Skip main navigation

Genetic Study of PHACTR1 and Fibromuscular Dysplasia, Meta-Analysis and Effects on Clinical Features of Patients

The ARCADIA-POL Study
Originally publishedhttps://doi.org/10.1161/HYPERTENSIONAHA.120.14793Hypertension. 2020;76:e4–e7

Fibromuscular dysplasia (FMD) is a common cause of renovascular hypertension in middle-aged women lacking cardiovascular risk factors. FMD is an intriguing group of noninflammatory arterial deformations, which results in stenosis, and in some cases, aneurysms and dissections of medium-sized arteries. Mostly asymptomatic, the clinical manifestations of FMD depends on the arterial bed affected and is usually diagnosed in the context of resistant hypertension or after a stroke, particularly cervical artery dissection.1 FMD is increasingly identified after the occurrence of spontaneous coronary artery dissection, an underdiagnosed cause of acute myocardial infarction.1 We have previously established a complex genetic basis for FMD and reported a first genetic susceptibility locus on chromosome 6 involving rs9349379, a common genetic locus within the phosphatase and active regulatory 1 gene (PHACTR1).2 This pleiotropic genetic locus is also known to associate with several neurovascular diseases,3 atherosclerotic acute myocardial infarction,4 and spontaneous coronary artery dissection.5

Here, we aimed to replicate the association between PHACTR1 and FMD in a Polish case-control study and assess the effect of this locus on FMD, hypertension, and cardiac features among cases.

Patients were recruited through the ARCADIA-POL (Assessment of Renal and Cervical Artery Dysplasia - POLAND) study, a nation-wide Polish registry for FMD. Lesions in at least one vascular bed were confirmed in all patients using whole-body angio-computed tomography as previously described.6 We analyzed 129 patients for whom we excluded atherosclerotic stenosis and syndromes where FMD-like lesions are often observed (eg, Ehlers Danlos, Loeys-Dietz). Patients were 80.6% women, with a mean age at inclusion of 45.9±14.1 years, and a mean age at FMD diagnosis of 42.7±14.2 years. One hundred thirteen patients (87.6%) were hypertensive. The mean age at the diagnosis of hypertension was 36.5±14.3 years, and FMD was diagnosed on average 6.2±3.6 years later. Other most frequent FMD-related symptoms were headaches (38.8%), dizziness (37.2%), pulsatile tinnitus (15.5%), and stroke (9.3%). Among analyzed patients with FMD, 41.1% were current or ex-smokers. In this group of patients, FMD was identified in renal arteries in 109 (84.5%) patients, as well as in cerebrovascular, visceral, and lower extremities arteries in 34 (26.4%), 20 (15.5%), and 15 (11.6%) patients, respectively. Arterial dissection(s) or aneurysms in various vascular beds were present in 16 (12.4%) and in 31 (24%) patients, respectively. One hundred and ten (85.3%) patients presented multifocal FMD lesions. Multisite FMD was found in 42 patients (32.6%) and was defined as the presence of FMD stenosis, FMD-related aneurysms, FMD-related dissections, FMD-related S-shaped cervical arteries in 2 or more of the vascular beds. The median number of vascular beds affected was 1 (interquartile range 1–2, minimum =1, maximum =5). We extracted DNA using Chemagic (Perkin Elmer) and performed genotyping using direct sequencing (Applied Biosystems 3730XL) in cases and controls. We compared DNAs sequences from patients to 298 controls, a randomly ascertained sample from the WOBASZ II (Multicentre National Population Health Examination Survey) study, a population-based Polish cohort.7

We confirmed the association between rs9349379 genotypes and FMD using logistic regression. We found 56% increased risk for FMD per A allele (odds ratio [OR]=1.56 [95% CI, 1.14–2.13]; P=5.5×10−3). We provide the estimated effect of this variant on FMD risk through an updated meta-analysis of the previously published data2 and the current case-control study using METAL.8 We confirmed the genome-wide significant association with FMD (OR=1.40 [95% CI, 1.27–1.55]; P=1.8×10−11) that now involves 1283 FMD cases and 4193 controls.

PHACTR1 robust association with FMD is intriguing, as this variant also associates with risk for several neurovascular and cardiovascular diseases involving diverse risk factors. We sought to characterize the effects of this variant among FMD patients of the ARCADIA-POL who have benefited from extensive clinical phenotyping related to FMD, hypertension and cardiac morphology and function. To gain power, we analyzed AG and AA genotypes carriers under the same category (dominant model) using unconditional logistic regression for binary traits and linear regression or Mann-Whitney test for continuous traits, adjusted for sex and age when relevant (Tables 1 through 5).

Table 1. Association of rs9349379 Variant with Demographic Traits Among FMD Patients

DemographicsSexBMISmokingOral Contraception (Only Women)Age FMD DiagnosisAge HTN DiagnosisYears HTN to FMD Diagnosis*
WomenMenNkg/m2YesNoYesNonYearsnYearsnYears
GG (ref)14 (13%)6 (24%)2023.74±3.358 (15%)12 (16%)5 (14%)9 (14%)1942.05±13.951636.75±12.50130.62±0.77
AG+AA90 (87%)19 (76%)10224.57±3.7245 (85%)61 (84%)30 (86%)55 (86%)10143.05±15.048736.06±14.04776.22±7.65
Effect (AG+AA vs GG)2.03 (0.69–5.96)1.081.15 (0.42–3.13)1.06 (0.33–3.80)−0.74−3.325.61
P value0.200.180.780.920.570.170.01

Binary variables are expressed as the number of patients (percentage) and quantitative variables as mean±SD. P value for the association between the clinical trait and rs9349379 under the genetic dominant model adjusted for sex and age, except for the sex analysis. Effect for binary dependent variables odds ratio (95% CI) from unconditional logistic regression is reported and for quantitative dependent variable the β from linear regression is reported. BMI indicates body mass index; HTN, hypertension; and FMD, fibromuscular dysplasia.

*For this trait, we report Δ between genotype groups means and Mann-Whitney (nonparametric test) P values; this analysis was not adjusted for sex and age.

Among the 129 Polish patients, we found equal distribution of rs9349379 genotypes in ever and nonsmokers (P=0.78) and for oral contraception treatment among women (P=0.92). No differences were found between mean age of FMD and hypertension diagnosis. We report a nonsignificant younger age at hypertension diagnosis (β=−3.32 years, P=0.17). Interestingly, patients with AG+AA genotypes tend to associate with an average of 5.61 years difference between hypertension and FMD diagnosis (P=0.01), which may reflect a milder hypertension phenotype, and thus discourage vascular imaging exploration overtime that is needed to reveal FMD lesions. Patients carrying the AG+AA genotypes tend to be more likely to present with multifocal versus focal FMD subtype (OR=4.82, P=0.15). We found a trend for a decreased odd among AG+AA genotype group for multisite lesions (OR=0.41, P=0.08), especially at both renal arteries (OR=0.37, P=0.05), supporting a potentially milder presentation. Systolic and diastolic blood pressure values were similar by genotype, as was hypertension status among patients. We found the risk allele for FMD to tend to associate with decreased levels of creatinine (β=−6.32, P=0.053), although renal function estimated by glomerular filtration rate did not differ by genotype (P=0.69).

FMD stenosis may evolve in aneurysm and dissections in the affected arteries. In our cohort, 13 patients presented aneurysm in the renal and 11 at the cerebrovascular bed. We found no influence of the genotype on the risk of aneurysm in patients globally (P=0.64), or specifically at the renal (P=0.91) or cerebrovascular (P=0.20) beds. However, patients carrying the risk genotypes AG+AA had a 3.81-fold risk for dissection, although this association did not reach statistical significance (P=0.22), given the limited size of the cohort and the low number of dissections observed. Our study is probably lacking statistical power to confirm the previously reported association where the A allele associated with a higher risk for cervical dissection3 and coronary dissection.5 Larger samples are needed to address the specific association of rs9349379 with dissection events in FMD.

Patients with FMD from the ARCADIA-POL study underwent a series of echocardiographic measurements, including intraventricular septum thickness, posterior wall thickness, left ventricular mass, which were all comparable according to genotypes as were ejection fraction and global longitudinal strain (Tables 1 through 5). We noted nonsignificant trends for lower E-wave length during diastole (β=−5.067, P=0.17) but not A-wave, resulting in a lower E- to A-wave ratio (β=−0.146, P=0.09) in the AG+AA group.

Our study is limited by the small size of the cohort of patients and the low power to robustly estimate the genetic associations with a large number of the traits and lack of significance after multiple testing corrections. Except for the global association with FMD, additional exploration of PHACTR1 on hypertension and cardiac traits is needed in larger cohorts of patients with FMD to validate our results.

In summary, we provide confirmatory association between PHACTR1 locus and FMD in this first genetic study in a Polish case-control study and an updated global effect through the largest existing genetic meta-analysis for this vascular disease. Additional risk loci are expected to be revealed through ongoing genome-wide association studies. We found evidence for an association with delayed diagnosis for FMD among patients who present with HTN. These patients could be at higher risk for FMD misdiagnosis and less risk to present multisite lesions. We overall exclude a major genetic effect of this locus on blood pressure and cardiac morphology parameters among patients with FMD.

Table 2. Association of rs9349379 Variant with Traits Related to FMD Presentation Among Patients

FMD PresentationFMD SubtypeFMD Lesions LocationRenal Artery Affected2 Renal Arteries AffectedCerebrovascular ArteryDigestive Artery
FocalMultifocalMultisiteSingle-SiteYesNoYesNoYesNoYesNo
GG (ref)1 (5%)19 (17%)10 (24%)10 (11%)17 (16%)3 (15%)13 (23%)7 (10%)7 (21%)13 (14%)5 (25%)15 (14%)
AG+AA18 (95%)91 (83%)32 (76%)77 (89%)92 (84%)17 (85%)44 (77%)65 (90%)27 (79%)82 (86%)15 (75%)94 (86%)
Effect (AG+AA vs GG)4.82 (0.57–40.50)0.41 (0.15–1.10)0.87 (0.22–3.39)0.37 (0.14–1.01)0.64 (0.23–1.81)0.49 (0.15–1.57)
P value0.150.080.840.0530.400.23

Variables are expressed as the number of patients (percentage). P value for the association between the clinical trait and rs9349379 under the genetic dominant model adjusted for sex and age. Effect for dependent variables odds ratio (95% CI) from unconditional logistic regression is reported. FMD indicates fibromuscular dysplasia.

Table 3. Association of rs9349379 Variant with Traits Related to Blood Pressure Among FMD Patients

Blood Pressure TraitsSBPDBPHypertensionAnti-HTN DrugsCreatinine LevelsGFR (CDK-Epi)
Nmm HgNmm HgYesNoYesNonμmol/LnUnit
GG (ref)19132.18±24.361982.50±10.3117 (15%)3 (19%)13 (14%)5 (18%)1880.97±15.181788.63±19.33
AG+AA98134.85±18.199883.08±10.2496 (85%)13 (81%)79 (86%)23 (82%)10872.56±13.4410290.64±19.06
Effect (AG+AA vs GG)2.4420.5371.51 (0.38–6.05)1.42 (0.45–4.51)−6.322.02
P value0.620.840.560.550.0530.69

Binary variables are expressed as the number of patients (percentage) and quantitative variables as mean±SD. P value for the association between the clinical trait and rs9349379 under the genetic dominant model adjusted for sex and age, except for GFR (CDK-Epi). Effect for binary dependent variables odds ratio (95% CI) from unconditional logistic regression is reported and for quantitative dependent variable the β from linear regression is reported. DBP indicates diastolic blood pressure; FMD, fibromuscular dysplasia; GFR (CDK-Epi), glomerular filtration rate using the Chronic Kidney Disease - Epidemiology Collaboration formula; HTN, hypertension; and SBP, systolic blood pressure.

Table 4. Association of rs9349379 Variant with Vascular Complications Among FMD Patients

Vascular ComplicationsAneurysmRenal Artery AneurysmCerebrovascular AneurysmDissectionResistive Index (Left Renal Artery)Resistive Index (Right Renal Artery)History of Stent Implantation
YesNoYesNoYesNoYesNonUnitnUnitYesNo
GG (ref)5 (16%)15 (15%)2 (15%)18 (16%)3 (27%)17 (14%)1 (6%)19 (17%)190.55±0.07180.56±0.081 (9%)19 (16%)
AG+AA26 (84%)83 (85%)11 (85%)98 (84%)8 (73%)101 (86%)15 (94%)94 (83%)1050.58±0.091050.58±0.0810 (91%)99 (84%)
Effect (AG+AA vs GG)0.76 (0.2–2.4)0.92 (0.18–4.65)0.39 (0.09–1.66)3.81 (0.4–32.0)0.0170.0052.55 (0.3–22.8)
P value0.640.910.200.220.370.780.40

Variables are expressed as mean±SD. P value for the association between the clinical trait and rs9349379 under the genetic dominant model. Effect for binary dependent variables odds ratio (95% CI) from unconditional logistic regression is reported and for quantitative dependent variable the β from linear regression is reported. FMD indicates fibromuscular dyspasia.

Table 5. Association of rs9349379 Variant with Echocardiographic Parameters Among FMD Patients

Echocardiographic ParametersIntraventricular Septum Thickness (Diastole)Posterior Wall Thickness (Diastole)Left Ventricular Mass IndexEjection Fraction (S)Global Longitudinal Strain (S)E-Wave (Diastolic)A-Wave (Diastolic)E- to A-Wave Ratio (Diastolic)*
NUnitNUnitnUnitnUnitnUnitnUnitnUnitNUnit
GG (ref)199.47±1.72189.86±1.821583.1±17.01868.0±5.915−18.4±1.72080.3±17.52062.8±16.1201.3±0.3
AG+AA999.66±1.75999.92±1.749490.9±27.78969.1±6.770−19.3±3.19476.0±14.59468.0±20.2921.2±0.4
Effect (AG+AA vs GG)0.2530.1928.4280.720−0.858−5.0672.845−0.146
P value0.560.660.260.670.280.170.480.09

The number of patients available for analysis is indicated for each trait. Binary variables are expressed as the number of patients (percentage) and quantitative variables as mean±SD. Effect for binary dependent variables odds ratio (95% CI) from unconditional logistic regression is reported and for quantitative dependent variable the β from linear regression is reported. Analysis was adjusted for sex and age except for sex and glomerular filtration rate, which included these variables in the calculation formula. P value for the association between the clinical trait and rs9349379 under the genetic dominant model adjusted for sex and age. Effect for quantitative dependent variable the β from linear regression is reported.

*For this trait, we report Δ between genotype groups means and Mann-Whitney (nonparametric test) P values and the analysis was not adjusted for sex and age.

Footnotes

*These authors are joint first authors.

†These authors are joint last authors.

Correspondence to Nabila Bouatia-Naji, PARCC Inserm, 56 rue Leblanc, F75015, Paris, France. Email

References

  • 1. Gornik HL, Persu A, Adlam D, Aparicio LS, Azizi M, Boulanger M, Bruno RM, De Leeuw P, Fendrikova-Mahlay N, Froehlich J, et al; Working Group ‘Hypertension and the Kidney’ of the European Society of Hypertension (ESH) and the Society for Vascular Medicine (SVM). First international consensus on the diagnosis and management of fibromuscular dysplasia.J Hypertens. 2019; 37:229–252. doi: 10.1097/HJH.0000000000002019CrossrefMedlineGoogle Scholar
  • 2. Kiando SR, Tucker NR, Castro-Vega LJ, Katz A, D’Escamard V, Tréard C, Fraher D, Albuisson J, Kadian-Dodov D, Ye Z, et al. PHACTR1 is a genetic susceptibility locus for fibromuscular dysplasia supporting its complex genetic pattern of inheritance.PLoS Genet. 2016; 12:e1006367. doi: 10.1371/journal.pgen.1006367CrossrefMedlineGoogle Scholar
  • 3. Debette S, Kamatani Y, Metso TM, Kloss M, Chauhan G, Engelter ST, Pezzini A, Thijs V, Markus HS, Dichgans M, et al; International Stroke Genetics Consortium; CADISP Group. Common variation in PHACTR1 is associated with susceptibility to cervical artery dissection.Nat Genet. 2015; 47:78–83. doi: 10.1038/ng.3154CrossrefMedlineGoogle Scholar
  • 4. Nikpay M, Goel A, Won HH, Hall LM, Willenborg C, Kanoni S, Saleheen D, Kyriakou T, Nelson CP, Hopewell JC, et al. A comprehensive 1,000 genomes-based genome-wide association meta-analysis of coronary artery disease.Nat Genet. 2015; 47:1121–1130. doi: 10.1038/ng.3396CrossrefMedlineGoogle Scholar
  • 5. Adlam D, Olson TM, Combaret N, Kovacic JC, Iismaa SE, Al-Hussaini A, O’Byrne MM, Bouajila S, Georges A, Mishra K, et al; DISCO Consortium; CARDIoGRAMPlusC4D Study Group. Association of the PHACTR1/EDN1 genetic locus with spontaneous coronary artery dissection.J Am Coll Cardiol. 2019; 73:58–66. doi: 10.1016/j.jacc.2018.09.085CrossrefMedlineGoogle Scholar
  • 6. Warchoł-Celińska E, Pieluszczak K, Pappaccogli M, Soplińska A, Prejbisz A, Dobrowolski P, Klisiewicz A, Kądziela J, Falkowski A, Śmigielski W, et al. Dissecting visceral fibromuscular dysplasia reveals a new vascular phenotype of the disease: a report from the ARCADIA-POL study.J Hypertens. 2020; 38:737–744. doi: 10.1097/HJH.0000000000002327CrossrefMedlineGoogle Scholar
  • 7. Drygas W, Niklas AA, Piwońska A, Piotrowski W, Flotyńska A, Kwaśniewska M, Nadrowski P, Puch-Walczak A, Szafraniec K, Bielecki W, et al. Multi-centre national population health examination survey (WOBASZ II study): assumptions, methods, and implementation.Kardiol Pol. 2016; 74:681–690. doi: 10.5603/KP.a2015.0235CrossrefMedlineGoogle Scholar
  • 8. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans.Bioinformatics. 2010; 26:2190–2191. doi: 10.1093/bioinformatics/btq340CrossrefMedlineGoogle Scholar

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.