PGE2 is an essential step in the initiation of GTN-induced migraine-like attacks.

ObjectiveThis study was undertaken to identify susceptibility loci for cluster headache and obtain insights into relevant disease pathways.MethodsWe carried out a genome‐wide association study, where 852 UK and 591 Swedish cluster headache cases were compared with 5,614 and 1,134 controls, respectively. Following quality control and imputation, single variant association testing was conducted using a logistic mixed model for each cohort. The 2 cohorts were subsequently combined in a merged analysis. Downstream analyses, such as gene‐set enrichment, functional variant annotation, prediction and pathway analyses, were performed.ResultsInitial independent analysis identified 2 replicable cluster headache susceptibility loci on chromosome 2. A merged analysis identified an additional locus on chromosome 1 and confirmed a locus significant in the UK analysis on chromosome 6, which overlaps with a previously known migraine locus. The lead single nucleotide polymorphisms were rs113658130 (p = 1.92 × 10−17, odds ratio [OR] = 1.51, 95% confidence interval [CI] = 1.37–1.66) and rs4519530 (p = 6.98 × 10−17, OR = 1.47, 95% CI = 1.34–1.61) on chromosome 2, rs12121134 on chromosome 1 (p = 1.66 × 10−8, OR = 1.36, 95% CI = 1.22–1.52), and rs11153082 (p = 1.85 × 10−8, OR = 1.30, 95% CI = 1.19–1.42) on chromosome 6. Downstream analyses implicated immunological processes in the pathogenesis of cluster headache.InterpretationWe identified and replicated several genome‐wide significant associations supporting a genetic predisposition in cluster headache in a genome‐wide association study involving 1,443 cases. Replication in larger independent cohorts combined with comprehensive phenotyping, in relation to, for example, treatment response and cluster headache subtypes, could provide unprecedented insights into genotype–phenotype correlations and the pathophysiological pathways underlying cluster headache. ANN NEUROL 2021;90:193–202

We report detailed functional analyses and genotype-phenotype correlations in 433 individuals carrying disease-causing variants in SCN8A, encoding the voltage-gated Na+ channel NaV1.6. Five different clinical subgroups could be identified: 1) Benign familial infantile epilepsy (BFIE) (n=17, normal cognition, treatable seizures), 2) intermediate epilepsy (n=36, mild ID, partially pharmacoresponsive), 3) developmental and epileptic encephalopathy (DEE, n=191, severe ID, majority pharmacoresistant), 4) generalized epilepsy (n=21, mild to moderate ID, frequently with absence seizures), and 5) affected individuals without epilepsy (n=25, mild to moderate ID). Groups 1-3 presented with early-onset (median: four months) focal or multifocal seizures and epileptic discharges, whereas the onset of seizures in group 4 was later (median: 39 months) with generalized epileptic discharges. The epilepsy was not classifiable in 143 individuals. We performed functional studies expressing missense variants in ND7/23 neuroblastoma cells and primary neuronal cultures using recombinant tetrodotoxin insensitive human NaV1.6 channels and whole-cell patch clamping. Two variants causing DEE showed a strong gain-of-function (GOF, hyperpolarising shift of steady-state activation, strongly increased neuronal firing rate), and one variant causing BFIE or intermediate epilepsy showed a mild GOF (defective fast inactivation, less increased firing). In contrast, all three variants causing generalized epilepsy induced a loss-of-function (LOF, reduced current amplitudes, depolarising shift of steady-state activation, reduced neuronal firing). Including previous studies, functional effects were known for 165 individuals. All 133 individuals carrying GOF variants had either focal (76, groups 1-3), or unclassifiable epilepsy (37), whereas 32 with LOF variants had either generalized (14), no (11) or unclassifiable (5) epilepsy; only two had DEE. Computational modeling in the GOF group revealed a significant correlation between the severity of the electrophysiological and clinical phenotypes. GOF variant carriers responded significantly better to sodium channel blockers (SCBs) than to other anti-seizure medications, and the same applied for all individuals of groups 1-3. In conclusion, our data reveal clear genotype-phenotype correlations between age at seizure onset, type of epilepsy and gain- or loss-of-function effects of SCN8A variants. Generalized epilepsy with absence seizures is the main epilepsy phenotype of LOF variant carriers and the extent of the electrophysiological dysfunction of the GOF variants is a main determinant of the severity of the clinical phenotype in focal epilepsies. Our pharmacological data indicate that SCBs present a therapeutic treatment option in early onset SCN8A-related focal epilepsy.

Migraine is a common brain disorder but reliable diagnostic biomarkers in blood are still lacking. Our aim was to identify, using proton nuclear magnetic resonance (1H-NMR) spectroscopy, metabolites in serum that are associated with lifetime and active migraine by comparing metabolic profiles of patients and controls. Fasting serum samples from 313 migraine patients and 1512 controls from the Erasmus Rucphen Family (ERF) study were available for 1H-NMR spectroscopy. Data was analysed using elastic net regression analysis. A total of 100 signals representing 49 different metabolites were detected in 289 cases (of which 150 active migraine patients) and 1360 controls. We were able to identify profiles consisting of 6 metabolites predictive for lifetime migraine status and 22 metabolites predictive for active migraine status. We estimated with subsequent regression models that after correction for age, sex, BMI and smoking, the association with the metabolite profile in active migraine remained. Several of the metabolites in this profile are involved in lipid, glucose and amino acid metabolism. This study indicates that metabolic profiles, based on serum concentrations of several metabolites, including lipids, amino acids and metabolites of glucose metabolism, can distinguish active migraine patients from controls.

Objective Identifying common genetic variants that confergenetic risk for cluster headache. Methods We conducted a case-control study in the Dutch Leiden University Cluster headache neuro-Analysis program (LUCA) study population (n = 840) and unselected controls from the Netherlands Epidemiology of Obesity Study (NEO; n = 1,457). Replication was performed in a Norwegian sample of 144 cases from the Trondheim Cluster headache sample and 1,800 controls from the Nord-Trondelag Health Survey (HUNT). Gene set and tissue enrichment analyses, blood cell-derived RNA-sequencing of genes around the risk loci and linkage disequilibrium score regression were part of the downstream analyses. Results An association was found with cluster headache for 4 independent loci (r(2) < 0.1) with genomewide significance (p < 5 x 10(-8)), rs11579212 (odds ratio [OR] = 1.51, 95% confidence interval [CI] = 1.33-1.72 near RP11-815 M8.1), rs6541998 (OR = 1.53, 95% CI = 1.37-1.74 near MERTK), rs10184573 (OR = 1.43, 95% CI = 1.26-1.61 near AC093590.1), and rs2499799 (OR = 0.62, 95% CI = 0.54-0.73 near UFL1/FHL5), collectively explaining 7.2% of the variance of cluster headache. SNPs rs11579212, rs10184573, and rs976357, as proxy SNP for rs2499799 (r(2) = 1.0), replicated in the Norwegian sample (p < 0.05). Gene-based mapping yielded ASZ1 as possible fifth locus. RNA-sequencing indicated differential expression of POLR1B and TMEM87B in cluster headache patients. Interpretation This genomewide association study (GWAS) identified and replicated genetic risk loci for cluster headaches with effect sizes larger than those typically seen in complex genetic disorders. ANN NEUROL 2021

The Mediator is an evolutionarily conserved, multi-subunit complex that regulates multiple steps of transcription. Mediator activity is regulated by the reversible association of a four-subunit module comprising CDK8 or CDK19 kinases, together with cyclin C, MED12 or MED12L, and MED13 or MED13L. Mutations in MED12, MED13, and MED13L were previously identified in syndromic developmental disorders with overlapping phenotypes. Here, we report CDK8 mutations (located at 13q12.13) that cause a phenotypically related disorder. Using whole-exome or whole-genome sequencing, and by international collaboration, we identified eight different heterozygous missense CDK8 substitutions, including 10 shown to have arisen de novo, in 12 unrelated subjects; a recurrent mutation, c. 185C>T (p.Ser62Leu), was present in five individuals. All predicted substitutions localize to the ATP-binding pocket of the kinase domain. Affected individuals have overlapping phenotypes characterized by hypotonia, mild to moderate intellectual disability, behavioral disorders, and variable facial dysmorphism. Congenital heart disease occurred in six subjects; additional features present in multiple individuals included agenesis of the corpus callosum, ano-rectal malformations, seizures, and hearing or visual impairments. To evaluate the functional impact of the mutations, we measured phosphorylation at STAT1-Ser727, a known CDK8 substrate, in a CDK8 and CDK19 CRISPR double-knockout cell line transfected with wild-type (WT) or mutant CDK8 constructs. These experiments demonstrated a reduction in STAT1 phosphorylation by all mutants, in most cases to a similar extent as in a kinase-dead control. We conclude that missense mutations in CDK8 cause a developmental disorder that has phenotypic similarity to syndromes associated with mutations in other subunits of the Mediator kinase module, indicating probable overlap in pathogenic mechanisms.

Genomewide association studies (GWASs) have contributed greatly to unraveling the genetic basis of Alzheimer's disease (AD). However, a large amount of "missing heritability" remains. In this exploratory study, we investigated the effect of cytosine-adenine-guanine (CAG) repeats in polyglutamine disease-associated genes (PDAGs) on the risk of AD and its expression. In a cohort of 959 patients diagnosed with AD (Amsterdam Dementia cohort) and 4106 cognitively healthy participants (Leiden 85-plus Study and the Prospective Study of Pravastatin in the Elderly at Risk), we determined the CAG repeat sequences in ATXN1, ATXN2, ATXN3, CACNA1A, ATXN7, TBP, HTT, ATN1, and AR. We did not find a significant association between the risk of AD and variations in CAG repeat numbers of PDAGs. However, we found that differences in CAG repeat numbers in ATXN1, ATXN2, and AR were significantly associated with several clinical and imaging features in AD patients. Specifically, the association between memory performance in patients with AD and the CAG repeat size in the longer ATXN1 allele, and the association between atrophy in the medial temporal lobes and the CAG repeat number in the longer AR allele remained significant after correction for multiple testing. Our findings suggest that repeat polymorphisms in ATXN1 and AR can act as important genetic modifiers of AD, warranting further scrutiny of their role in its missing heritability and pathogenesis.