Journal of Cerebral Blood Flow & Metabolism (2015), 1–4 © 2015 ISCBFM All rights reserved 0271-678X/15 $32.00 www.jcbfm.com
BRIEF COMMUNICATION
APOE ε2 is associated with white matter hyperintensity volume in CADASIL Benno Gesierich1,15, Christian Opherk1,2,15, Jonathan Rosand3,4, Mariya Gonik1, Rainer Malik1, Eric Jouvent5, Dominique Hervé6, Poneh Adib-Samii7, Steve Bevan7, Luigi Pianese8, Serena Silvestri8, Maria T Dotti9, Nicola De Stefano9, Jeroen van der Grond10, Elles MJ Boon11, Francesca Pescini12, Natalia Rost3, Leonardo Pantoni12, Saskia A Lesnik Oberstein11, Antonio Federico9, Michele Ragno8, Hugh S Markus13, Elisabeth Tournier-Lasserve6, Hugues Chabriat5, Martin Dichgans1,14, Marco Duering1,15 and Michael Ewers1,15 Apolipoprotein E (APOE) increases the risk for Alzheimer’s disease (ε4 allele) and cerebral amyloid angiopathy (ε2 and ε4), but its role in small vessel disease (SVD) is debated. Here we studied the effects of APOE on white matter hyperintensity volume (WMHV) in CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy), a nonamyloidogenic angiopathy and inherited early-onset form of pure SVD. Four hundred and eighty-eight subjects were recruited through a multicenter consortium. Compared with APOE ε3/ε3, WMHV was increased in APOE ε2 (P = 0.02) but not APOE ε4. The results remained significant when controlled for genome-wide genetic background variation. Our findings suggest a modifying influence of APOE ε2 on WMHV caused by pure SVD. Journal of Cerebral Blood Flow & Metabolism advance online publication, 29 April 2015; doi:10.1038/jcbfm.2015.85 Keywords: APOE; CADASIL; multicenter study; small vessel disease; white matter hyperintensities
INTRODUCTION Cerebral small vessel disease (SVD) is the most common cause of vascular cognitive impairment. White matter damage detectable as white matter hyperintensities (WMH) on T2-weighted magnetic resonance imaging (MRI) scans is a hallmark of SVD.1 Twin and family history studies have demonstrated a strong genetic contribution to WMH.2 In a recent meta-analysis of studies mostly on elderly individuals, both the Apolipoprotein E (APOE) ε2 and ε4 alleles were found to be associated with WMH burden.3 However, this might in part relate to the well-established role of APOE in Alzheimer’s disease and cerebral amyloid angiopathy, which are both common in the elderly population and associated with WMH.4,5 In fact, disentangling the impact of APOE on WMH in the
context of SVD, Alzheimer’s disease, and cerebral amyloid angiopathy has been difficult. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a nonamyloidogenic angiopathy caused by mutations in NOTCH3. CADASIL is considered a model of pure SVD because the primary pathology is restricted to microvessels. WMH volumes in CADASIL show a striking variability that is not explained by demographic factors or different mutations at the NOTCH3 locus.6 Previous studies on potential genetic modifiers found no influence of the APOE ε4 allele on WMH volumes.7,8 However, sample sizes were relatively small and the APOE ε2 allele was not examined. The current study aimed at investigating the
1 Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians-University, Munich, Germany; 2Department of Neurology, SLK-Kliniken Heilbronn, Heilbronn, Germany; 3Center for Human Genetic Research and Department of Neurology, Massachusetts General Hospital, Boston, Massachusetts, USA; 4Program in Medical and Population Genetics, Broad Institute, Cambridge, Massachusetts, USA; 5Department of Neurology, DHU NeuroVasc, Hopital Lariboisiere, APHP, Paris, France; 6 Université Paris Diderot, Sorbonne Paris Cité, Génétique des Maladies vasculaires INSERM UMR-S740, Paris, France; 7Stroke and Dementia Research Centre, St. George’s, University of London, London, UK; 8Neurology Division and Molecular Medicine Section, Mazzoni Hospital, Ascoli Piceno, Italy; 9Department of Medicine, Surgery and Neurosciences, University of Siena, Siena, Italy; 10Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands; 11Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands; 12Stroke Unit and Neurology, Azienda Ospedaliero Universitaria Careggi, Florence, Italy; 13Clinical Neurosciences, University of Cambridge, Cambridge, UK and 14Munich Cluster for Systems Neurology (SyNergy), Munich, Germany. Correspondence: Professor Dr M Ewers, Institute for Stroke and Dementia Research (ISD), Klinikum der Universität München, Feodor-Lynen-Str. 17, 81377 München, Germany. E-mail: [email protected] This study was supported by the Corona Foundation, the Vascular Dementia Research Foundation, the Deutsche Forschungsgemeinschaft (OP 212/1-1), the Dr Werner Jackstädt Foundation, Fondation Leducq (Transatlantic Network of Excellence on the Pathogenesis of Small Vessel Disease of the Brain), the American Heart Association–Bugher Foundation, the Massachusetts General Hospital Deane Institute for Integrative Research in Atrial Fibrillation and Stroke, and the National Institute of Neurological Disorders and Stroke (5P50NS051343; R01NS059727). Dr Markus is supported by a National Institute for Health Research Senior Investigator award. Drs Pianese and Pescini were supported by MIUR (Ministero dell’Istruzione, dell’Università e della Ricerca) programmi di ricerca cofinanziati-2006 (MIUR 2006-prot. 2006065719) and programmi di ricerca cofinanziati-2009 (MIUR 2009-prot. 20095JPSNA) and Regione Toscana, programma per la Ricerca Regionale in Materia di Salute, 2009 (Evaluation of NOTCH3 mutations and correlation with clinical phenotypes; prot. reg. AOO GRT 190899/Q.20.70.20). Dr Dotti was supported by MIUR (prot. 20095JPSNA_005). Dr Rosand has received significant research funding from the Deane Institute for Integrated Research on Atrial Fibrillation and Stroke at Massachusetts General Hospital and the National Institutes of Health. Dr Rost is supported by the National Institute of Neurological Disorders and Stroke (K23NS064052). Dr SAJ Lesnik Oberstein has received significant research funding from The Brain Foundation of The Netherlands and The Netherlands Organisation for Health Research and Development (ZonMw). 15 These authors contributed equally to this work. Received 23 December 2014; revised 20 March 2015; accepted 3 April 2015
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
2
effects of the APOE ε2 and ε4 alleles on WMH burden in the context of pure SVD. This was performed through a large multicenter study in patients with CADASIL. MATERIALS AND METHODS Subjects A total of 552 CADASIL patients were assessed at seven university hospitals across Europe as part of the CADASIL genome-wide association study consortium.6 Nineteen patients were excluded for technical problems during genotyping, another 20 patients because of missing MRI data, and 25 patients because of missing demographic/clinical data. Thus, the final sample size consisted of 488 patients (Supplementary Table). The aims and recruitment criteria of the consortium have been reported in detail.6 In brief, all subjects underwent a detailed clinical interview after physical examination, blood draw for isolation of DNA, and structural MRI. The study was conducted in adherence to the Good Clinical Practice guidelines and to the Declaration of Helsinki including all revisions. Each site obtained ethical approval from the respective local institutional review board. All participants of the study gave written informed consent for study participation, MRI scanning, and use of DNA. Genotyping APOE carrier status was determined centrally using an iPlex assay (Sequenom, San Diego, CA, USA). Data from a genome-wide set of 466 single nucleotide polymorphism (SNP) markers from the Affymetrix 6.0 platform (Affymetrix UK, High Wycombe, UK)6 was used to control for variations in the genetic background of the study cohort (see Statistics section). Because genotyping of the large genome-wide set of SNP markers was less robust than the targeted genotyping of APOE, SNP data were available for only 407 out of the 488 subjects. None of these SNP markers was in linkage disequilibrium (r240.1) with the APOE genotype. Magnetic resonance imaging data processing MR scans were acquired at each site, using scanners operating at field strengths between 0.5 and 3 T. White matter hyperintensities were determined as previously described.6,9 In short, lesions were segmented on FLAIR images by a semiautomated method and then checked and corrected by trained raters. The interrater reliability for this procedure was excellent with an intraclass correlation coefficient of 0.996. The WMH volumes were normalized to the intracranial cavity volume by dividing the WMH volume by the intracranial cavity volume, which had been derived from T2- or proton density-weighted images. Given the skewed distribution of the normalized WMH volumes, a square-root transformation was performed. The normalized and transformed WMH volumes are referred to as nWMHV. A fit of the normal
Table 1.
distribution curve onto the distribution of nWMHV is shown in the Supplementary Figure 1. Statistical analysis Demographics, vascular risk factors, and WMH volume were compared across APOE groups, using analysis of variance for continuous variables and Χ2-tests of independence for bimodal variables. Multiple linear regression analysis was used to determine the effect of APOE status on nWMHV. Specifically, we examined whether carriers of the APOE ε2 or ε4 allele showed a difference in nWMHV compared with noncarriers. Because of the small number of subjects homozygous for ε4 (n = 8) and the absence of subjects homozygous for ε2 we limited our analyses to ε2- and ε4-carrier status. We encoded ε2 carriers (i.e., ε2/ε3) and ε4 carriers (i.e., ε3/ε4 and ε4/ε4) as dummy variables, with noncarriers (i.e., ε3/ε3) being the reference group. We performed a stepwise regression analysis with forward selection. nWMHV was entered as dependent variable and APOE carrier status, age, sex, hypertension, hypercholesterolemia, diabetes, and past/current smoker were entered as independent variables. To minimize the possibility that any APOE effects are driven by differences in genetic background, we conducted an additional regression analysis controlling for genetic background variation in a subset of 407 patients in which the genome-wide set of SNP markers were available.6 To reduce dimensionality, identity-bystate–based principal component analysis was performed on the SNP markers.6 On the basis of visual inspection of a plot showing the decrease of the Eigenvalues across the rank-ordered principal components, the first five principal components were selected. Further, we ran a second stepwise regression with forward selection, entering these five principal components in addition to the aforementioned independent variables. RESULTS Demographic information and vascular risk factors are reported in Table 1. The most frequent NOTCH3 mutations were R1006C and R182C (each n = 30). Stepwise multiple regression analysis resulted in a model including APOE carrier status and age as significant independent variables. APOE ε2 carriers showed increased nWMHV compared with the reference genotype ε3/ε3 (β = 0.031, t = 2.4, P = 0.0185). In contrast, there was no significant association between ε4-carrier status and nWMHV (β = 0.017, t = 1.7, P = 0.0935), compared with the same reference group. The sizes of the effects of individual APOE alleles on nWMHV are shown in Figure 1 (model 1). The covariate age was associated with higher nWMHV (β = 0.005, t = 15.1, P o 2E − 16). When further controlling for genetic background variation, the results remained essentially unchanged, i.e., nWMHV was significantly increased in ε2 carriers (β = 0.028, t = 2.0, P = 0.0496) but not in ε4 carriers (β = 0.008, t = 0.8, P = 0.4428;
Demographics, vascular risk factors, and WMH volume in the different APOE groups
APOE status Age, median (IQR) (years) Sex, male, n (%) Hypertension, n (%) Hypercholesterolemia, n (%) Diabetes melitus, n (%) Past/current smoker, n (%) WMHV, median (IQR) (%)
ε3/ε3, n = 350 (71.7%) 51 169 87 157 14 87 5.2
(44–60) (48.3) (24.9) (44.9) (4.0) (26.4) (2.6–8.5)
ε2 carriers, n = 46 (9.4%) 48 21 9 10 0 10 5.7
(42–54) (45.7) (19.6) (21.7) (0) (22.2) (2.3–12.7)
ε4 carriers, n = 92 (18.9%) 52 38 18 45 4 26 5.7
(41–59) (41.3) (19.6) (48.9) (4.3) (29.9) (2.6–10.1)
P value 0.606 0.484 0.461 0.006 0.374 0.628 0.233
Abbreviations: APOE, apolipoprotein E; IQR, interquartile range; WMHV, white matter hyperintensity volume (as percentage of intracranial cavity volume).
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
© 2015 ISCBFM
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
3
Figure 1. Effect sizes (i.e., regression coefficients) of the different APOE alleles on normalized white matter hyperintensities volumes, derived from the regression model with genotype ε3/ε3 being the reference and controlling for age (model 1) plus genetic background (model 2) as covariates. The bars indicate 95% confidence intervals of the regression coefficients.
Figure 1, model 2). Box plots for nWMHV after regressing out age and genetic background are provided in Supplementary Figure 2. DISCUSSION Our results demonstrate an association between the presence of the APOE ε2 allele and higher WMHV in a well-defined population of subjects with inherited pure SVD. In contrast, the APOE ε4 genotype was not associated with WMHV. Our finding on APOE ε2 is consistent with results from a recent meta-analysis in sporadic age-related SVD, showing an association between APOE ε2 allele and increased WMHV.3 For APOE ε4, we found a nonsignificant trend for an effect on WMHV, which disappeared when controlling for genetic background. These results on APOE ε4 are in agreement with previous studies in CADASIL that reported no effect of APOE ε4 on white matter damage.7,8 However, we cannot exclude an association between homozygous APOE ε4 genotype on WMHV3 as this could not be assessed in the current study owing to the low number of subjects with this genotype (n = 8). In summary, our results show that APOE ε2 allele contributes to the severity of white matter injury in subject with pure SVD. We can only speculate on the mechanisms by which the ε2 allele increases WMHV in this nonamyloidogenic form of SVD. The APOE ε4 and ε2 allele both have been found to increase the risk of cerebral amyloid angiopathy through an increased deposition of vascular amyloid beta (Aβ),10 which may in turn relate to impaired vascular drainage of Aβ.11 Although vascular Aβ deposition has no role in CADASIL, APOE ε2 may be associated with impaired drainage of other proteins leading to microvascular damage.12 Furthermore, the ε2 allele is associated with specific vessel pathologies in CAA, such as vessel dilation, fibrinoid necrosis, microaneurysms and double barreling.13 Some of these vasculopathic changes have also been observed in CADASIL. Hence, APOE ε2 might likewise influence these vascular pathologies in CADASIL. Another possibility is that the presence of the APOE ε2 allele is associated with increased WMH through its influence on inflammation. In population-based studies, higher plasma levels of lipoprotein-associated phospholipase A2, a marker of vascular inflammation, were associated with more severe SVD as measured by cerebral microbleeds and WMH.14,15 Although the association between lipoprotein-associated phospholipase A2 and higher © 2015 ISCBFM
prevalence of cerebral microbleeds was dependent on the presence of an APOE ε2 allele,14 such a modulatory effect of APOE ε 2 on the effect of inflammation on WMH remains to be tested in future studies. Specific strengths of the current study include the large cohort of carefully phenotyped patients with genetically defined SVD. Also, the WMH segmentations and genotyping were both performed on a central platform, thus minimizing the betweencenter variability. Our statistical results were robust, with the ε2 effect remaining significant even when controlling for variations in genetic background. Controlling for the genetic background is important, since subtle genetic variation within or between centers could introduce a major bias in the results. A limiting factor is that our study focused on WMH, but we did not assess other imaging features of SVD such as cerebral microbleeds and lacunes. Also, data regarding clinical disease severity (e.g., dementia, onset of cognitive impairment, and stroke) have not been collected. The consortium was established with the specific goal to assess genetic modifiers of WMH in a unique sample of patients with genetically determined SVD. The selection of patients was performed accordingly and imaging modalities needed to assess other SVD marker are sparsely available within the sample. DISCLOSURE/CONFLICT OF INTEREST The authors declare no conflict of interest.
DISCLAIMER Dr Pantoni is member of the editorial boards of Cerebrovascular Diseases and Acta Neurologica. REFERENCES 1 Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 2013; 12: 822–838. 2 Carmelli D, DeCarli C, Swan GE, Jack LM, Reed T, Wolf PA et al. Evidence for genetic variance in white matter hyperintensity volume in normal elderly male twins. Stroke 1998; 29: 1177–1181. 3 Schilling S, DeStefano AL, Sachdev PS, Choi SH, Mather KA, DeCarli CD et al. APOE genotype and MRI markers of cerebrovascular disease: systematic review and meta-analysis. Neurology 2013; 81: 292–300. 4 Brickman AM, Zahodne LB, Guzman VA, Narkhede A, Meier IB, Griffith EY et al. Reconsidering harbingers of dementia: progression of parietal lobe white matter hyperintensities predicts Alzheimer's disease incidence. Neurobiol Aging 2014; 36: 27–32. 5 Gurol ME, Viswanathan A, Gidicsin C, Hedden T, Martinez-Ramirez S, Dumas A et al. Cerebral amyloid angiopathy burden associated with leukoaraiosis: a positron emission tomography/magnetic resonance imaging study. Ann Neurol 2013; 73: 529–536. 6 Opherk C, Gonik M, Duering M, Malik R, Jouvent E, Herve D et al. Genome-wide genotyping demonstrates a polygenic risk score associated with white matter hyperintensity volume in CADASIL. Stroke 2014; 45: 968–972. 7 Singhal S, Bevan S, Barrick T, Rich P, Markus HS. The influence of genetic and cardiovascular risk factors on the CADASIL phenotype. Brain 2004; 127: 2031–2038. 8 van den Boom R, Lesnick Oberstein SAJ, van den Berg-Huysmans AA, Ferrari MD, van Buchem MA, Haan J. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy: structural MR imaging ghanges and apolipoprotein E genotype. Am J Neuroradiol 2006; 27: 359–362. 9 Duering M, Zieren N, Herve D, Jouvent E, Reyes S, Peters N et al. Strategic role of frontal white matter tracts in vascular cognitive impairment: a voxel-based lesion-symptom mapping study in CADASIL. Brain 2011; 134: 2366–2375. 10 Nelson PT, Pious NM, Jicha GA, Wilcock DM, Fardo DW, Estus S et al. APOEepsilon2 and APOE-epsilon4 correlate with increased amyloid accumulation in cerebral vasculature. J Neuropathol Exp Neurol 2013; 72: 708–715.
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
4 11 Thal DR, Larionov S, Abramowski D, Wiederhold KH, Van Dooren T, Yamaguchi H et al. Occurrence and co-localization of amyloid beta-protein and apolipoprotein E in perivascular drainage channels of wild-type and APP-transgenic mice. Neurobiol Aging 2007; 28: 1221–1230. 12 Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO. Review: cerebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum of protein elimination failure angiopathies (PEFA) in neurodegenerative disease with a focus on therapy. Neuropathol Appl Neurobiol 2013; 39: 593–611.
13 Greenberg SM, Vonsattel JP, Segal AZ, Chiu RI, Clatworthy AE, Liao A et al. Association of apolipoprotein E epsilon2 and vasculopathy in cerebral amyloid angiopathy. Neurology 1998; 50: 961–965. 14 Romero JR, Preis SR, Beiser AS, DeCarli C, Lee DY, Viswanathan A et al. Lipoprotein phospholipase A2 and cerebral microbleeds in the Framingham Heart Study. Stroke 2012; 43: 3091–3094. 15 Wright CB, Moon Y, Paik MC, Brown TR, Rabbani L, Yoshita M et al. Inflammatory biomarkers of vascular risk as correlates of leukoariosis. Stroke 2009; 40: 3466–3471.
Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http://www.nature.com/jcbfm)
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
© 2015 ISCBFM
BRIEF COMMUNICATION
APOE ε2 is associated with white matter hyperintensity volume in CADASIL Benno Gesierich1,15, Christian Opherk1,2,15, Jonathan Rosand3,4, Mariya Gonik1, Rainer Malik1, Eric Jouvent5, Dominique Hervé6, Poneh Adib-Samii7, Steve Bevan7, Luigi Pianese8, Serena Silvestri8, Maria T Dotti9, Nicola De Stefano9, Jeroen van der Grond10, Elles MJ Boon11, Francesca Pescini12, Natalia Rost3, Leonardo Pantoni12, Saskia A Lesnik Oberstein11, Antonio Federico9, Michele Ragno8, Hugh S Markus13, Elisabeth Tournier-Lasserve6, Hugues Chabriat5, Martin Dichgans1,14, Marco Duering1,15 and Michael Ewers1,15 Apolipoprotein E (APOE) increases the risk for Alzheimer’s disease (ε4 allele) and cerebral amyloid angiopathy (ε2 and ε4), but its role in small vessel disease (SVD) is debated. Here we studied the effects of APOE on white matter hyperintensity volume (WMHV) in CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy), a nonamyloidogenic angiopathy and inherited early-onset form of pure SVD. Four hundred and eighty-eight subjects were recruited through a multicenter consortium. Compared with APOE ε3/ε3, WMHV was increased in APOE ε2 (P = 0.02) but not APOE ε4. The results remained significant when controlled for genome-wide genetic background variation. Our findings suggest a modifying influence of APOE ε2 on WMHV caused by pure SVD. Journal of Cerebral Blood Flow & Metabolism advance online publication, 29 April 2015; doi:10.1038/jcbfm.2015.85 Keywords: APOE; CADASIL; multicenter study; small vessel disease; white matter hyperintensities
INTRODUCTION Cerebral small vessel disease (SVD) is the most common cause of vascular cognitive impairment. White matter damage detectable as white matter hyperintensities (WMH) on T2-weighted magnetic resonance imaging (MRI) scans is a hallmark of SVD.1 Twin and family history studies have demonstrated a strong genetic contribution to WMH.2 In a recent meta-analysis of studies mostly on elderly individuals, both the Apolipoprotein E (APOE) ε2 and ε4 alleles were found to be associated with WMH burden.3 However, this might in part relate to the well-established role of APOE in Alzheimer’s disease and cerebral amyloid angiopathy, which are both common in the elderly population and associated with WMH.4,5 In fact, disentangling the impact of APOE on WMH in the
context of SVD, Alzheimer’s disease, and cerebral amyloid angiopathy has been difficult. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a nonamyloidogenic angiopathy caused by mutations in NOTCH3. CADASIL is considered a model of pure SVD because the primary pathology is restricted to microvessels. WMH volumes in CADASIL show a striking variability that is not explained by demographic factors or different mutations at the NOTCH3 locus.6 Previous studies on potential genetic modifiers found no influence of the APOE ε4 allele on WMH volumes.7,8 However, sample sizes were relatively small and the APOE ε2 allele was not examined. The current study aimed at investigating the
1 Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians-University, Munich, Germany; 2Department of Neurology, SLK-Kliniken Heilbronn, Heilbronn, Germany; 3Center for Human Genetic Research and Department of Neurology, Massachusetts General Hospital, Boston, Massachusetts, USA; 4Program in Medical and Population Genetics, Broad Institute, Cambridge, Massachusetts, USA; 5Department of Neurology, DHU NeuroVasc, Hopital Lariboisiere, APHP, Paris, France; 6 Université Paris Diderot, Sorbonne Paris Cité, Génétique des Maladies vasculaires INSERM UMR-S740, Paris, France; 7Stroke and Dementia Research Centre, St. George’s, University of London, London, UK; 8Neurology Division and Molecular Medicine Section, Mazzoni Hospital, Ascoli Piceno, Italy; 9Department of Medicine, Surgery and Neurosciences, University of Siena, Siena, Italy; 10Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands; 11Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands; 12Stroke Unit and Neurology, Azienda Ospedaliero Universitaria Careggi, Florence, Italy; 13Clinical Neurosciences, University of Cambridge, Cambridge, UK and 14Munich Cluster for Systems Neurology (SyNergy), Munich, Germany. Correspondence: Professor Dr M Ewers, Institute for Stroke and Dementia Research (ISD), Klinikum der Universität München, Feodor-Lynen-Str. 17, 81377 München, Germany. E-mail: [email protected] This study was supported by the Corona Foundation, the Vascular Dementia Research Foundation, the Deutsche Forschungsgemeinschaft (OP 212/1-1), the Dr Werner Jackstädt Foundation, Fondation Leducq (Transatlantic Network of Excellence on the Pathogenesis of Small Vessel Disease of the Brain), the American Heart Association–Bugher Foundation, the Massachusetts General Hospital Deane Institute for Integrative Research in Atrial Fibrillation and Stroke, and the National Institute of Neurological Disorders and Stroke (5P50NS051343; R01NS059727). Dr Markus is supported by a National Institute for Health Research Senior Investigator award. Drs Pianese and Pescini were supported by MIUR (Ministero dell’Istruzione, dell’Università e della Ricerca) programmi di ricerca cofinanziati-2006 (MIUR 2006-prot. 2006065719) and programmi di ricerca cofinanziati-2009 (MIUR 2009-prot. 20095JPSNA) and Regione Toscana, programma per la Ricerca Regionale in Materia di Salute, 2009 (Evaluation of NOTCH3 mutations and correlation with clinical phenotypes; prot. reg. AOO GRT 190899/Q.20.70.20). Dr Dotti was supported by MIUR (prot. 20095JPSNA_005). Dr Rosand has received significant research funding from the Deane Institute for Integrated Research on Atrial Fibrillation and Stroke at Massachusetts General Hospital and the National Institutes of Health. Dr Rost is supported by the National Institute of Neurological Disorders and Stroke (K23NS064052). Dr SAJ Lesnik Oberstein has received significant research funding from The Brain Foundation of The Netherlands and The Netherlands Organisation for Health Research and Development (ZonMw). 15 These authors contributed equally to this work. Received 23 December 2014; revised 20 March 2015; accepted 3 April 2015
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
2
effects of the APOE ε2 and ε4 alleles on WMH burden in the context of pure SVD. This was performed through a large multicenter study in patients with CADASIL. MATERIALS AND METHODS Subjects A total of 552 CADASIL patients were assessed at seven university hospitals across Europe as part of the CADASIL genome-wide association study consortium.6 Nineteen patients were excluded for technical problems during genotyping, another 20 patients because of missing MRI data, and 25 patients because of missing demographic/clinical data. Thus, the final sample size consisted of 488 patients (Supplementary Table). The aims and recruitment criteria of the consortium have been reported in detail.6 In brief, all subjects underwent a detailed clinical interview after physical examination, blood draw for isolation of DNA, and structural MRI. The study was conducted in adherence to the Good Clinical Practice guidelines and to the Declaration of Helsinki including all revisions. Each site obtained ethical approval from the respective local institutional review board. All participants of the study gave written informed consent for study participation, MRI scanning, and use of DNA. Genotyping APOE carrier status was determined centrally using an iPlex assay (Sequenom, San Diego, CA, USA). Data from a genome-wide set of 466 single nucleotide polymorphism (SNP) markers from the Affymetrix 6.0 platform (Affymetrix UK, High Wycombe, UK)6 was used to control for variations in the genetic background of the study cohort (see Statistics section). Because genotyping of the large genome-wide set of SNP markers was less robust than the targeted genotyping of APOE, SNP data were available for only 407 out of the 488 subjects. None of these SNP markers was in linkage disequilibrium (r240.1) with the APOE genotype. Magnetic resonance imaging data processing MR scans were acquired at each site, using scanners operating at field strengths between 0.5 and 3 T. White matter hyperintensities were determined as previously described.6,9 In short, lesions were segmented on FLAIR images by a semiautomated method and then checked and corrected by trained raters. The interrater reliability for this procedure was excellent with an intraclass correlation coefficient of 0.996. The WMH volumes were normalized to the intracranial cavity volume by dividing the WMH volume by the intracranial cavity volume, which had been derived from T2- or proton density-weighted images. Given the skewed distribution of the normalized WMH volumes, a square-root transformation was performed. The normalized and transformed WMH volumes are referred to as nWMHV. A fit of the normal
Table 1.
distribution curve onto the distribution of nWMHV is shown in the Supplementary Figure 1. Statistical analysis Demographics, vascular risk factors, and WMH volume were compared across APOE groups, using analysis of variance for continuous variables and Χ2-tests of independence for bimodal variables. Multiple linear regression analysis was used to determine the effect of APOE status on nWMHV. Specifically, we examined whether carriers of the APOE ε2 or ε4 allele showed a difference in nWMHV compared with noncarriers. Because of the small number of subjects homozygous for ε4 (n = 8) and the absence of subjects homozygous for ε2 we limited our analyses to ε2- and ε4-carrier status. We encoded ε2 carriers (i.e., ε2/ε3) and ε4 carriers (i.e., ε3/ε4 and ε4/ε4) as dummy variables, with noncarriers (i.e., ε3/ε3) being the reference group. We performed a stepwise regression analysis with forward selection. nWMHV was entered as dependent variable and APOE carrier status, age, sex, hypertension, hypercholesterolemia, diabetes, and past/current smoker were entered as independent variables. To minimize the possibility that any APOE effects are driven by differences in genetic background, we conducted an additional regression analysis controlling for genetic background variation in a subset of 407 patients in which the genome-wide set of SNP markers were available.6 To reduce dimensionality, identity-bystate–based principal component analysis was performed on the SNP markers.6 On the basis of visual inspection of a plot showing the decrease of the Eigenvalues across the rank-ordered principal components, the first five principal components were selected. Further, we ran a second stepwise regression with forward selection, entering these five principal components in addition to the aforementioned independent variables. RESULTS Demographic information and vascular risk factors are reported in Table 1. The most frequent NOTCH3 mutations were R1006C and R182C (each n = 30). Stepwise multiple regression analysis resulted in a model including APOE carrier status and age as significant independent variables. APOE ε2 carriers showed increased nWMHV compared with the reference genotype ε3/ε3 (β = 0.031, t = 2.4, P = 0.0185). In contrast, there was no significant association between ε4-carrier status and nWMHV (β = 0.017, t = 1.7, P = 0.0935), compared with the same reference group. The sizes of the effects of individual APOE alleles on nWMHV are shown in Figure 1 (model 1). The covariate age was associated with higher nWMHV (β = 0.005, t = 15.1, P o 2E − 16). When further controlling for genetic background variation, the results remained essentially unchanged, i.e., nWMHV was significantly increased in ε2 carriers (β = 0.028, t = 2.0, P = 0.0496) but not in ε4 carriers (β = 0.008, t = 0.8, P = 0.4428;
Demographics, vascular risk factors, and WMH volume in the different APOE groups
APOE status Age, median (IQR) (years) Sex, male, n (%) Hypertension, n (%) Hypercholesterolemia, n (%) Diabetes melitus, n (%) Past/current smoker, n (%) WMHV, median (IQR) (%)
ε3/ε3, n = 350 (71.7%) 51 169 87 157 14 87 5.2
(44–60) (48.3) (24.9) (44.9) (4.0) (26.4) (2.6–8.5)
ε2 carriers, n = 46 (9.4%) 48 21 9 10 0 10 5.7
(42–54) (45.7) (19.6) (21.7) (0) (22.2) (2.3–12.7)
ε4 carriers, n = 92 (18.9%) 52 38 18 45 4 26 5.7
(41–59) (41.3) (19.6) (48.9) (4.3) (29.9) (2.6–10.1)
P value 0.606 0.484 0.461 0.006 0.374 0.628 0.233
Abbreviations: APOE, apolipoprotein E; IQR, interquartile range; WMHV, white matter hyperintensity volume (as percentage of intracranial cavity volume).
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
© 2015 ISCBFM
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
3
Figure 1. Effect sizes (i.e., regression coefficients) of the different APOE alleles on normalized white matter hyperintensities volumes, derived from the regression model with genotype ε3/ε3 being the reference and controlling for age (model 1) plus genetic background (model 2) as covariates. The bars indicate 95% confidence intervals of the regression coefficients.
Figure 1, model 2). Box plots for nWMHV after regressing out age and genetic background are provided in Supplementary Figure 2. DISCUSSION Our results demonstrate an association between the presence of the APOE ε2 allele and higher WMHV in a well-defined population of subjects with inherited pure SVD. In contrast, the APOE ε4 genotype was not associated with WMHV. Our finding on APOE ε2 is consistent with results from a recent meta-analysis in sporadic age-related SVD, showing an association between APOE ε2 allele and increased WMHV.3 For APOE ε4, we found a nonsignificant trend for an effect on WMHV, which disappeared when controlling for genetic background. These results on APOE ε4 are in agreement with previous studies in CADASIL that reported no effect of APOE ε4 on white matter damage.7,8 However, we cannot exclude an association between homozygous APOE ε4 genotype on WMHV3 as this could not be assessed in the current study owing to the low number of subjects with this genotype (n = 8). In summary, our results show that APOE ε2 allele contributes to the severity of white matter injury in subject with pure SVD. We can only speculate on the mechanisms by which the ε2 allele increases WMHV in this nonamyloidogenic form of SVD. The APOE ε4 and ε2 allele both have been found to increase the risk of cerebral amyloid angiopathy through an increased deposition of vascular amyloid beta (Aβ),10 which may in turn relate to impaired vascular drainage of Aβ.11 Although vascular Aβ deposition has no role in CADASIL, APOE ε2 may be associated with impaired drainage of other proteins leading to microvascular damage.12 Furthermore, the ε2 allele is associated with specific vessel pathologies in CAA, such as vessel dilation, fibrinoid necrosis, microaneurysms and double barreling.13 Some of these vasculopathic changes have also been observed in CADASIL. Hence, APOE ε2 might likewise influence these vascular pathologies in CADASIL. Another possibility is that the presence of the APOE ε2 allele is associated with increased WMH through its influence on inflammation. In population-based studies, higher plasma levels of lipoprotein-associated phospholipase A2, a marker of vascular inflammation, were associated with more severe SVD as measured by cerebral microbleeds and WMH.14,15 Although the association between lipoprotein-associated phospholipase A2 and higher © 2015 ISCBFM
prevalence of cerebral microbleeds was dependent on the presence of an APOE ε2 allele,14 such a modulatory effect of APOE ε 2 on the effect of inflammation on WMH remains to be tested in future studies. Specific strengths of the current study include the large cohort of carefully phenotyped patients with genetically defined SVD. Also, the WMH segmentations and genotyping were both performed on a central platform, thus minimizing the betweencenter variability. Our statistical results were robust, with the ε2 effect remaining significant even when controlling for variations in genetic background. Controlling for the genetic background is important, since subtle genetic variation within or between centers could introduce a major bias in the results. A limiting factor is that our study focused on WMH, but we did not assess other imaging features of SVD such as cerebral microbleeds and lacunes. Also, data regarding clinical disease severity (e.g., dementia, onset of cognitive impairment, and stroke) have not been collected. The consortium was established with the specific goal to assess genetic modifiers of WMH in a unique sample of patients with genetically determined SVD. The selection of patients was performed accordingly and imaging modalities needed to assess other SVD marker are sparsely available within the sample. DISCLOSURE/CONFLICT OF INTEREST The authors declare no conflict of interest.
DISCLAIMER Dr Pantoni is member of the editorial boards of Cerebrovascular Diseases and Acta Neurologica. REFERENCES 1 Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 2013; 12: 822–838. 2 Carmelli D, DeCarli C, Swan GE, Jack LM, Reed T, Wolf PA et al. Evidence for genetic variance in white matter hyperintensity volume in normal elderly male twins. Stroke 1998; 29: 1177–1181. 3 Schilling S, DeStefano AL, Sachdev PS, Choi SH, Mather KA, DeCarli CD et al. APOE genotype and MRI markers of cerebrovascular disease: systematic review and meta-analysis. Neurology 2013; 81: 292–300. 4 Brickman AM, Zahodne LB, Guzman VA, Narkhede A, Meier IB, Griffith EY et al. Reconsidering harbingers of dementia: progression of parietal lobe white matter hyperintensities predicts Alzheimer's disease incidence. Neurobiol Aging 2014; 36: 27–32. 5 Gurol ME, Viswanathan A, Gidicsin C, Hedden T, Martinez-Ramirez S, Dumas A et al. Cerebral amyloid angiopathy burden associated with leukoaraiosis: a positron emission tomography/magnetic resonance imaging study. Ann Neurol 2013; 73: 529–536. 6 Opherk C, Gonik M, Duering M, Malik R, Jouvent E, Herve D et al. Genome-wide genotyping demonstrates a polygenic risk score associated with white matter hyperintensity volume in CADASIL. Stroke 2014; 45: 968–972. 7 Singhal S, Bevan S, Barrick T, Rich P, Markus HS. The influence of genetic and cardiovascular risk factors on the CADASIL phenotype. Brain 2004; 127: 2031–2038. 8 van den Boom R, Lesnick Oberstein SAJ, van den Berg-Huysmans AA, Ferrari MD, van Buchem MA, Haan J. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy: structural MR imaging ghanges and apolipoprotein E genotype. Am J Neuroradiol 2006; 27: 359–362. 9 Duering M, Zieren N, Herve D, Jouvent E, Reyes S, Peters N et al. Strategic role of frontal white matter tracts in vascular cognitive impairment: a voxel-based lesion-symptom mapping study in CADASIL. Brain 2011; 134: 2366–2375. 10 Nelson PT, Pious NM, Jicha GA, Wilcock DM, Fardo DW, Estus S et al. APOEepsilon2 and APOE-epsilon4 correlate with increased amyloid accumulation in cerebral vasculature. J Neuropathol Exp Neurol 2013; 72: 708–715.
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
APOE ε2 increases lesion burden in pure SVD B Gesierich et al
4 11 Thal DR, Larionov S, Abramowski D, Wiederhold KH, Van Dooren T, Yamaguchi H et al. Occurrence and co-localization of amyloid beta-protein and apolipoprotein E in perivascular drainage channels of wild-type and APP-transgenic mice. Neurobiol Aging 2007; 28: 1221–1230. 12 Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO. Review: cerebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum of protein elimination failure angiopathies (PEFA) in neurodegenerative disease with a focus on therapy. Neuropathol Appl Neurobiol 2013; 39: 593–611.
13 Greenberg SM, Vonsattel JP, Segal AZ, Chiu RI, Clatworthy AE, Liao A et al. Association of apolipoprotein E epsilon2 and vasculopathy in cerebral amyloid angiopathy. Neurology 1998; 50: 961–965. 14 Romero JR, Preis SR, Beiser AS, DeCarli C, Lee DY, Viswanathan A et al. Lipoprotein phospholipase A2 and cerebral microbleeds in the Framingham Heart Study. Stroke 2012; 43: 3091–3094. 15 Wright CB, Moon Y, Paik MC, Brown TR, Rabbani L, Yoshita M et al. Inflammatory biomarkers of vascular risk as correlates of leukoariosis. Stroke 2009; 40: 3466–3471.
Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http://www.nature.com/jcbfm)
Journal of Cerebral Blood Flow & Metabolism (2015), 1 – 4
© 2015 ISCBFM
