Losartan in Marfan syndrome. Alexander Willem den Hartog

Transcription

1 Losartan in Marfan syndrome Alexander Willem den Hartog

2

3 Losartan in Marfan syndrome Alexander Willem den Hartog

4 Losartan in Marfan syndrome Thesis, University of Amsterdam, the Netherlands Author: Alexander W. den Hartog ISBN: Cover: istock.com Layout: Optima Grafische Communicatie, Rotterdam, the Netherlands ( Printed by: Optima Grafische Communicatie, Rotterdam, the Netherlands ( Financial support by the Dutch Heart Foundation for publication of this thesis is gratefully acknowledged. The research described in this thesis was supported by a grant of the Dutch Heart Foundation (DHF 2008B115) Additional financial support for printing this thesis was kindly provided by: Academical Medical Center - University of Amsterdam, OLVG Unit Cardiologie, OLVG Cardiologie Research B.V., Contactgroep Marfan Nederland, The Marfan Foundation, Sanofi Nederland B.V., Bayer B.V., Boehringer Ingelheim B.V., Chipsoft, Servier Nederland Farma B.V., Astellas Pharma B.V.

5 LOSARTAN IN MARFAN SYNDROME ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. ir. K.I.J. Maex ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op dinsdag 9 mei 2017, te uur door Alexander Willem den Hartog geboren te Haarlem

6 Promotiecommissie: Promotor: Prof. dr. B.J.M. Mulder Universiteit van Amsterdam Prof. dr. A.H. Zwinderman Universiteit van Amsterdam Copromotores: Dr. M. Groenink Universiteit van Amsterdam Dr. V. de Waard Universiteit van Amsterdam Overige leden: Dr. W. Stooker UvA en OLVG Prof. dr. R. Balm Universiteit van Amsterdam Dr. D.R. Koolbergen AMC-UvA Dr. R.N. Planken AMC-UvA Prof. dr. J. De Backer Universiteit Gent Prof. dr. A.C. van Rossum Vrije Universiteit Amsterdam Prof. dr. M.M.A.M. Mannens Universiteit van Amsterdam Faculteit der Geneeskunde

7 The problem with the world is that the intelligent people are full of doubts while the stupid ones are full of confidence Charles Bukowksi

8

9 table of contents Introduction and outline of the thesis 9 Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Current and future pharmacological treatment strategies with regard to aortic disease in Marfan syndrome. Expert Opinion on Pharmacotherapy 2012 Apr;13(5): Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome. International Journal of Cardiology 2013 Oct;168(3): Aortic disease in patients with Marfan syndrome: aortic volume assessment for surveillance. Radiology 2013 Nov;269(2):370-7 The risk for type B aortic dissection in Marfan syndrome. Journal of the American College of Cardiology 2015 Jan;65(3): Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. European Heart Journal 2013 Dec;34(45): The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations. Netherlands Heart Journal 2016 Nov;24(11): Chapter 7 Summary 131 Future Perspectives 135 Nederlandse Samenvatting 139 Publication List 146 Author Contributions 147 PhD Portfolio 149 Curriculum Vitae 151 Dankwoord 153

10

11 Introduction and outline of the thesis

12

13 Introduction and outline of the thesis 11 introduction Marfan syndrome (MFS) is a heritable disorder of the connective tissue with a prevalence of 1 per 5000 and is known to affect multiple organ system, including lungs, eyes, skeleton en the entire cardiovascular system.(1,2) MFS is caused by a deficiency of the protein fibrillin-1 that is imposed by heterozygous mutations in FBN1 and excessive transforming growth factor-β (TGF-β) expression. The MFS diagnosis is established by the revised Ghent criteria, which includes the family history, aortic dilatation and dissection, ectopia lentis and a list of systemic features.(3) The most common cardiovascular manifestation of MFS include aortic (root) dilatation (Figure 1), which may lead to aortic dissection (Stanford type A or B; Figure 2), thickening and prolapse of the atrioventricular valves with or without valve regurgitation and ventricular dysfunction. figure 1 Aortic root aneurysm In order to prevent aortic dissection current MFS guidelines advocate prophylactic replacement of the aortic root when the diameter exceeds mm.(4) Furthermore, treatment with β-blockers is recommended in order to lower blood pressure, chronotropy and the inotropic state of the heart in order to diminish the cyclic repetitive forces exerted on the aortic wall and thereby reducing the onset and progression of aortic dilatation and possibly preventing aortic dissection.(5) However, MFS studies assessing the effect of β-blockers on aortic dilatation are conflicting and this pharmacological strategy focusses on prevention of aortic dilatation without targeting the underlying cause of the progressive aortic disease in MFS. The pathological substrate for aortic disease in MFS is referred as aortic media degeneration and consists of vascular smooth-muscle

14 12 Introduction and outline of the thesis cell apoptosis, adventitial inflammation, interstitial mucoid deposits and degeneration of elastin fibrils along with accumulation of collagen and proteoglycans. Aortic media degeneration is caused by a deficiency of normal fibrillin-1 protein that leads to intense elastolysis and increased TGF-β expression. Angiotensin II contributes to aortic medial degeneration by promoting endothelial dysfunction, hypertrophy of the aortic wall and inducing pro-inflammatory responses. The effect of angiotensin II is mediated by angiotensin II receptor type 1, which increases TGF-β expression by stimulating autocrine production in cardiac fibroblasts and Smad expression. Overexpression of TGF-β in fibrillin-1-deficient mice leads to MFS phenotypical features and treatment of these mice with losartan has showed to prevent aortic media degeneration and dilatation.(2) In this thesis the effect of losartan on both aortic dilatation rate and ventricular function in MFS patients is addressed. Furthermore, the occurrence of type B aortic dissection in MFS is investigated. figure 2 Classification of aortic dissection localization

15 Introduction and outline of the thesis 13 outline of this thesis chapter 1 provides an overview of the current pharmacological treatment options used to prevent or delay aortic complications in adults with MFS. Furthermore, this chapter discusses insights on the biochemical pathways leading to aortic disease, new targets for pharmacological therapy and ongoing pharmacological trials as well, which can be viewed as a general introduction into the field of MFS. MFS is caused by mutations in the gene that encodes for fibrillin-1 protein, a structural protein that normally binds TGF-β. MFS mice studies have shown an association between aortic diameter and TGF-β level.(2) chapter 2 addresses the role of TGF-β as a biomarker for aortic dilatation, aortic root growth and dissection in adult MFS patients. A major clinical manifestation of MFS is type A aortic dissection. With the introduction of prophylactic aortic root surgery at an aortic root diameter of cm, type A aortic dissection can be prevented. This aggressive surgical approach and medicinal therapy has improved life expectancy of MFS patients. With this improved survival an increased incidence of type B aortic dissections has been reported. Therefore, current guidelines advocate regular visualization of the total aorta. However, type B aortic dissection, especially after surgical aortic replacement, shows to be very difficult to foresee by measuring aortic diameters and most likely does not reflect the extent of aortic disease.(6) Volume measurement of the aorta could reflect the extent of aortic disease more sensitively. In chapter 3 the reproducibility of aortic volume estimates as a novel method to measure the extent of aortic disease is assessed. Furthermore, we will test its use serially to determine global aortic dilatation in MFS patients. In chapter 4 we investigate the association of type B aortic dissection in patients with MFS with various clinical parameters, including genetic analysis, aortic dimensions, and aortic elastic properties. Furthermore, we establish a risk model to predict the occurrence of type B aortic dissection in adult patients with MFS. In the last two chapters we focus on the effect of pharmacological treatment with losartan to prevent cardiovascular complications in MFS. In chapter 5 a multicentre, open-label randomized control trial with blinded endpoints is presented. Here the potential role of losartan treatment in reducing aortic complications in MFS is investigated. In chapter 6 the effect of losartan on ventricular volumes and function in genetically classified subgroups of asymptomatic MFS patients is addressed. Finally, we complete this thesis with a summary, future perspectives and a Dutch summary.

16 14 Introduction and outline of the thesis reference list 1. Judge DP, Dietz HC. Marfan s syndrome. Lancet Lond Engl Dec 3;366(9501): Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science Apr 7;312( ): Loeys BL, Dietz HC, Braverman AC, Callewaert BL, de BJ, Devereux RB, et al. The revised Ghent nosology for the Marfan syndrome. JMedGenet Jul;47( ): ESC Guidelines on the diagnosis and treatment of aortic diseases: Document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adultthe Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC). Eur Heart J Nov 1;35(41): Hartog AW, Franken R, Zwinderman AH, Groenink M, Mulder BJ. Current and future pharmacological treatment strategies with regard to aortic disease in Marfan syndrome. Expert Opin Pharmacother Apr;13(5): Engelfriet P. The spectrum of adult congenital heart disease in Europe: morbidity and mortality in a 5 year follow-up period: The Euro Heart Survey on adult congenital heart disease. Eur Heart J Sep 29;26(21):

17

18

19 Chapter 1 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome Alexander W. den Hartog, Romy Franken, Aeilko H. Zwinderman, Maarten Groenink, Barbara J.M. Mulder Adapted from: Expert Opinion on Pharmacotherapy 2012 Apr;13(5):

20 18 Chapter 1 abstract introduction Marfan syndrome is a multisystemic connective tissue disorder caused mainly by mutations in the fibrillin-1 gene. The entire cardiovascular system is affected in patients with Marfan syndrome. Aortic root dilatation, aortic valve regurgitation or - the most feared and life-threatening symptom - aortic root dissection are the most common manifestations. Therapeutic strategies, such as prophylactic aortic root surgery and pharmacological therapy, focus on the prevention of aortic dissection. Currently, the standard medicinal treatments targeting aortic dilatation and dissection consist of agents generally used to lower blood pressure and/or the inotropic state of the heart. By these means, the cyclic repetitive forces exerted on the aortic wall are diminished and thus the onset of aortic dilatation is potentially prevented. Although these pharmacological agents may offer some benefit in reduction of aortic aneurysm expansion rate, they do not target the underlying cause of the progressive aortic degradation. areas covered This review discusses the effectiveness of frequently prescribed medications used to prevent and delay aortic complications in Marfan syndrome. New insights on the biochemical pathways leading to aortic disease are also discussed to highlight new targets for pharmacological therapy. expert opinion Recent insights in the transforming growth factor beta signaling pathway and inflammatory mechanisms in a well-established mouse model of Marfan syndrome, have led to studies exploring new pharmacological treatment strategies with doxycycline, statins and angiotensin II receptor blockers. Pharmacological therapy is focused more on prevention than on delay of aortic wall pathology in Marfan syndrome. Of the new pharmacological treatment strategies targeting aortic pathology in Marfan syndrome, angiotensin receptor type 1 blockers are promising candidates, with several clinical trials currently ongoing.

21 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 19 introduction Marfan syndrome (MFS) is a multi-systemic disorder of the connective tissue with a prevalence of 1 per 5000 individuals and is mainly caused by mutations in the fibrillin-1 (FBN1) gene.(1) MFS is inherited in an autosomal dominant manner. Currently, more than 1700 genetic mutations have been described in patients with MFS and over 90% of these mutations are unique to an individual or family.(2) To date, no genotype-phenotype correlations or predictors for MFS severity have been found on the basis of the location or type of FBN1 mutations. Even within families, in which affected individuals share the same mutation, extreme variable phenotype expression has been shown.(3) The diagnosis of MFS as established by the Ghent Nosology of 1996 is based on family history and a combination of clinical manifestations in various organs such as the skeletal, cardiovascular and ocular system (Table 1).(4) The criteria are divided into major and minor criteria based on their frequency in patients with MFS and in the general population in the presence of other conditions. As several features of MFS are age-dependent, the 1996 Ghent Nosology had a relative low sensitivity. Therefore, the criteria have been revised and published as the Revised Ghent Nosology of The revised criteria give more weight to aortic root dilatation, FBN1 mutation, other relevant genes and ectopia lentis. All other features have become minor criteria and are described by a systemic score (Table 1).(5;6) According to the Revised Ghent Nosology, the diagnosis of MFS will be established in the presence of: (1) aortic root enlargement (Z score 2) and ectopia lentis, (2) aortic root enlargement (Z score 2) and a FBN1 mutation, (3) aortic root enlargement (Z score 2) and systemic features of MFS resulting in a systemic score 7 points, or (4) ectopia lentis and an FBN1 mutation that is known to be associated with aortic disease in MFS. 1 cardiovascular manifestations of marfan syndrome Cardiovascular manifestations of MFS comprise aortic root dilatation (Figure 1) with or without aortic valve regurgitation (this is present in 80 % of patients with MFS),(7;8) thickening and prolapse of the atrioventricular valves with or without regurgitation and aortic dissection.(9) Less common cardiovascular manifestations of MFS include arrhythmias,(10-13) systolic (14;15) and diastolic dysfunction (16) as well as calcification of aortic- and atrioventricular valves.(17;18) In addition to these features, aortic elastic behaviour is generally affected in MFS of which distensibility and pulse wave velocity are in vivo measurable derivatives.(19-21) Mitral valve disease has become a minor criterion in the Revised Ghent Nosology due to its lack of specificity for MFS, rather than lack of importance. The prevalence of mitral valve disease in MFS is highly divergent

22 20 Chapter 1 table 1 Revised Ghent criteria for diagnosis of Marfan syndrome and related conditions criteria for diagnosis of marfan syndrome In the absence of family history of Marfan syndrome: 1. Ao (Z 2) AND EL = MFS* 2. Ao (Z 2) AND FBN1 = MFS 3. Ao (Z 2) AND Syst ( 7 points) = MFS* 4. EL AND FBN1 associated with Ao = MFS EL with or without Syst AND with an FBN1 not previously been associated with Ao or no FBN1 = ELS Ao (Z < 2) AND Syst ( 5 points) with at least one skeletal feature, without EL = MASS MVP AND Ao (Z < 2) AND Syst (< 5) without EL = MVPS In the presence of family history of Marfan syndrome: 5. EL AND FH = MFS 6. Syst ( 7 points) AND FH = MFS* 7. Ao (Z 2 above age of 20 years, 3 below age of 20 years) AND FH = MFS* scoring of systemic features Wrist AND thumb sign 3 (wrist OR thumb sign 1) Pectus carinatum deformity 2 (pectus excavatum or chest asymmetry 1) Hindfoot deformity 2 (plain pes planus 1) Pneumothorax 2 Dural ectasia 2 Protrusio acetabuli 2 Reduced US/LS AND increased arm/height AND no severe scoliosis 1 Scoliosis or thoracolumbar kyphosis 1 Reduced elbow extension 1 Facial features (3/5) 1 (dolichocephaly, enophthalmos, downslanting palpebral fissures, malar hypoplasia, retrognathia) Skin striae 1 Myopia > 3 diopters 1 Mitral valve prolapse (all types) 1 Maximum total: 20 points; score 7 indicates systemic involvement *Caveat: without discriminating features of Shprintzen-Goldberg syndrome, Loeys-Dietz syndrome or vascular EDS syndrome. If present then TGFBR1/2 testing, collagen biochemistry, COL3A1 testing. Other conditions/genes will follow with time. Abbreviations: Ao, aortic diameter at the sinuses of Valsalva above indicated Z-score or aortic root dissection; Z, Z-score; EL, ectopia lentis; MFS, Marfan syndrome; FBN1, fibrilline-1 mutation; Syst, systemic score; ELS, ectopia lentis syndrome; MASS, myopia, mitral valve prolapse, borderline (Z<2) aortic root dilatation, striae, skeletal findings phenotype; MVP, mitral valve prolapse; MVPS, mitral valve prolapse syndrome; FH, family history; US/LS, upper segment/lower segment ratio

23 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 21 (28-77%),(17;22;23) which may be explained by small cohorts and by differences in the criteria used to assign the presence of mitral valve disease. In MFS, one or both of the mitral leaflets are frequently prolonged, often accompanied by myxomatous thickening and billowing. Neonatal MFS is often complicated by severe mitral valve prolapse and regurgitation, which is a major contributor to poor prognosis and is the leading cause of morbidity.(24;25) Morse et al. reported in 22 infants with MFS that mitral valve prolapse was present in 82% in the first 3 months of life.(24) Investigators of the Universal MFS database reported among 320 children with MFS at age <18 years a prevalence of mitral valve prolapse of 73% among those classified as having neonatal MFS (diagnosed prior to age 4 weeks), 62% among those diagnosed between age 4 weeks and 10 years, and 59% among those diagnosed between 10 and 18 years of age.(26) 1 figure 1 Echocardiographic long axis image of a Marfan patient with aortic root dilatation survival Since the introduction of surgical techniques for aortic root repair, the life expectancy of patients with MFS has nearly normalized.(27) Over the years, improvements in surgical techniques and post-operative care have resulted in a continuous lowering of the threshold for prophylactic aortic root repair, which is currently advised at an aortic root diameter of 50 mm. Aortic valve repair is getting increasingly favoured over prosthetic

24 22 Chapter 1 valve replacement for a variety of lesions causing aortic regurgitation.(28-34) Advantages of repair over replacement include a lower incidence of thrombo-embolic events, the avoidance of long-term anticoagulation, and reduced risks of endocarditis.(35) However, the durability of aortic repair procedures has been a matter of great concern.(28) le Polain de Waroux et al. showed that intraoperative transesophageal echocardiography can be used to identify MFS patients undergoing aortic root repair who are at increased risk for developing recurrent severe aortic regurgitation.(36) These parameters include the persistence of a residual aortic root jet, a coaptation length <4 mm, a coaptation level occurring below the level of the aortic annulus and an enlarged aortic annulus. (36) In an European survey, events and diameters involving the distal aorta were greater in patients who had undergone aortic root surgery as compared to those who had not. (37) A possible explanation for this expansion of the distal aorta after prophylactic aortic surgery may be a more advanced stage of MFS in these patients, aortic root replacement has an impact on the distal aorta as a result of altered hemodynamic factors and altered wall mechanics or due to clamping of the aorta during the procedure.(38) histology and biochemical Pathways The pathological substrate for aortic disease in MFS consists of vascular smooth muscle cell apoptosis, adventitial inflammation, interstitial mucoid desposits as well as degeneration of elastin fibrils along with accumulation of collagen and proteoglycans in the aortic wall, often referred to as aortic medial degeneration.(39-41) Fibrillin-1 constitutes a major component of microfibrils in the extracellular matrix and has a role in maintaining the structural integrity and cell development in arteries. This complex glycoprotein confers elasticity to the arterial wall and facilitates the vessel to retain its original geometry after expansion due to the repetitive cyclic hemodynamic forces by linking vascular smooth muscle cells to adjacent elastin fibrils.(42) Furthermore, fibrillin-1-rich-microfibrils have a role in modulating cell development and homeostasis through regulation of transforming growth factor beta (TGF-β) and matrix metalloproteinases (MMPs).(43) A deficiency of normal fibrillin-1 in tissues of MFS patients may lead to ineffective tissue repair due to intense elastolysis and increased TGF-β expression. The ultimate consequence of this cascade is aortic medial degeneration development. Overexpression of TGF-β in fibrillin-1 deficient mice leads to phenotypical features also present in patients with MFS, such as myxomatous changes with mitral valve pathology, aortic dilatation and emphysematous pulmonary changes.(44-46) In addition to fibrillin-1 abnormalities, angiotensin II may contribute to the development of aortic medial degeneration by promoting endothelial dysfunction, hypertrophy of the aortic wall (47-49) and inducing pro-inflammatory responses.(50;51) Generally,

25 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 23 the effect of angiotensin II is mediated by two receptors. Angiotensin II receptor type 1 (AT1R) increases TGF-β expression by stimulating the autocrine production in cardiac fibroblasts (52) and Smad expression.(53) Angiotensin II receptor type 2 (AT2R) signaling may induce vascular smooth muscle cell apoptosis and is expressed significantly higher in the aorta of patients with MFS (Figure 2).(41;54) AT2R signalling has the capacity to attenuate both Smad-dependent and mitogen activated protein kinase TGF-β signalling cascades, most notably the extracellular signal regulated kinase (ERK)(55;56). However, in some cell types and tissues, AT2R signaling has shown to oppose AT1R mediated enhancement of TGF-β expression.(54;57) Thus, AT1R stimulation drives ERK1/2 activation, whereas AT2R signalling can both augment and inhibit aneurysm formation in preclinical models. A third pathway contributing to the development of aortic medial degeneration is MMP overexpression. Abdominal aortic aneurysms studies led to the hypothesis that increased production of proteinases in the aneurismal wall is involved,(58;59) especially the MMP-2 and MMP-9.(60-63) MMPs represent a family of zinc dependent, calcium activated, locally produced endopeptidases that are expressed weakly in healthy aortic tissue. Fragmentation of FBN1 leads to overexpression of MMPs, in particular MMP-2 and MMP-9, in patients with MFS and are secreted by mesenchymal cells as well as aneurysm infiltrating macrophages.(64-67) Elevated expression of MMPs results in extensive degeneration of elastic fibres, endothelium dysfunction, reduction of smooth muscle contractility, has show to degraded arterial elastin and correlates with aortic aneurysm diameter. (68-71) Aortic medial degeneration does not exclusively occur in MFS. It is rather considered to be a degenerative end-stage of a multitude of conditions including the normal aging process, bicuspid aortic valve disease, atherosclerosis and hypertension.(72) 1 Pharmacological treatment of aortic disease in marfan syndrome Pharmacotherapy is routinely undertaken to slow down aortic dilatation in patients with MFS, thereby preventing or postponing the need for surgical intervention. Improved understanding of the histological abnormalities causing MFS can potentially identify other pharmacotherapeutic agents that target the pathogenic basis for MFS, ultimately preventing cardiovascular complications, which would improve life expectancy and quality of life.

26 24 Chapter 1 beta-adrenoceptor antagonist (β-blockers) Currently, β-blockers are regarded as the standard medical therapy in order to delay and prevent aortic dilatation and aortic dissection in patients with MFS. The rationale of the use of β-blockers to attenuate aortic pathology in MFS is based on reduction of stress on the proximal aorta.(figure 2) Although there is no supporting data, it is thought that relieving wall stress from the aortic wall is achieved by decreasing inotropy (diminishing rate of volume and pressure change in the ascending aorta, dp/dt), chronotropy and blood pressure. The use of β-blockers began in 1960, for the management of aortic aneurysms prone to dissection in patients with malignant hypertension.(73;74) In 1970 Simpson et al. demonstrated that turkeys, prone to spontaneous aortic dissection, had a reduced incidence of aortic dissection when propranolol was added to their feed.(75) Based on these results, Halpern et al. described in 1971 that propranolol might reduce the risk of aortic dilatation and dissection in six adult patients with MFS. (76) However, in 1977 after a mean 6 year follow up of 25 patients with MFS the same group concluded, that propranolol did not protect against aortic dilatation or dissection, after 20% of patients died of aortic dissection.(77) In 1994, Shores et al. published a open-label randomized trial of propranolol in 70 adolescents and adults with MFS. (78) They demonstrated no significant difference between the placebo and propranolol treated groups with regard to cardiovascular endpoints, such as aortic root regurgitation, dissection and rupture. However, over a follow up period of 10 years the rate of aortic dilatation in the propranolol treated group was significantly lower than in the placebo group (p<0.001).(78) To date, several studies have been published regarding the effectiveness of β-blockers on cardiovascular endpoints in MFS.(27;79-82) These studies show conflicting results and none were prospective, double blinded randomized trials. A large (n = 417) retrospective study by Silverman et al. noticed no beneficial effect of either propranolol, atenolol, metoprolol, nadolol or a combination of β-blockers on aortic root dilatation. However, this study was limited by a lack of understanding of the cause of death and only data on aortic surgery was used as a clinical endpoint. (27) In a prospective study, Salim et al. demonstrated a beneficial effect of atenolol and propranolol on aortic root dilatation in 113 patients with MFS,(82) while Tahernia et al. noticed in a small prospective study (n=6) that there was no progression of aortic root dilatation in patients with MFS receiving β-blockers.(81) In a meta-analysis Gersony et al. concluded that there is no evidence for the use of β-blockers to prevent cardiovascular endpoints in MFS.(83) In MFS, the aorta dilates gradually and aortic distensibility decreases due to degeneration of the aortic media.(84;85) The effect of β-blockers on these aortic properties has been assessed thoroughly and has also shown conflicting results. If β-blockers benefit the aorta through its negative chronotropic and inotropic actions, the elastic aortic wall properties should improve after administration of a β-blocker. A study by Groenink et al. reported an increase in aortic distensibility and a decrease in

27 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 25 pulse wave velocity in the groups treated with a β-blocker (either metoprolol or atenolol), thereby potentially reducing the risk of aortic dilatation and aortic dissection.(20) However, several studies found no beneficial effect of β-blockers on reduction of aortic stiffness in MFS patients, including 2 studies of patients undergoing cardiac catheterization showed an increased stiffness and decreased distensibility after administration of propranolol.(86-91) A small study by Nollen et al. found no difference in aortic diameter or distensibility as measured by magnetic resonance imaging between patients with MFS treated with β-blockers and those not treated with β-blockers.(87) In patients after entire aortic replacement, the use of atenolol or labetalol did not decrease aortic pressure.(88) Haouzi et al. and Rios et al. described a heterogeneous response in aortic root elastic properties of MFS patients treated with either metoprolol or atenolol.(84;91) Pharmacological intervention routinely used in adult patients with MFS is also applicable in children and pregnant women with MFS. However, there are conflicting data regarding the therapeutic potential of β-blockers to treat aortic root dilatation in children with MFS.(92;93) In a retrospective study with 63 paediatric MFS patients Selamet et al. found no significant difference in rate of aortic root dilatation between the β-blocker treated group and the control group after a mean 6 year follow up.(92) Conversely, in a retrospective study with 115 paediatric MFS patients, Ladouceur et al. showed a reduction in aortic root expansion rate in the β-blocker treated group as compared to the control group.(93) During pregnancy hormonal alterations in pregnant MFS patients lead to histological changes of the aortic wall, such as fragmentation of the reticulum fibres, diminishing amount of mucopolysaccharides and loss of the normal corrugation of elastic fibres.(94) Pregnancy has shown to increase the risk of aortic dissection in women with MFS, probably due to both hemodynamic and hormonal alterations.(95-102) However, pregnancy seems to be relatively safe up to an aortic root diameter of 45 mm.(97) Propranolol and metoprolol are regarded as safe and non-teratogenic for the unborn child and are regard as the standard pharmacological therapy in order to prevent aortic complications.(98;103;104) However, no trial has ever been performed assessing the efficacy of β-blockers on aortic root dilatation in women with MFS during pregnancy. A potential undesirable effect of β-blockers is increasing systemic vascular resistance, which could increase the amplitude of reflected pulse waves and decrease total aortic compliance. Yin et al. demonstrated an increase in magnitude of wave reflection in patients with MFS as compared to a healthy control group and this wave reflection is further increased by propranolol.(90) Patients with MFS may develop aneurysm and dissection of the aorta beyond the sinuses of Valsalva.( ) Yetman et al. demonstrated that distal peripheral arterial aneurysms are present in one third of adults with MFS. (105) β-blocker therapy is often continued after aortic aneurysms surgery, although its efficacy has never been demonstrated.(37;108) On the contrary, increased rigidity due 1

28 26 Chapter 1 to the artificial grafts and elevated vascular resistance during β-adrenergic blockade may even augment wave reflections in the aorta, thereby increasing shear stress in the main side branches.( ) Meijboom et al. doubted the use of atenolol, a commonly prescribed β-blocker, in patients with MFS after entire aortic replacement. This doubt stemmed from the result of a small prospective study showing that pulse pressure, systolic pressure, wave speed and wave reflections were significantly increased in patients with MFS after entire aortic replacement as compared to controls.(88) In these operated MFS patients, Meijboom et al. found no decrease in aortic pressure and a trend towards increase in wave reflection when treated with β-blockers. doxycycline Doxycycline, a tetracycline-class antibiotic, is known to inhibit MMP secretion. The inhibitory effect of doxycycline on MMP expression may be attributed to the following mechanisms: binding to zinc/calcium at the catalytic sites on MMPs, inhibition of pro-mmp activity, promoting proteolysis of pro-mmps and decreased levels of MMP mrna (Figure 2).(112;113) Several aortic aneurysm studies in a mouse model of MFS and an aneurysm rat model assessing the effects of doxycycline noted a stabilizing effect on the aortic wall by preventing aortic elastic fibre fragmentation and normalize endothelium dependent relaxation.(67;88; ) In addition, these studies reported that doxycycline may inhibit MMP-2 and MMP-9 production,(112;113) TGF-β expression and Smad2 signalling.(67;129;130) Chung et al. noted that doxycycline is more effective in preventing aortic aneurysm formation than atenolol in a well established mouse model of MFS and reported that the aneurysm suppressing properties of doxycycline are in part attributable to its inhibitory effects op MMP-2 and MMP-9 expression.(118) Whereas in animal studies doxycycline appears to be beneficial in preventing aneurysm disease, the evidence in human studies is, according to the meta analysis by Dodd et al., conflicting.(131) Therefore, further studies are required in order to examine the efficacy of doxycycline in preventing aneurysm development in patients with MFS. statin Statin treatment is one of the cornerstone therapies in coronary artery disease by reducing the progression of atherosclerosis. In addition to their lipoprotein-reducing properties, statin treatment may have a beneficial effect on aneurysm formation by decreasing MMP expression and have a positive effect on aortic distensibility trough an increase in endothelial nitric oxide synthase (enos) and a decrease in reactive oxygen species (ROS). In an in- vivo model of diabetic glucose intolerance, statins have been shown to reduce expression of TGF-β, as well as diminish oxidative stress by blocking the effects of ROS on aneurysms.(132;133) Statins achieve these results through the suppression of the NADH/NADPH oxidase system (Figure 2).(134) To date, statins have shown to reduce

29 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 27 aortic aneurysms in an observational study,(135) although another study noticed no beneficial effect of statin treatment on aortic aneurysm expansion rate.(136) A recent study by McLoughlin et al. assessing the effect of pravastatine in a mouse model of MFS demonstrated that pravastatine treatment attenuated aortic root dilatation to the same degree as losartan treatment but preserves elastin content within the aortic wall significantly better than losartan.(137) Although this study is only a proof of principle and substantial effort is still needed to elucidate the precise mechanism involved, it is a potential beneficial treatment strategy. Molecular mechanisms of aneurysm formation and the effects of different medications. Angiotensin promotes aneurysm formation through angiotensin 1 (AT1) receptors. Increased angiotensin II causes an increase in reactive oxygen species (ROS) through the NADH/NADPH system, which in turn increases cyclophilin A and matrix metalloproteinase (MMP) levels. This promotes inflammatory reaction and subsequent medial degeneration, leading to aneurysm formation. Fibrillin gene mutations cause enhanced transforming growth factor (TGF)-β signaling. This results in cellular proliferation and matrix degradation probably through signaling via the psmad2 system. Angiotensin receptor blockers are thought to inhibit the above pathways via inhibition of the AT1 receptors. Angiotensin-converting enzyme inhibitors (ACE-I) block angiotensin II. Statins block the NADH/NADPH system; tetracyclines and macrolides reduce MMP activity. β-blockers reduce shear stress on the vessel.(138) 1 figure 2 Pathways leading to aortic medial degeneration in MFS and the effects of different medications

30 28 Chapter 1 renin-angiotensin-aldosteron system inhibitors Angiotensin II and its two receptors (AT1R and AT2R) have shown to play an important role in the development of aortic medial degeneration in MFS. Inhibition of the reninangiotensin-aldosteron system (RAAS) cascade may therefore prevent aortic medial degeneration and thereby preclude aneurysm formation. angiotensin converting enzyme inhibitors Angiotensin converting enzyme inhibitors (ACE-I) such as enalapril and perindopril have been utilized in numerous cardiovascular diseases including heart failure, hypertension, coronary artery disease, dilated cardiomyopathy and post myocardial infarction.(139) The beneficial effects of ACE-I in MFS may be attributed to different mechanisms. First, ACE-I may stabilize the aortic wall by down regulating angiotensin II and hence limit signalling through both the AT1R and the AT2R pathway (Figure 2). Furthermore, ACE-I may reduce TGF-β and MMP expression, reduce Smad2 phosphorylation, diminish angiotensin II mediated vascular smooth muscle cell apoptosis and block the effects of hyperhomocysteinemia.(140) In addition, the use of ACE-I augments collagen syntheses, increases elastin en fibrillin deposition, reduces vascular inflammation, decreases pulse wave velocity and improves aortic compliances.(141) Other beneficial effects of ACE-I include inhibition of kininase, which results in increased bradykinin that leads in turn to bradykinin-mediated improvement in aortic elastic tone by vasodilatation.(142;143) Two clinical trials have assessed the effect of ACE-I as compared to β-blockers in patients with MFS. A non-randomized, controlled, prospective trial investigated the effect of enalapril, atenolol or propanolol in 58 patients with MFS.(144) After a follow-up of 3 years, patients treated with enalapril showed an improvement of aortic distensibility and a reduction in the aortic stiffness relative to patients treated with β-blockers. These hemodynamic changes were associated with a smaller increase in aortic root diameter and fewer clinical endpoints during follow up as compared with either β-blockers or no medical therapy. Another randomized, prospective trial assessed the effects of perindopril in 17 patients with MFS.(141) A significant improvement in the ACE-I treated group (n=10) as compared to the control group (n=7) was noticed in relation to arterial compliance, central and peripheral pulse wave velocities as well as aortic root diameters. In addition, perindopril reduced TGF-β expression after 24 weeks of therapy as compared to placebo (p=0.01).(141) These two clinical trials suggest that ACE-I may offer therapeutic benefit in MFS patients by inhibiting angiotensin II expression. However, a large prospective randomized controlled trial is needed to confirm these results. ACE-I have shown to prevent aortic expansion and rupture in animal models of aortic aneurysms.(145) In addition, ACE-I may lower the incidence of distal aortic disease and may reduce the risk of abdominal aortic aneurysm rupture in humans.(105;146) Although the short-term survival of type B acute aortic dissection is excellent, 40 % of

31 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 29 patients experience aortic events, such as rupture, recurrent dissection and aneurismal dilatation.( ) A retrospective study by Takeshita et al. showed that treatment with ACE-I was associated with a significantly reduced risk of dissection related aortic events in patients who survived aortic type B dissection.(152) 1 angiotensin ii type 1 receptor blocker AT1R blockers, such as losartan and irbesartan, are used to treat hypertension and other cardiovascular diseases when ACE-Is are not tolerated due to bradykinin-mediated side effects like dry hacking cough. AT1R blockers have drawn interest for treatment of MFS since Habashi et al. demonstrated in a MFS mouse model with FBN1 C1039G/+ mice, the most common class of FBN1 mutation in people with MFS, that aortic medial degeneration development and TGF-β signalling were diminished by losartan (Figure 2).(45) In addition, Habashi et al. noticed that aortic dilatation and absolute aortic root diameter were indistinguishable between losartan treated transgenic mice and wild-type mice and losartan seemed to prevent elastic fibre fragmentation, which is crucial in preserving aortic wall architecture. Following this study, Brooke et al. assessed in a retrospective unblinded trial the effect of an AT1R blocker in 18 paediatric MFS patients relative to 65 paediatric MFS patients treated with β-blockers.(153) AT1R blocker therapy, either losartan (n=17) or irbesartan, was prescribed in cases of rapid aortic root dilatation despite other medical therapy (n=11), aortic root diameter approaching the threshold for surgical intervention (n=6) or unacceptable adverse events (n=1) in conventional pharmacology. Brooke et al. noticed a significant decrease in aortic root growth rate and absolute sinotubular junction diameter in each member of the group using AT1R blockers. Unlike studies with the frequently used FBN1 C1039G/+ mouse models, Nistala et al. showed a less pronounced effect of losartan on aortic wall degradation in FBN1 mgr/mgr mice, whereas the effect on aneurysm expansion were comparable to treatment with beta-blockers.(154) Nistala et al. have suggested that inhibition of aortic medial degeneration in severe MFS may require higher losartan dosage. Habashi et al. demonstrated that dual blockade of AT1R mediated ERK1/2 activation and AT2R mediated ERK1/2 inhibition, as occurs either with the use of ACE-I in FBN1 C1039G/+ mice or the use of losartan in AT2KO:FBN1 C1039G/+ mice, results in no net change in ERK1/2 activation status.(155). This indicates that, in the presence of AT1 receptor blockade, ongoing AT2 receptor signalling is required for the attenuation of ERK cascade. Losartan reduces ERK1/2 activation by blocking both the AT1R cascade while simultaneously shunting angiotensin II signalling through the AT2 receptor. The lack of effect of enalapril lack on ERK is probably attributable to the loss of AT2 receptor signalling. The differential effect of enalapril treatment suggests that TGF-β mediated ERK1/2 activation is the predominant driver of aneurysm progression in MFS.

32 30 Chapter 1 Yang et al. assessed the effect of losartan combined with doxycycline in a mouse model of MFS and demonstrated that neither losartan nor doxycycline alone could restore vascular integrity and cell function.(128) In addition, they noticed that the combination of losartan and doxycycline reduced the size of aneurysm throughout the life spam more effective than single drug therapy. It did this by improving elastic fibre organization, normalizing vasomotor function, inhibiting MMP-2 and MMP-9 expression and TGF-β activation.(128) Currently, multiple randomized controlled trials have been initiated in humans with MFS examining the efficacy of losartan on aortic root growth (Table 2). calcium channel blockers Between 10 and 20% of MFS patients do not tolerate β-blockers or these are contraindicated because of asthma bronchiale, diabetes mellitus or depression. Although prescribed in patients with MFS, clinical data to support the use of calcium antagonists in MFS are lacking. To date, only one small prospective observational study assessed the effect of calcium antagonists in children with MFS.(79) In this study a calcium antagonist, most often verapamil, was prescribed to 6 patients with MFS. Due to the limited number of patients receiving calcium antagonists, these patients were combined with those receiving β-blockers and defined as the treatment group. Overall, the treatmentgroup was found to have a slower rate of aortic root growth when compared to an older patient population. nitrate The endothelium constitutively releases vasodilators, including nitric oxide (NO), to regulate smooth muscle contractility and thus arteriolar tone.(156) Chung et al. demonstrated that the endothelium-function and signalling of NO production is impaired in the thoracic aorta of mice with MFS. This endothelium dysfunction is most likely due to the downregulation of calcium-dependent endothelial NO synthase.(157) However, the endothelium-independent vasodilatory response in a small study of 10 MFS patients with sodium nitroprusside, a nitrate, was similar to that of 10 healthy age- and sexmatched controls, indicating normal sensitivity of smooth muscle cells to NO in MFS. (158) To assess the effectiveness of nitroprusside in MFS on prevention of aortic dilatation, a randomised, placebo controlled trial is currently running with 30 MFS patients. (159) conclusion The cardiovascular manifestations of MFS include aortic root dilatation, aortic valve regurgitation and aortic dissection. Aortic dissection is the main cause of morbidity

33 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 31 and mortality in MFS and therapeutic strategies, such as prophylactic aortic root surgery and pharmacological therapy, focus primarily on preventing the onset of aortic dissection. Despite limited and conflicting evidence, β-blockers are regarded as the standard medical therapy in order to delay and prevent cardiovascular complications in MFS, including in children, in pregnant women as well as after aortic surgery. Although there are no supporting data, the rationale of the use of β-blockers to attenuate aortic pathology in MFS is based on reduction of stress on the proximal aorta by decreasing inotropy, chronotropy and blood pressure.(figure 2) So far, the only open-label randomized clinical trial assessing the effect of propranolol demonstrated a reduction in the rate of aortic root dilatation. However, propranolol did not show any beneficial effect on clinical cardiovascular endpoints such as aortic dissection or aortic rupture. To date, several studies have been published regarding the effectiveness of β-blockers on cardiovascular endpoints in MFS.(27;79-82) These studies show conflicting results and none were prospective, double blinded randomized trials. Evidence for beneficial effect of calcium channel blockers on aortic root dilatation in MFS is still lacking. The only trial ever assessing calcium channel blocker efficacy was a small study whereby the treatment group was defined as patients receiving calcium antagonists combined with patients receiving β-blockers. A randomized trial assessing the effectiveness of nitroprusside on reducing arterial stiffness in patients with MFS is ongoing, after few studies demonstrated an impairment of the NO production in patients with MFS. Although β-blockers are thought to be beneficial by reducing the hemodynamic stress on the aorta, they do not prevent the development of aortic medial degeneration, the pathological substrate for aortic disease in MFS. Recent publications have demonstrated several pathways contributing to the development of aortic medial degeneration such as fibrillin-1 abnormalities, along with overexpression of angiotensin II, TGF-β and MMPs (Figure 2). Improved understanding of the histological abnormalities causing MFS has identified several potential pharmacotherapies directly targeting aortic medial degeneration. These therapies include doxycycline, statins, ACE-I and AT1R blockers.(figure 2) In a mouse model of MFS, doxycycline has shown to be more effective in preventing aortic aneurysm formation than atenolol by preserving elastic fibre integrity, normalizing vasomotor function, suppressing TGF-β activation and inhibiting MMP-2 and 9 expression. However, the efficacy of doxycycline in preventing aneurysms in human MFS studies is conflicting. Another potential beneficial therapy, statin, has demonstrated to slow down aortic root dilatation rate by preserving elastin with in the aortic wall, decreasing MMP expression and reducing the expression of TGF-β in a MFS mouse model. A recent study in a mouse model of MFS demonstrated that pravastatine treatment attenuated aortic root dilatation to the same degree as losartan treatment. Although few clinical trials have been conducted assessing the effects of ACE-I in MFS patients, two studies have sug- 1

34 32 Chapter 1 table 2 Ongoing trials assessing the effect of losartan on aortic properties in Marfan syndrome start date follow up design age range (years) target no. (months) cases clinical endpoints tool boston children s hospital, Pediatric heart network lacro et al. January DB, RCT; losartan vs atenolol 0, Change in AoR diameter Ultrasound national taiwan university hospital (losartan) wu et al. February 2007 unknown OF, RCT; losartan and atenolol vs propanolol 1 44 Change in Aor diameter Ultrasound brigham and women s hospital October DB, RCT; losartan creager et al. vs atenolol Aortic biophysical properties Ultrasound + arterial tonometry heart and stroke foundation of canada January DB, RCT; losartan vs atenolol Aortic biophysical properties Ultrasound sandor et al. academic medical center (compare) February OB, RCT; losartan vs no losartan Change in AoR diameter Ultrasound + MRI radonic et al. Policlinico st. matteo hospital July OB, RCT losartan gambarin et al. vs nebivolol or combined Change in AoR diameter Ultrasound

35 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 33 table 2 Ongoing trials assessing the effect of losartan on aortic properties in Marfan syndrome (continued) clinical endpoints tool start date follow up design age range (years) target no. (months) cases Ultrasound Change in AoR diameter September DB, RCT losartan vs placebo hospital bichat Paris (marfan sartan) detaint et al. Ultrasound + MRI AoR diameter at any level, CA diameter June DB, RCT losartan vs placebo ghent hospital (ghent marfan trial) moberg et al. Ultrasound Change in AoR diameter September DB, RCT irbesartan vs placebo royal brompton harefield foundation (the aims trial) mullen et al. Ultrasound + MRI Change in AoR, CA diameter October DB, RCT losartan vs atenolol hospital universitario vall d hebron forteza et al. AoR Aortic root; CA carotid artery; DB double-blind; echo: echocardiography; MRI magnetic resonance imaging; OB open label, blinded endpoints; OF open factorial; RCT randomized controlled trial; 1

36 34 Chapter 1 gested that patients treated with an ACE-I had a reduction in the aortic stiffness index, a slower rate of aortic growth and reduced mean arterial pressure. In addition, ACE-I has shown to reduced the risk of dissection related events in patients who survived aortic type B dissection. Another promising pharmacological treatment strategy targeting aortic medial degeneration in MFS are AT1R blockers. These agents have been shown to diminish TGF-β signalling, restore aortic wall architecture in FBN1 C1039G/+ mice. In the study by Habashi et al., aortic dilatation and absolute aortic root diameter were indistinguishable between losartan treated transgenic and wild-type mice. Another study by Habashi et al., demonstrated that TGF-β mediated ERK1/2 activation is the predominant driver of aneurysm progression in MFS and is reduced by losartan, which blocks the AT1R cascade while simultaneously shunting angiotensin II signalling through the AT2R. In contrast to earlier studies, Nistala et al. showed a less pronounced effect of losartan on aortic wall degradation in FBN1 mgr/mgr mice, and suggested that inhibition of aortic medial degeneration in severe MFS might require higher losartan dosage. Yang et al. showed that the combination of losartan and doxycycline reduced the size of aneurysm more effective than single drug therapy bin a mouse model. It did this by improving elastic fiber organization, normalizing vasomotor function, inhibiting MMP-2, MMP-9 and TGF-β activation. In a retrospective, unblinded trial, Brook et al. noticed a significant decrease in aortic root growth rate and absolute sinotubular junction diameter in 18 paediatric patients using AT1R blockers. These findings resulted in the initiation of multicentre randomized controlled clinical trials assessing the effect of losartan on aortic root growth in humans with MFS (Table 2). expert opinion The evidence for the efficacy of β-blockers on aortic root dilatation is controversial and limited. In order to assess the efficacy of β-blockers, a randomized blinded trial with a comparison of β-blockers and ACE-I and/or AT2 antagonists in MFS patients is needed. However, this has become impossible as it is considered unethical to discontinue β-blockers in patients with MFS following several studies that demonstrated a beneficial effect of β-blockers on aortic root dilatation rate. Although β-blockers may offer some benefit in reduction of aortic root dilatation rate in MFS patients, they do not prevent the development of aortic medial degeneration, the underlying pathological cause for aortic disease in MFS. Recent studies in MFS have demonstrated that a deficiency of normal fibrillin-1, due to FBN1 mutations, may lead to increased TGF-β and MMP expression, thereby contributing to the development of aortic medial degeneration. The elucida-

37 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome 35 tion of these pathways has resulted in new potential pharmacotherapies targeting the pathogenic basis for MFS for example doxycycline, statins, ACE-I and AT1R blockers. Doxycycline and pravastatine have demonstrated to slow aortic root dilatation rate and to preserve elastin within the aortic wall by hindering the development of aortic medial degeneration in a MFS mouse model. Although both have shown to be beneficial in a mouse model of MFS, translation of these agents in preventing aortic root dilatation in patients with MFS requires prospective randomized clinical trials. The AT1R blockers are another promising new pharmacological treatment strategy in MFS targeting aortic medial degeneration. Several studies with the AT1R blocker losartan have shown that medical treatment strategies targeting the specific pathogenic factors may prevent aortic root dilatation by diminishing TGF-β activity. However, a major limitation of the study by Habashi et al., investigating efficacy of losartan, was that prevention of aortic medial degeneration development was demonstrated in a mouse model with one specific FBN1 mutation. In addition, Nistala et al. showed a less pronounced effect of losartan on aortic wall degradation in FBN1 mgr/mgr mice, a more aggressive mutation, and have suggested that inhibition of aortic medial degeneration in severe MFS may require higher losartan dosage. In humans more than 1700 different FBN1 mutations have been reported and it is therefore uncertain whether losartan therapy has the same effect in patients with different mutations causing MFS. It is possible that different types of FBN1 mutations cause different molecular and biophysical outcomes, thus MFS patients using losartan might be divided into a responder and non-responder group. These findings have resulted in the initiation of multiple multicentre randomized controlled trials assessing the effects of AT1R blocker on aortic root growth in humans with MFS (Table 2). The primary goal of these trials will be to determine whether aortic root dilatation and its associated complications are reduced in patients with MFS receiving AT1R blocker, relative to those receiving β-blockers therapy or placebo. Secondary aims of these trials are to answer questions such as what is the long term efficacy of AT1R blocker, what are the possible side effects and to assess the potential of AT1R blocker to reverse aortic damage. These trials are crucial in assessing the efficacy and safety of AT1R blocker in humans with MFS. If AT1R blocker demonstrates therapeutic capacity, it may be the first potential pharmacotherapy directed towards preventing phenotypes associated with MFS. 1

38 36 Chapter 1 reference list 1. Judge DP, Dietz HC. Marfan s syndrome. Lancet 2005 Dec 3;366(9501): Arslan-Kirchner M, Arbustini E, Boileau C, Child A, Collod-Beroud G, De PA, et al. Clinical utility gene card for: Marfan syndrome type 1 and related phenotypes [FBN1]. Eur J Hum Genet 2010 Sep;18(9). 3. Loeys B, Nuytinck L, Delvaux I, De BS, De PA. Genotype and phenotype analysis of 171 patients referred for molecular study of the fibrillin-1 gene FBN1 because of suspected Marfan syndrome. Arch Intern Med 2001 Nov 12;161(20): De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996 Apr 24;62(4): Loeys BL, Dietz HC, Braverman AC, Callewaert BL, De BJ, Devereux RB, et al. The revised Ghent nosology for the Marfan syndrome. J Med Genet 2010 Jul;47(7): Faivre L, Collod-Beroud G, Ades L, Arbustini E, Child A, Callewaert B, et al. The new Ghent criteria for Marfan syndrome: what do they change? Clin Genet 2011 May Milewicz DM, Dietz HC, Miller DC. Treatment of aortic disease in patients with Marfan syndrome. Circulation 2005 Mar 22;111(11):e150-e Faivre L, Collod-Beroud G, Loeys BL, Child A, Binquet C, Gautier E, et al. Effect of mutation type and location on clinical outcome in 1,013 probands with Marfan syndrome or related phenotypes and FBN1 mutations: an international study. Am J Hum Genet 2007 Sep;81(3): van Karnebeek CD, Naeff MS, Mulder BJ, Hennekam RC, Offringa M. Natural history of cardiovascular manifestations in Marfan syndrome. Arch Dis Child 2001 Feb;84(2): Chen S, Fagan LF, Nouri S, Donahoe JL. Ventricular dysrhythmias in children with Marfan s syndrome. Am J Dis Child 1985 Mar;139(3): Yetman AT, Bornemeier RA, McCrindle BW. Long-term outcome in patients with Marfan syndrome: is aortic dissection the only cause of sudden death? J Am Coll Cardiol 2003 Jan 15;41(2): Chen S, Fagan LF. Wolff-Parkinson-White syndrome in Marfan s syndrome. J Pediatr 1974 Feb;84(2): Yetman AT, Temple J, Erickson CC. Radiofrequency ablation of a left-sided atrioventricular pathway in a patient with Marfan syndrome. Cardiol Young 2002 Oct;12(5): Meijboom LJ, Timmermans J, van Tintelen JP, Nollen GJ, De BJ, van den Berg MP, et al. Evaluation of left ventricular dimensions and function in Marfan s syndrome without significant valvular regurgitation. Am J Cardiol 2005 Mar 15;95(6): de Witte P, Aalberts JJ, Radonic T, Timmermans J, Scholte AJ, Zwinderman AH, et al. Intrinsic biventricular dysfunction in Marfan syndrome. Heart 2011 Oct Das BB, Taylor AL, Yetman AT. Left ventricular diastolic dysfunction in children and young adults with Marfan syndrome. Pediatr Cardiol 2006 Mar;27(2): Pyeritz RE, Wappel MA. Mitral valve dysfunction in the Marfan syndrome. Clinical and echocardiographic study of prevalence and natural history. Am J Med 1983 May;74(5): Roberts WC, Honig HS. The spectrum of cardiovascular disease in the Marfan syndrome: a clinicomorphologic study of 18 necropsy patients and comparison to 151 previously reported necropsy patients. Am Heart J 1982 Jul;104(1): Hirata K, Triposkiadis F, Sparks E, Bowen J, Wooley CF, Boudoulas H. The Marfan syndrome: abnormal aortic elastic properties. J Am Coll Cardiol 1991 Jul;18(1):57-63.

39 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome Groenink M, de Roos A, Mulder BJ, Spaan JA, van der Wall EE. Changes in aortic distensibility and pulse wave velocity assessed with magnetic resonance imaging following beta-blocker therapy in the Marfan syndrome. Am J Cardiol 1998 Jul 15;82(2): Jeremy RW, Huang H, Hwa J, McCarron H, Hughes CF, Richards JG. Relation between age, arterial distensibility, and aortic dilatation in the Marfan syndrome. Am J Cardiol 1994 Aug 15;74(4): Detaint D, Faivre L, Collod-Beroud G, Child AH, Loeys BL, Binquet C, et al. Cardiovascular manifestations in men and women carrying a FBN1 mutation. Eur Heart J 2010 Sep;31(18): Taub CC, Stoler JM, Perez-Sanz T, Chu J, Isselbacher EM, Picard MH, et al. Mitral valve prolapse in Marfan syndrome: an old topic revisited. Echocardiography 2009 Apr;26(4): Morse RP, Rockenmacher S, Pyeritz RE, Sanders SP, Bieber FR, Lin A, et al. Diagnosis and management of infantile marfan syndrome. Pediatrics 1990 Dec;86(6): Sisk HE, Zahka KG, Pyeritz RE. The Marfan syndrome in early childhood: analysis of 15 patients diagnosed at less than 4 years of age. Am J Cardiol 1983 Aug;52(3): Faivre L, Masurel-Paulet A, Collod-Beroud G, Callewaert BL, Child AH, Stheneur C, et al. Clinical and molecular study of 320 children with Marfan syndrome and related type I fibrillinopathies in a series of 1009 probands with pathogenic FBN1 mutations. Pediatrics 2009 Jan;123(1): Silverman DI, Burton KJ, Gray J, Bosner MS, Kouchoukos NT, Roman MJ, et al. Life expectancy in the Marfan syndrome. Am J Cardiol 1995 Jan 15;75(2): Carr JA, Savage EB. Aortic valve repair for aortic insufficiency in adults: a contemporary review and comparison with replacement techniques. Eur J Cardiothorac Surg 2004 Jan;25(1): David TE. Aortic valve repair for management of aortic insufficiency. Adv Card Surg 1999;11: El KG, Vanoverschelde JL, Glineur D, Poncelet A, Verhelst R, Astarci P, et al. Repair of aortic valve prolapse: experience with 44 patients. Eur J Cardiothorac Surg 2004 Sep;26(3): El KG, Vanoverschelde JL, Glineur D, Pierard F, Verhelst RR, Rubay J, et al. Repair of bicuspid aortic valves in patients with aortic regurgitation. Circulation 2006 Jul 4;114(1 Suppl):I610-I Langer F, Graeter T, Nikoloudakis N, Aicher D, Wendler O, Schafers HJ. Valve-preserving aortic replacement: does the additional repair of leaflet prolapse adversely affect the results? J Thorac Cardiovasc Surg 2001 Aug;122(2): Lausberg HF, Aicher D, Langer F, Schafers HJ. Aortic valve repair with autologous pericardial patch. Eur J Cardiothorac Surg 2006 Aug;30(2): Schafers HJ, Aicher D, Langer F. Correction of leaflet prolapse in valve-preserving aortic replacement: pushing the limits? Ann Thorac Surg 2002 Nov;74(5):S1762-S Minakata K, Schaff HV, Zehr KJ, Dearani JA, Daly RC, Orszulak TA, et al. Is repair of aortic valve regurgitation a safe alternative to valve replacement? J Thorac Cardiovasc Surg 2004 Mar;127(3): le Polain de Waroux JB, Pouleur AC, Robert A, Pasquet A, Gerber BL, Noirhomme P, et al. Mechanisms of recurrent aortic regurgitation after aortic valve repair: predictive value of intraoperative transesophageal echocardiography. JACC Cardiovasc Imaging 2009 Aug;2(8): Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart 2006 Sep;92(9): Mulder BJ. The distal aorta in the Marfan syndrome. Neth Heart J 2008 Nov;16(11): Carlson RG, Lillehei CW, Edwards JE. Cystic medial necrosis of the ascending aorta in relation to age and hypertension. Am J Cardiol 1970 Apr;25(4): Dean JC. Management of Marfan syndrome. Heart 2002 Jul;88(1):

40 38 Chapter Nagashima H, Sakomura Y, Aoka Y, Uto K, Kameyama K, Ogawa M, et al. Angiotensin II type 2 receptor mediates vascular smooth muscle cell apoptosis in cystic medial degeneration associated with Marfan s syndrome. Circulation 2001 Sep 18;104(12 Suppl 1):I282-I Dingemans KP, Teeling P, Lagendijk JH, Becker AE. Extracellular matrix of the human aortic media: an ultrastructural histochemical and immunohistochemical study of the adult aortic media. Anat Rec 2000 Jan 1;258(1): Ramirez F, Rifkin DB. Cell signaling events: a view from the matrix. Matrix Biol 2003 Apr;22(2): Neptune ER, Frischmeyer PA, Arking DE, Myers L, Bunton TE, Gayraud B, et al. Dysregulation of TGF-beta activation contributes to pathogenesis in Marfan syndrome. Nat Genet 2003 Mar;33(3): Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science 2006 Apr 7;312(5770): Ng CM, Cheng A, Myers LA, Martinez-Murillo F, Jie C, Bedja D, et al. TGF-beta-dependent pathogenesis of mitral valve prolapse in a mouse model of Marfan syndrome. J Clin Invest 2004 Dec;114(11): Pueyo ME, Michel JB. Angiotensin II receptors in endothelial cells. Gen Pharmacol 1997 Nov;29(5): Dikalova A, Clempus R, Lassegue B, Cheng G, McCoy J, Dikalov S, et al. Nox1 overexpression potentiates angiotensin II-induced hypertension and vascular smooth muscle hypertrophy in transgenic mice. Circulation 2005 Oct 25;112(17): Rueckschloss U, Quinn MT, Holtz J, Morawietz H. Dose-dependent regulation of NAD(P)H oxidase expression by angiotensin II in human endothelial cells: protective effect of angiotensin II type 1 receptor blockade in patients with coronary artery disease. Arterioscler Thromb Vasc Biol 2002 Nov 1;22(11): Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes atherosclerotic lesions and aneurysms in apolipoprotein E-deficient mice. J Clin Invest 2000 Jun;105(11): Su EJ, Lombardi DM, Siegal J, Schwartz SM. Angiotensin II induces vascular smooth muscle cell replication independent of blood pressure. Hypertension 1998 Jun;31(6): Lee AA, Dillmann WH, McCulloch AD, Villarreal FJ. Angiotensin II stimulates the autocrine production of transforming growth factor-beta 1 in adult rat cardiac fibroblasts. J Mol Cell Cardiol 1995 Oct;27(10): Rodriguez-Vita J, Sanchez-Lopez E, Esteban V, Ruperez M, Egido J, Ruiz-Ortega M. Angiotensin II activates the Smad pathway in vascular smooth muscle cells by a transforming growth factorbeta-independent mechanism. Circulation 2005 May 17;111(19): Jones ES, Black MJ, Widdop RE. Angiotensin AT2 receptor contributes to cardiovascular remodelling of aged rats during chronic AT1 receptor blockade. J Mol Cell Cardiol 2004 Nov;37(5): Akishita M, Ito M, Lehtonen JY, Daviet L, Dzau VJ, Horiuchi M. Expression of the AT2 receptor developmentally programs extracellular signal-regulated kinase activity and influences fetal vascular growth. J Clin Invest 1999 Jan;103(1): Ulmasov B, Xu Z, Tetri LH, Inagami T, Neuschwander-Tetri BA. Protective role of angiotensin II type 2 receptor signaling in a mouse model of pancreatic fibrosis. Am J Physiol Gastrointest Liver Physiol 2009 Feb;296(2):G284-G294.

41 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome Habashi JP, Doyle JJ, Holm TM, Aziz H, Schoenhoff F, Bedja D, et al. Angiotensin II type 2 receptor signaling attenuates aortic aneurysm in mice through ERK antagonism. Science 2011 Apr 15;332(6027): Busuttil RW, Abou-Zamzam AM, Machleder HI. Collagenase activity of the human aorta. A comparison of patients with and without abdominal aortic aneurysms. Arch Surg 1980 Nov;115(11): Busuttil RW, Rinderbriecht H, Flesher A, Carmack C. Elastase activity: the role of elastase in aortic aneurysm formation. J Surg Res 1982 Mar;32(3): Murphy G, Cockett MI, Ward RV, Docherty AJ. Matrix metalloproteinase degradation of elastin, type IV collagen and proteoglycan. A quantitative comparison of the activities of 95 kda and 72 kda gelatinases, stromelysins-1 and -2 and punctuated metalloproteinase (PUMP). Biochem J 1991 Jul 1;277 ( Pt 1): Senior RM, Griffin GL, Fliszar CJ, Shapiro SD, Goldberg GI, Welgus HG. Human 92- and 72-kilodalton type IV collagenases are elastases. J Biol Chem 1991 Apr 25;266(12): Thompson RW, Holmes DR, Mertens RA, Liao S, Botney MD, Mecham RP, et al. Production and localization of 92-kilodalton gelatinase in abdominal aortic aneurysms. An elastolytic metalloproteinase expressed by aneurysm-infiltrating macrophages. J Clin Invest 1995 Jul;96(1): Thompson RW, Parks WC. Role of matrix metalloproteinases in abdominal aortic aneurysms. Ann N Y Acad Sci 1996 Nov 18;800: Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N, Baxter BT. Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysms. J Clin Invest 2002 Sep;110(5): Segura AM, Luna RE, Horiba K, Stetler-Stevenson WG, McAllister HA, Jr., Willerson JT, et al. Immunohistochemistry of matrix metalloproteinases and their inhibitors in thoracic aortic aneurysms and aortic valves of patients with Marfan s syndrome. Circulation 1998 Nov 10;98(19 Suppl):II331- II Bunton TE, Biery NJ, Myers L, Gayraud B, Ramirez F, Dietz HC. Phenotypic alteration of vascular smooth muscle cells precedes elastolysis in a mouse model of Marfan syndrome. Circ Res 2001 Jan 19;88(1): Xiong W, Knispel RA, Dietz HC, Ramirez F, Baxter BT. Doxycycline delays aneurysm rupture in a mouse model of Marfan syndrome. J Vasc Surg 2008 Jan;47(1): Chung AW, Au YK, Sandor GG, Judge DP, Dietz HC, van BC. Loss of elastic fiber integrity and reduction of vascular smooth muscle contraction resulting from the upregulated activities of matrix metalloproteinase-2 and -9 in the thoracic aortic aneurysm in Marfan syndrome. Circ Res 2007 Aug 31;101(5): Freestone T, Turner RJ, Coady A, Higman DJ, Greenhalgh RM, Powell JT. Inflammation and matrix metalloproteinases in the enlarging abdominal aortic aneurysm. Arterioscler Thromb Vasc Biol 1995 Aug;15(8): Chung AW, Yang HH, van BC. Imbalanced synthesis of cyclooxygenase-derived thromboxane A2 and prostacyclin compromises vasomotor function of the thoracic aorta in Marfan syndrome. Br J Pharmacol 2007 Oct;152(3): Chung AW, Au YK, Cortes SF, Sandor GG, Judge DP, Dietz HC, et al. Endothelial dysfunction and compromised enos/akt signaling in the thoracic aorta during the progression of Marfan syndrome. Br J Pharmacol 2007 Apr;150(8): Cecconi M, Nistri S, Quarti A, Manfrin M, Colonna PL, Molini E, et al. Aortic dilatation in patients with bicuspid aortic valve. J Cardiovasc Med (Hagerstown ) 2006 Jan;7(1):

42 40 Chapter Wheat MW, Palmer RF, Bartley TD, Seelman RC. Treatment of dissecting aneurysms of the aorta with surgery. J Thorac Cardiovasc Surg 1965 Sep;50: Harris PD, Malm JR, Bigger JT, Jr., Bowman FO, Jr. Follow-up studies of acute dissecting aortic aneurysms managed with antihypertensive agents. Circulation 1967 Apr;35(4 Suppl):I183-I Simpson CF, Kling JM, PALMER RF. Beta-aminopropionitrile-induced dissecting aneurysms of turkeys: treatment with propranolol. Toxicol Appl Pharmacol 1970 Jan;16(1): Halpern BL, Char F, Murdoch JL, Horton WB, McKusick VA. A prospectus on the prevention of aortic rupture in the Marfan syndrome with data on survivorship without treatment. Johns Hopkins Med J 1971 Sep;129(3): Ose L, McKusick VA. Prophylactiv use of propranolol in the Marfan syndrome to prevent aortic dissection. Birth Defects Orig Artic Ser 1977;13(3C): Shores J, Berger KR, Murphy EA, Pyeritz RE. Progression of aortic dilatation and the benefit of longterm beta-adrenergic blockade in Marfan s syndrome. N Engl J Med 1994 May 12;330(19): Rossi-Foulkes R, Roman MJ, Rosen SE, Kramer-Fox R, Ehlers KH, O Loughlin JE, et al. Phenotypic features and impact of beta blocker or calcium antagonist therapy on aortic lumen size in the Marfan syndrome. Am J Cardiol 1999 May 1;83(9): Roman MJ, Rosen SE, Kramer-Fox R, Devereux RB. Prognostic significance of the pattern of aortic root dilation in the Marfan syndrome. J Am Coll Cardiol 1993 Nov 1;22(5): Tahernia AC. Cardiovascular anomalies in Marfan s syndrome: the role of echocardiography and beta-blockers. South Med J 1993 Mar;86(3): Salim MA, Alpert BS, Ward JC, Pyeritz RE. Effect of beta-adrenergic blockade on aortic root rate of dilation in the Marfan syndrome. Am J Cardiol 1994 Sep 15;74(6): Gersony DR, McClaughlin MA, Jin Z, Gersony WM. The effect of beta-blocker therapy on clinical outcome in patients with Marfan s syndrome: a meta-analysis. Int J Cardiol 2007 Jan 18;114(3): Rios AS, Silber EN, Bavishi N, Varga P, Burton BK, Clark WA, et al. Effect of long-term beta-blockade on aortic root compliance in patients with Marfan syndrome. Am Heart J 1999 Jun;137(6): Groenink M, de Roos A, Mulder BJ, Verbeeten B, Jr., Timmermans J, Zwinderman AH, et al. Biophysical properties of the normal-sized aorta in patients with Marfan syndrome: evaluation with MR flow mapping. Radiology 2001 May;219(2): Reed CM, Fox ME, Alpert BS. Aortic biomechanical properties in pediatric patients with the Marfan syndrome, and the effects of atenolol. Am J Cardiol 1993 Mar 1;71(7): Nollen GJ, Westerhof BE, Groenink M, Osnabrugge A, van der Wall EE, Mulder BJ. Aortic pressurearea relation in Marfan patients with and without beta blocking agents: a new non-invasive approach. Heart 2004 Mar;90(3): Meijboom LJ, Westerhof BE, Nollen GJ, Spaan JA, de Mol BA, Jacobs MJ, et al. Beta-blocking therapy in patients with the Marfan syndrome and entire aortic replacement. Eur J Cardiothorac Surg 2004 Nov;26(5): Yetman AT, Bornemeier RA, McCrindle BW. Usefulness of enalapril versus propranolol or atenolol for prevention of aortic dilation in patients with the Marfan syndrome. Am J Cardiol 2005 May 1;95(9): Yin FC, Brin KP, Ting CT, Pyeritz RE. Arterial hemodynamic indexes in Marfan s syndrome. Circulation 1989 Apr;79(4): Haouzi A, Berglund H, Pelikan PC, Maurer G, Siegel RJ. Heterogeneous aortic response to acute beta-adrenergic blockade in Marfan syndrome. Am Heart J 1997 Jan;133(1):60-3.

43 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome Selamet Tierney ES, Feingold B, Printz BF, Park SC, Graham D, Kleinman CS, et al. Beta-blocker therapy does not alter the rate of aortic root dilation in pediatric patients with Marfan syndrome. J Pediatr 2007 Jan;150(1): Ladouceur M, Fermanian C, Lupoglazoff JM, Edouard T, Dulac Y, Acar P, et al. Effect of betablockade on ascending aortic dilatation in children with the Marfan syndrome. Am J Cardiol 2007 Feb 1;99(3): Manalo-Estrella P, Barker AE. Histopathologic findings in human aortic media associated with pregnancy. Arch Pathol 1967 Apr;83(4): Immer FF, Bansi AG, Immer-Bansi AS, McDougall J, Zehr KJ, Schaff HV, et al. Aortic dissection in pregnancy: analysis of risk factors and outcome. Ann Thorac Surg 2003 Jul;76(1): Lipscomb KJ, Smith JC, Clarke B, Donnai P, Harris R. Outcome of pregnancy in women with Marfan s syndrome. Br J Obstet Gynaecol 1997 Feb;104(2): Meijboom LJ, Vos FE, Timmermans J, Boers GH, Zwinderman AH, Mulder BJ. Pregnancy and aortic root growth in the Marfan syndrome: a prospective study. Eur Heart J 2005 May;26(9): Meijboom LJ, Drenthen W, Pieper PG, Groenink M, van der Post JA, Timmermans J, et al. Obstetric complications in Marfan syndrome. Int J Cardiol 2006 Jun 7;110(1): Mulder BJ, Bleker OP. Valvular heart disease in pregnancy. N Engl J Med 2003 Oct 2;349(14): Pyeritz RE. Maternal and fetal complications of pregnancy in the Marfan syndrome. Am J Med 1981 Nov;71(5): Rossiter JP, Repke JT, Morales AJ, Murphy EA, Pyeritz RE. A prospective longitudinal evaluation of pregnancy in the Marfan syndrome. Am J Obstet Gynecol 1995 Nov;173(5): Zeebregts CJ, Schepens MA, Hameeteman TM, Morshuis WJ, de la Riviere AB. Acute aortic dissection complicating pregnancy. Ann Thorac Surg 1997 Nov;64(5): Ferrero S, Colombo BM, Ragni N. Maternal arrhythmias during pregnancy. Arch Gynecol Obstet 2004 May;269(4): Yankowitz J. Pharmacologic treatment of hypertensive disorders during pregnancy. J Perinat Neonatal Nurs 2004 Jul;18(3): Yetman AT, Roosevelt GE, Veit N, Everitt MD. Distal aortic and peripheral arterial aneurysms in patients with marfan syndrome. J Am Coll Cardiol 2011 Dec 6;58(24): Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart 2006 Sep;92(9): Mulder BJ. The distal aorta in the Marfan syndrome. Neth Heart J 2008 Nov;16(11): Detter C, Mair H, Klein HG, Georgescu C, Welz A, Reichart B. Long-term prognosis of surgicallytreated aortic aneurysms and dissections in patients with and without Marfan syndrome. Eur J Cardiothorac Surg 1998 Apr;13(4): London GM, Guerin AP. Influence of arterial pulse and reflected waves on blood pressure and cardiac function. Am Heart J 1999 Sep;138(3 Pt 2): Safar ME, Tual JL, Safavian A, London GM. Antihypertensive therapy and wave reflections. Prostaglandins Leukot Essent Fatty Acids 1996 Jan;54(1): Ting CT, Chou CY, Chang MS, Wang SP, Chiang BN, Yin FC. Arterial hemodynamics in human hypertension. Effects of adrenergic blockade. Circulation 1991 Sep;84(3): Curci JA, Mao D, Bohner DG, Allen BT, Rubin BG, Reilly JM, et al. Preoperative treatment with doxycycline reduces aortic wall expression and activation of matrix metalloproteinases in patients with abdominal aortic aneurysms. J Vasc Surg 2000 Feb;31(2): Thompson RW, Baxter BT. MMP inhibition in abdominal aortic aneurysms. Rationale for a prospective randomized clinical trial. Ann N Y Acad Sci 1999 Jun 30;878:

44 42 Chapter Bartoli MA, Parodi FE, Chu J, Pagano MB, Mao D, Baxter BT, et al. Localized administration of doxycycline suppresses aortic dilatation in an experimental mouse model of abdominal aortic aneurysm. Ann Vasc Surg 2006 Mar;20(2): Petrinec D, Liao S, Holmes DR, Reilly JM, Parks WC, Thompson RW. Doxycycline inhibition of aneurysmal degeneration in an elastase-induced rat model of abdominal aortic aneurysm: preservation of aortic elastin associated with suppressed production of 92 kd gelatinase. J Vasc Surg 1996 Feb;23(2): Holmes DR, Petrinec D, Wester W, Thompson RW, Reilly JM. Indomethacin prevents elastaseinduced abdominal aortic aneurysms in the rat. J Surg Res 1996 Jun;63(1): Mosorin M, Juvonen J, Biancari F, Satta J, Surcel HM, Leinonen M, et al. Use of doxycycline to decrease the growth rate of abdominal aortic aneurysms: a randomized, double-blind, placebocontrolled pilot study. J Vasc Surg 2001 Oct;34(4): Chung AW, Yang HH, Radomski MW, van BC. Long-term doxycycline is more effective than atenolol to prevent thoracic aortic aneurysm in marfan syndrome through the inhibition of matrix metalloproteinase-2 and -9. Circ Res 2008 Apr 25;102(8):e73-e Curci JA, Petrinec D, Liao S, Golub LM, Thompson RW. Pharmacologic suppression of experimental abdominal aortic aneurysms: acomparison of doxycycline and four chemically modified tetracyclines. J Vasc Surg 1998 Dec;28(6): Manning MW, Cassis LA, Daugherty A. Differential effects of doxycycline, a broad-spectrum matrix metalloproteinase inhibitor, on angiotensin II-induced atherosclerosis and abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol 2003 Mar 1;23(3): Prall AK, Longo GM, Mayhan WG, Waltke EA, Fleckten B, Thompson RW, et al. Doxycycline in patients with abdominal aortic aneurysms and in mice: comparison of serum levels and effect on aneurysm growth in mice. J Vasc Surg 2002 May;35(5): Pyo R, Lee JK, Shipley JM, Curci JA, Mao D, Ziporin SJ, et al. Targeted gene disruption of matrix metalloproteinase-9 (gelatinase B) suppresses development of experimental abdominal aortic aneurysms. J Clin Invest 2000 Jun;105(11): Sheth RA, Maricevich M, Mahmood U. In vivo optical molecular imaging of matrix metalloproteinase activity in abdominal aortic aneurysms correlates with treatment effects on growth rate. Atherosclerosis 2010 Sep;212(1): Sho E, Chu J, Sho M, Fernandes B, Judd D, Ganesan P, et al. Continuous periaortic infusion improves doxycycline efficacy in experimental aortic aneurysms. J Vasc Surg 2004 Jun;39(6): Tedesco MM, Terashima M, Blankenberg FG, Levashova Z, Spin JM, Backer MV, et al. Analysis of in situ and ex vivo vascular endothelial growth factor receptor expression during experimental aortic aneurysm progression. Arterioscler Thromb Vasc Biol 2009 Oct;29(10): Turner GH, Olzinski AR, Bernard RE, Aravindhan K, Karr HW, Mirabile RC, et al. In vivo serial assessment of aortic aneurysm formation in apolipoprotein E-deficient mice via MRI. Circ Cardiovasc Imaging 2008 Nov;1(3): Vinh A, Gaspari TA, Liu HB, Dousha LF, Widdop RE, Dear AE. A novel histone deacetylase inhibitor reduces abdominal aortic aneurysm formation in angiotensin II-infused apolipoprotein E-deficient mice. J Vasc Res 2008;45(2): Yang HH, Kim JM, Chum E, van BC, Chung AW. Effectiveness of combination of losartan potassium and doxycycline versus single-drug treatments in the secondary prevention of thoracic aortic aneurysm in Marfan syndrome. J Thorac Cardiovasc Surg 2010 Aug;140(2): Raffetto JD, Khalil RA. Matrix metalloproteinases and their inhibitors in vascular remodeling and vascular disease. Biochem Pharmacol 2008 Jan 15;75(2):

45 Current and future pharmacological treatment strategies regarding aortic disease in Marfan syndrome Chew DK, Conte MS, Khalil RA. Matrix metalloproteinase-specific inhibition of Ca2+ entry mechanisms of vascular contraction. J Vasc Surg 2004 Nov;40(5): Dodd BR, Spence RA. Doxycycline inhibition of abdominal aortic aneurysm growth - a systematic review of the literature. Curr Vasc Pharmacol 2011 Jul 1;9(4): Crisby M, Nordin-Fredriksson G, Shah PK, Yano J, Zhu J, Nilsson J. Pravastatin treatment increases collagen content and decreases lipid content, inflammation, metalloproteinases, and cell death in human carotid plaques: implications for plaque stabilization. Circulation 2001 Feb 20;103(7): Yu Y, Ohmori K, Chen Y, Sato C, Kiyomoto H, Shinomiya K, et al. Effects of pravastatin on progression of glucose intolerance and cardiovascular remodeling in a type II diabetes model. J Am Coll Cardiol 2004 Aug 18;44(4): Ejiri J, Inoue N, Tsukube T, Munezane T, Hino Y, Kobayashi S, et al. Oxidative stress in the pathogenesis of thoracic aortic aneurysm: protective role of statin and angiotensin II type 1 receptor blocker. Cardiovasc Res 2003 Oct 1;59(4): Sukhija R, Aronow WS, Sandhu R, Kakar P, Babu S. Mortality and size of abdominal aortic aneurysm at long-term follow-up of patients not treated surgically and treated with and without statins. Am J Cardiol 2006 Jan 15;97(2): Ferguson CD, Clancy P, Bourke B, Walker PJ, Dear A, Buckenham T, et al. Association of statin prescription with small abdominal aortic aneurysm progression. Am Heart J 2010 Feb;159(2): McLoughlin D, McGuinness J, Byrne J, Terzo E, Huuskonen V, McAllister H, et al. Pravastatin reduces Marfan aortic dilation. Circulation 2011 Sep 13;124(11 Suppl):S168-S Danyi P, Elefteriades JA, Jovin IS. Medical therapy of thoracic aortic aneurysms: are we there yet? Circulation 2011 Sep 27;124(13): Izzo JL, Jr., Weir MR. Angiotensin-converting enzyme inhibitors. J Clin Hypertens (Greenwich ) 2011 Sep;13(9): Hadi HA, Carr CS, Al SJ. Endothelial dysfunction: cardiovascular risk factors, therapy, and outcome. Vasc Health Risk Manag 2005;1(3): Ahimastos AA, Aggarwal A, D Orsa KM, Formosa MF, White AJ, Savarirayan R, et al. Effect of perindopril on large artery stiffness and aortic root diameter in patients with Marfan syndrome: a randomized controlled trial. JAMA 2007 Oct 3;298(13): Hornig B, Kohler C, Drexler H. Role of bradykinin in mediating vascular effects of angiotensinconverting enzyme inhibitors in humans. Circulation 1997 Mar 4;95(5): Yang HY, Erdos EG, Levin Y. Characterization of a dipeptide hydrolase (kininase II: angiotensin I converting enzyme). J Pharmacol Exp Ther 1971 Apr;177(1): Yetman AT, Bornemeier RA, McCrindle BW. Usefulness of enalapril versus propranolol or atenolol for prevention of aortic dilation in patients with the Marfan syndrome. Am J Cardiol 2005 May 1;95(9): Liao S, Miralles M, Kelley BJ, Curci JA, Borhani M, Thompson RW. Suppression of experimental abdominal aortic aneurysms in the rat by treatment with angiotensin-converting enzyme inhibitors. J Vasc Surg 2001 May;33(5): Hackam DG, Thiruchelvam D, Redelmeier DA. Angiotensin-converting enzyme inhibitors and aortic rupture: a population-based case-control study. Lancet 2006 Aug 19;368(9536): Genoni M, Paul M, Tavakoli R, Kunzli A, Lachat M, Graves K, et al. Predictors of complications in acute type B aortic dissection. Eur J Cardiothorac Surg 2002 Jul;22(1):

46 44 Chapter Glower DD, Fann JI, Speier RH, Morrison L, White WD, Smith LR, et al. Comparison of medical and surgical therapy for uncomplicated descending aortic dissection. Circulation 1990 Nov;82(5 Suppl):IV39-IV Glower DD, Speier RH, White WD, Smith LR, Rankin JS, Wolfe WG. Management and long-term outcome of aortic dissection. Ann Surg 1991 Jul;214(1): Neya K, Omoto R, Kyo S, Kimura S, Yokote Y, Takamoto S, et al. Outcome of Stanford type B acute aortic dissection. Circulation 1992 Nov;86(5 Suppl):II1-II Sueyoshi E, Sakamoto I, Hayashi K, Yamaguchi T, Imada T. Growth rate of aortic diameter in patients with type B aortic dissection during the chronic phase. Circulation 2004 Sep 14;110(11 Suppl 1):II256-II Takeshita S, Sakamoto S, Kitada S, Akutsu K, Hashimoto H. Angiotensin-converting enzyme inhibitors reduce long-term aortic events in patients with acute type B aortic dissection. Circ J 2008 Nov;72(11): Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz HC, III. Angiotensin II blockade and aorticroot dilation in Marfan s syndrome. N Engl J Med 2008 Jun 26;358(26): Nistala H, Lee-Arteaga S, Carta L, Cook JR, Smaldone S, Siciliano G, et al. Differential effects of alendronate and losartan therapy on osteopenia and aortic aneurysm in mice with severe Marfan syndrome. Hum Mol Genet 2010 Dec 15;19(24): Habashi JP, Doyle JJ, Holm TM, Aziz H, Schoenhoff F, Bedja D, et al. Angiotensin II type 2 receptor signaling attenuates aortic aneurysm in mice through ERK antagonism. Science 2011 Apr 15;332(6027): Wilkinson IB, Qasem A, McEniery CM, Webb DJ, Avolio AP, Cockcroft JR. Nitric oxide regulates local arterial distensibility in vivo. Circulation 2002 Jan 15;105(2): Chung AW, Au YK, Cortes SF, Sandor GG, Judge DP, Dietz HC, et al. Endothelial dysfunction and compromised enos/akt signaling in the thoracic aorta during the progression of Marfan syndrome. Br J Pharmacol 2007 Apr;150(8): Nakamura M, Itoh S, Makita S, Ohira A, Arakawa N, Hiramori K. Peripheral resistance vessel dysfunction in Marfan syndrome. Am Heart J 2000 Apr;139(4): Current controlled trials. The effect of angiotensin 2 blockade and long acting nitrates on arterial stiffness in patients with marfan syndrome. A placebo controlled studt. (ISRCTN )

47

48

49 Chapter 2 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome Romy Franken, Alexander W. den Hartog, Vivian de Waard, Leo Engele, Teodora Radonic, Rene Lutter, Janneke Timmermans, Arthur J. Scholte, Maarten P. van den Berg, Aeilko H. Zwinderman, Maarten Groenink, Barbara J.M. Mulder Adapted from: International Journal of Cardiology 2013 Oct;168(3):2441-6

50 48 Chapter 2 abstract background Patients with Marfan syndrome (MFS) are at risk for cardiovascular disease. Marfan associated mutations in the FBN1 gene lead to increased transforming growth factor-β (TGF-β) activation. The aim of this study was to investigate the role of plasma TGF-β as a biomarker for progressive aortic root dilatation and dissection. methods Plasma TGF-β level and aortic root diameter by means of echocardiography were assessed in 99 MFS patients. After 38 months of follow-up measurement of the aortic root was repeated and individual aortic root growth curves were constructed. Clinical events were evaluated. The primary composite endpoint was defined as aortic dissection and prophylactic aortic root replacement. results TGF-β levels were higher in MFS patients as compared to healthy controls (109 pg/ml versus 54 pg/ml, p < 0.001). Higher plasma TGF-β levels correlated with larger aortic root dimensions (r = 0.26, p = 0.027), previous aortic root surgery (161 pg/ml versus 88 pg/ml, p = 0.007) and faster aortic root growth rate (r = 0.42, p < 0.001). During 38 months of follow-up, 17 events were observed (four type B dissections and 13 aortic root replacements). Patients with TGF-β levels above 140 pg/ml had a 6.5 times higher risk of experiencing the composite endpoint compared to patients with TGF-β levels below 140 pg/ml (95% CI: 2.1 to 20.1, p = 0.001) with 65% sensitivity and 78% specificity. conclusion Elevated TGF-β level in patients with Marfan syndrome is correlated with larger aortic root diameters and faster aortic root growth. Level of plasma TGF-β predicts cardiovascular events and might serve as a prognostic biomarker in MFS.

51 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 49 introduction Marfan syndrome (MFS) is an autosomal connective tissue disorder caused by mutations in the FBN1 gene with high penetrance, but great clinical variability (1,2). Patients with MFS suffer from an increased risk of cardiovascular manifestations like (asymmetric) aortic root dilatation, mitral valve prolapse, impaired biventricular function and aortic dissection, the latter being the main cause of premature death (3 6). However, it is hard to predict which MFS patient is prone for aortic dissection, since dissections may occur unexpectedly in non-dilated aortas (7) and after prophylactic aortic root surgery (8). Originally, it was thought that MFS was due to structural deficiency of fibrillin-1, leading to weakened microfibrils (9). This hypothesis provided a plausible explanation for aortic disease and eye lens dislocation, but other clinical features such as long bone overgrowth remained incompletely understood. Besides being a structural protein, fibrillin-1 normally binds to a large latent complex (LLC), which comprises the inactive form of transforming growth factor-β (TGF-β) (10). Defective or deficient fibrillin-1, by FBN1 mutations, alters the matrix sequestration of the LLC resulting in activation and enhanced release of TGF-β into the extracellular environment (10). Free and active TGF-β is involved in many cellular processes, including growth inhibition, cell migration and extracellular matrix remodeling (11). Studies with MFS mouse models have consistently demonstrated an increased TGF-β signaling (12 14). Treatment of these MFS mice with TGF-β neutralizing antibodies prevented the development of pathological changes in the aortic wall, progressive aortic root dilatation (15), as well as attenuating of other clinical manifestations of MFS, including pulmonary alveolar septation (13), and myxomatous degeneration of the mitral valve (14). High circulating TGF-β is correlated with increased age and aortic diameters, and decreases upon losartan treatment in MFS mice (16). To our knowledge no clear correlation between circulating TGF-β and aortic diameter in humans with a classic MFS phenotype has been shown. MFS is a highly variable disease in phenotype and age of onset of various manifestations, so a biomarker which predicts aortic disease activity and pathogenic events would be of tremendous value. The purpose of the current study was therefore to investigate whether increased circulating TGF-β levels at baseline can predict progressive aortic root dilatation and aortic dissections. 2 methods Patient population In this predefined sub-study of the COMPARE trial, we included 99 adult MFS patients from four academic centers in the Netherlands between 2008 and 2009 (Academic Medi-

52 50 Chapter 2 cal Center Amsterdam; St. Radboud University Medical Center Nijmegen; Leiden University Medical Center; University Medical Center Groningen) (17). In short, the COMPARE study is a multi-center randomized clinical trial, investigating the effects of losartan on aortic dimensions. One of the secondary objectives is to determine whether TGF-β can be associated with different phenotypic expressions. Inclusion criteria include diagnosis of MFS according to the Ghent criteria (18) and age 18 years. Exclusion criteria are angiotensin converting enzyme (ACE) inhibitor usage and previous replacement of more than one part of the aorta. All patients were included in this substudy, irrespective of the medication they used. The trial was conducted with approval from institutional review boards in four participating academic hospitals in The Netherlands. Written informed consent was obtained from all participants. baseline examination During the baseline visit, all patients were evaluated with extended clinical examination by two investigators (determination of MFS). History of cardiovascular surgery and use of cardiovascular drugs were obtained from patients medical records and questionnaires. For analysis of plasma TGF-β, venous serum blood was collected and measured in EDTA plasma samples. Total TGF-β1, i.e. latent and active TGF-β, was determined by means of ELISA using a commercially available kit (Bio-Rad, Richmond, Canada) following the manufacturer s recommendations. In order to be able to measure the neo-epitope in active TGF-β, the latent TGF-β1 samples were first acid-activated before assaying. The TGF-β levels of MFS patients, measured in the same laboratory, were compared to TGF-β levels of 22 healthy controls. Healthy controls were mostly recruited among the spouses, who were not suspicious of having MFS. A couple of healthy controls were non-affected siblings of MFS patients, who were excluded of having MFS by a clinical geneticist, cardiologist and ophthalmologist from the Marfan outpatient screening clinic in the Academic Medical Centre Amsterdam (The Netherlands). follow-up Doppler echocardiography was performed at baseline and after every year of followup. Aortic root growth was calculated on basis of the aortic root diameter of the most recent echocardiogram minus the aortic root diameter of the baseline echocardiogram. Elective aortic root replacement was blinded for TGF-β levels and completely dependent on the judgment of treating cardiologists based on the 2010 ACC/AHA/AATS and ESC guidelines for the management of patients with Thoracic Aortic Disease (19,20). Patients who had undergone aortic root replacement before the start of the study were excluded in this analysis. In cases with aortic root replacement during study, the echocardiogram made before surgery was used if available, patients with aortic root replacement just after the first echocardiography were excluded from this analysis. The follow-up period

53 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 51 varied between two and five years, therefore the absolute growth in millimeter (mm) was divided by time in months and multiplied by 12 to obtain a growth rate per year (mm/year). Data on events such as aortic dissection and elective aortic root replacement were taken from the databases of the participating hospitals and were evaluated after the two to five years of follow-up. transthoracic echocardiography Transthoracic Doppler echocardiography was performed with a Vivid 7 (GE, Vingmed Ultrasound, Horton, Norway) ultrasound system by multiple echocardiographers. Aortic root diameters were measured in end-diastole at the level of the sinus of Valsalva. Aortic dimensions were measured using the leading edge to leading edge technique, consistent with the current American Society of Echocardiography guidelines (21,22). All echocardiographic images were acquired and recorded digitally, and analyzed by a single observer (RF). 2 statistical analysis Data are presented as mean value ± standard error of the mean or number of patients (percent) or as median and 95% confidence interval (CI), where appropriate. Comparisons between continuous variables were made by two-tailed Student t-tests. Linear regression analysis was used to identify determinants for progressive aortic dilatation and clinical events. Receiver operator characteristic (ROC) analysis was performed to determine sensitivity and specificity of TGF-β in predicting clinical events. Kaplan Meier analysis was used for survival comparison between patients with high (TGF-β > 140 pg/ ml) and low TGF-β (TGF-β < 140 pg/ml). All statistical tests were two-sided and differences were considered statistically significant at p < Data analysis was performed using the SPSS statistical package (19.0 for windows; SPSS Inc., Chicago, Illinois, USA). ROC analysis and figures were generated using GraphPad Prism (version 5.01 for Windows, GraphPad Software, San Diego, California, USA). The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology. results A total of 99 MFS patients (mean age: 36 years (range years)), diagnosed according to the Ghent criteria of 1996 (18), were enrolled, with a mean follow-up of 38 months (range months). Thirty-one patients (31%) had a previous aortic root repair at baseline, with a mean age at first cardiovascular surgery of 31 years (range years). In 84% of the MFS patients a FBN1 mutation was found. The MFS patients and healthy

54 52 Chapter 2 controls were comparable for age (36 years, SEM = 1 versus 34 years, SEM = 2, p = 0.27) and gender (58% versus 55%, p = 0.8). The use of cardiovascular drug therapy and previous cardiovascular surgery were only observed in MFS patients. Baseline demographics of the study population are provided in Table 1. table 1 Baseline characteristics of the study population. variables mfs patients n = 99 Age (years) 36 ± 12 Male gender 57 (58%) BSA (m 2 ) 2.0 ± 0.3 aortic root dilatation Absolute diameter (mm) 43 ± 4 Mean z-score 3.8 ± 1.7 FBN1 mutation (a) 79 (84%) Ectopia lentis 50 (51%) Skeletal involvement 70 (71%) Dural ectasia 51 (52%) Cardiovascular surgery 31 (31%) Aortic root replacement 28 (28%) David procedure 7 (7%) Bentall procedure 21 (21%) Mitral valve correction (b) 3 (3%) Distal aortic graft 1 (1%) cardiovascular medicine β-blockers 73 (74%) Calcium channel blockers 3 (3%) Flecainide 1 (1%) Diuretics 1 (1%) Data presented as number (percentage) or mean ± standard deviation Abbreviations: BSA = body surface area, FBN1 = fibrillin 1 gene. a = 5 of 99 patients did not have FBN1 diagnostics b = 1 of 3 patients also had had a Bentall procedure age and cardiovascular medical treatment Baseline plasma TGF-β levels were significantly higher in patients with MFS compared to healthy controls (109 pg/ml, SEM = 12 pg/ml versus 54 pg/ml, SEM = 9 pg/ml, p < 0.001) (Figure 1A). In our cohort 78 patients used cardiovascular drug therapy, 74% of the patients were using β-blockers, three patients were on calcium channel blockers, one patient used a diuretic and one patient used flecainide (Table 1). Patients using

55 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 53 β-blocker therapy did not have lower TGF-β levels compared to MFS patients without using cardiovascular medicine (109 pg/ml, SEM = 12 pg/ml versus 81 pg/ml, SEM = 15 pg/ml, p = 0.17) (Figure 1A). No significant difference in aortic root diameter was found between patients using β-blockers (43.8 mm, SEM =0.6 mm) and patients without β-blocker treatment (41.6 mm, SEM = 1.1 mm, p = 0.09). Figure 1B shows no relation between age and level of plasma TGF-β in MFS patients (r = 0.17, p = 0.07) as well as in healthy controls (r = 0.2, p = 0.35). 2 figure 1 A: Mean and standard error of the mean of total TGF-β plasma levels in patients with Marfan syndrome (MFS) (black box), healthy controls (white box), MFS patients without cardiovascular (CV) drugs (dark gray box) and MFS patients with β-blocker use (light gray box). B: Relation of plasma TGF-β levels and age in patients with Marfan syndrome (MFS) in black dots and healthy controls in white dots. aortic root dimensions, Previous aortic root surgery and skeletal features Within our MFS cohort 31% of the patients had undergone aortic root surgery before the start of the study. TGF-β levels were studied in the patient group with aortic root repair and compared to the patients with their native aortic root. In patients without

56 54 Chapter 2 previous aortic root surgery, aortic root diameter was significantly correlated with circulating total TGF-β levels (n = 71, r = 0.26, p = 0.027) (Figure 2A). Furthermore, patients with an aortic root replacement before inclusion had significant higher circulating TGF-β levels in plasma compared to patients without prior cardiovascular surgery (164 pg/ml, SEM = 27 versus 88 pg/ml, SEM = 13, p = 0.005) (Figure 2B). No correlation was found between circulating TGF-β and body surface area (BSA) (r < 0.001, p = 0.978). Moreover, no difference was demonstrated between TGF-β levels and patients with or without skeletal involvement (106 pg/ml, SEM = 14 versus 115 pg/ml, SEM = 25, p = 0.759), with or without a known FBN1 mutation (106 pg/ml, SEM = 13 versus 89 pg/ml. SEM = 28, p = 0.598) and between TGF-β levels and patients with or without ectopia lentis (101 pg/ml, SEM = 15 versus 116 pg/ml, SEM = 19, p = 0.539). figure 2 A: Correlation between total plasma TGF-β levels and echocardiographic aortic root diameters in patients with Marfan syndrome. B: Mean and standard of the mean of total TGF-β plasma levels in MFS patients without previous aortic root (AoR) surgery (black box) and MFS patients with previous AoR surgery (white box). tgf-β levels and follow-up Figure 3A shows a significant correlation between circulating TGF-β levels and aortic root growth in patients with MFS after a mean follow-up of 3 years (n = 67, r = 0.42, p < 0.001). In total 28 patients were excluded from this analysis because they already had undergone aortic root surgery previous to the study, another four patients were

57 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 55 excluded from this analysis because they were sent to surgery just after the first echocardiography. Univariate analysis of gender, age, BSA and aortic root dimensions as predictors for aortic root growth (mm/year) are described in Table 2. Besides plasma levels of TGF-β no other significant predictor for aortic root growth in our cohort was found. Correction for β -blocker usage revealed that TGF-β level was the sole independent parameter associated with aortic root growth (β = 0.45, SEM < 0.001, p < 0.001). Subanalyses in patients with and without β-blocker treatment revealed similar results with regard to TGF-β and aortic root growth (n = 49, r = 0.50, p < versus n = 18, r = 0.32, p = 0.20). Subanalyses in patients with and without losartan treatment showed similar results with regard to TGF-β and aortic root growth (n = 40, r = 0.42, p = versus n = 27, r = 0.41, p = 0.04). 2 table 2 Univariate analysis for predictors for aortic root growth and clinical events. aortic root growth clinical events β p β p baseline characteristics Gender Age Body surface area β-blocker use TGF-β level 0.42 < Aortic root diameter (mm) by echo During a mean period of 38 months follow-up 17 events were observed (four aortic type B dissections and 13 elective aortic root replacements indicated following the current guidelines), with a mean time to event of 28 months, SEM = 4 months. ROC curve analysis revealed an optimum cut-off value of 140 pg/ml for the composite clinical endpoint, with an area under the curve (AUC) of 0.71 (95% CI = 0.58 to 0.84; p = 0.006), yielding a sensitivity of 65% (95% CI = 38% to 86%) and a specificity of 78% (95% CI = 68% to 86%) (Figure 3B). This cut-off value corresponded to the mean TGF-β level plus two times the standard deviation of our healthy controls (mean 54 pg/ml, standard deviation 41 pg/ ml, cut-off value: 136 pg/ml). For detailed analysis of the effect of TGF-β plasma level, the total patient cohort was divided into two groups (Group 1: < 140 pg/ml and Group 2: >140 pg/ml). Kaplan Meier survival analysis revealed that patients with a TGF-β level above 140 pg/ml had a significant higher probability of meeting the combined endpoint (log rank-test, p = 0.006) (Figure 3C), with a Hazard Ratio of 6.5 (95% CI: 2.1 to 20.1, p = 0.001).

58 56 Chapter 2 figure 3 A: Correlation between total plasma TGF-β levels and aortic root growth in mm/year measured by echocardiogram in patients with Marfan syndrome. B: Area under curve (AUC) shows sensitivity and specificity of TGF-β as predictive value for the combined primary endpoint elective aortic root replacement and aortic dissection. C: Kaplan Meier curve shows percentage freedom of events of TGF-β < 140 pg/ml and TGF-β > 140 pg/ml. Numbers of patients at risk of events are demonstrated below the figure.

59 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 57 discussion Our study is the first to demonstrate that TGF-β might serve as a prognostic biomarker, since a single determination of TGF-β level at baseline is an independent predictor for cardiovascular events in MFS. Patients with a TGF-β level above 140 pg/ml had a 6.5 times higher likelihood to reach the combined endpoint (aortic dissection and elective aortic root surgery). Moreover, we demonstrated a positive correlation between circulating TGF-β levels and aortic root diameters at baseline, as well as aortic root growth rate after a mean follow-up of 38 months. The importance of cardiovascular features in MFS was already emphasized by Murdoch in 1972, showing that lifespan in MFS patients was markedly shortened due to unpredictable cardiovascular events (23). Patients with MFS may develop an aortic dissection due to progressive dilatation in the entire aorta (24), which can be predicted by both aortic diameter and aortic elasticity as independent parameters (6,25,26). Furthermore, the asymmetry of the aortic root might be of clinical importance in unexpected aortic root dissection (5). Increased clinical awareness and availability of a surgical technique for prophylactic aortic root replacement have greatly increased life expectancy by 30 years in the MFS population (27). However, patients with MFS are still at risk for aortic dissection, predominantly in the distal aorta after successful prophylactic aortic root replacement (4,28,29). Moreover, aortic dissection may also occur in patients without prior aortic dilatation (4,7,28,29). In the last decades tremendous progress has been made in elucidating the common molecular mechanisms underlying the pathogenesis of MFS. However, despite increased FBN1 mutation screening, clinical severity cannot be predicted yet, due to the unclear genotype phenotype correlations. These observations highlight the need for an informative distinctive prognostic marker to monitor aortic disease in MFS. Previous studies have shown that the aortic vessel of humans and mice with MFS had excessive TGF-β activation by studying phosphorylation of downstream signaling molecules, transcription factors SMAD 2/3 (13,15,16,30). Haskett et al. demonstrated in MFS mice that disruption in the elastic lamina of nonaneurismal descending aortas due to biaxial testing is accompanied by an increasing TGF-β signaling and that TGF-β can regulate collagen organization (31). Furthermore, in a MFS mouse dissection model, it has been shown that serum TGF-β levels were significantly increased before a dissection occurred (32). These data suggest that pathological changes in the aortic wall, ultimately leading to dissection, might lead to dysregulation of TGF-β. This is consistent with the current model of MFS in which large latent complex fails to be sequestered by defective fibrilline-1, resulting in more bioavailable TGF-β, leading to chronic and excessive activation of the SMAD transcription factors (10,30,33). Along the line, TGF-β antagonism through TGF-β neutralizing antibodies or inhibition of the angiotensin II receptor type 1 (that stimulates TGF-β production) by losartan treatment prevented progressive aortic dilatation (15) and other clinical features in MFS 2

60 58 Chapter 2 mouse models. The therapeutic benefits of TGF-β antagonism, by losartan treatment, were confirmed in humans in a small MFS paediatric cohort (34). Plasma TGF-β levels might be increased in other diseases such as myocardial fibrosis, however in this young cohort with cardiovascular management, myocardial infarction or hypertrophic cardiomyopathy is not very likely. In the study of Matt et al., strong correlations were found between circulating TGF-β levels and aortic root diameters in MFS mice, however no correlations could be demonstrated in their human cohort (16). In our study, we did find a significant correlation between aortic root diameters and TGF-β levels, which is probably due to the clinically more severely affected MFS population as expressed in larger aortic root dimensions and z-scores. Furthermore, 26% of the MFS population studied by Matt et al. used losartan treatment, with significantly lower TGF-β levels and another 7% used ACE inhibitors with a tendency toward lower TGF-β levels (16). ACE inhibitors prevent angiotensin II conversion leading to reduced signaling via both angiotensin receptor types 1 and 2, while losartan only inhibits the type 1 receptor. Both medications will give lower TGF-β production, however, the signaling via the type 2 receptor is beneficial for the vessel wall. For this reason we did not include MFS patients on ACE inhibitors in our cohort. In this highly variable population a clear correlation between TGF-β levels and aortic root diameters could not be found and no follow-up data was available. In our subanalysis in this study we found similar results for patients with and without losartan or β-blocker treatment regarding TGF-β and aortic root growth. This suggests that TGF-β corresponds to aortic root growth irrespective to the use of cardiovascular medication. To our knowledge, the observed correlation in our study between TGF-β and progressive aortic root growth and even more remarkable the predictive value of TGF-β levels on clinical events have not been reported before. Plasma TGF-β might thus serve as a prognostic biomarker in patients with MFS. Aortic root replacement in medical history showed higher plasma TGF-β levels compared to patients with their native aortic root. This may be caused by aortic stiffness alteration after prosthetic aortic root or it may be due to permanent wall damage in other parts of the aorta. Furthermore, TGF-β levels and aortic root diameters were not very strongly correlated, nevertheless TGF-β level is highly predictive for aortic events. This also suggests that high TGF-β levels are rather a consequence of severe wall damage than the cause of the aortic pathology. This implies that there might be an even more comprehensive pathway leading to the cardiovascular features of MFS. It has been suggested that inflammation plays a modifying role in the pathogenesis of MFS, and that inflammation might be a novel therapeutic target in MFS patients (35). More fundamental research to unravel new molecular pathways remains necessary to optimize treatment in MFS patients. In the mean time results of several multicentre, randomized, controlled clinical trials are awaited.

61 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 59 limitations In this observational follow-up study, relatively small patient numbers were included and only four patients reached the clinical endpoint of aortic dissection. Therefore, the combined endpoint of aortic dissections with physician driven elective aortic root surgery was chosen. Our study aimed to predict aortic root growth and aortic events by a single baseline determination of plasma TGF-β level. An interesting point would have been the repetitive testing of TGF-β to garner precise information about the time dependency and stability of these values in each patient. A drawback for the implementation of TGF-β measurements as a prognostic tool for aortic damage may be the large variability in TGF-β measurements in different laboratories. Therefore, our data cannot be easily extrapolated to other centers, and normalized standards should be determined by each individual centre. Asymmetrical shape of aortic root in MFS can bias the echocardiographic measurements, other imaging modalities such as cardiac magnetic resonance imaging or computed tomography are better exam methods for aorta root measurements and follow-up. 2 conclusion Plasma TGF-β levels are correlated with aortic dimensions of adults with MFS. Our study is the first to demonstrate the prognostic value of TGF-β on aortic root growth and clinical events, such as aortic dissection and elective aortic root surgery. Plasma TGF-β might therefore serve as a prognostic biomarker in MFS.

62 60 Chapter 2 references 1. De Backer J, Loeys B, Leroy B, Coucke P, Dietz H, De Paepe A. Utility of molecular analyses in the exploration of extreme intrafamilial variability in the Marfan syndrome. Clin Genet Sep 2007;72(3): Dietz HC, Cutting GR, Pyeritz RE, et al. Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature Jul ;352(6333): De Witte P, Aalberts JJ, Radonic T, et al. Intrinsic biventricular dysfunction in Marfan syndrome. Heart Dec 2011;97(24): Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart Sep 2006;92(9): Meijboom LJ, Groenink M, van der Wall EE, Romkes H, Stoker J, Mulder BJ. Aortic root asymmetry in marfan patients; evaluation by magnetic resonance imaging and comparison with standard echocardiography. Int J Card Imaging Jun 2000;16(3): Meijboom LJ, Timmermans J, Zwinderman AH, Engelfriet PM, Mulder BJ. Aortic root growth in men and women with the Marfan s syndrome. Am J Cardiol Nov ;96(10): Groenink M, Lohuis TA, Tijssen JG, et al. Survival and complication free survival in Marfan s syndrome: implications of current guidelines. Heart Oct 1999;82(4): Engelfriet P, Mulder B. Is there benefit of beta-blocking agents in the treatment of patients with the Marfan syndrome? Int J Cardiol Jan ;114(3): Robinson PN, Arteaga-Solis E, Baldock C, et al. The molecular genetics of Marfan syndrome and related disorders. J Med Genet Oct 2006;43(10): Ramirez F, Dietz HC. Marfan syndrome: from molecular pathogenesis to clinical treatment. Curr Opin Genet Dev Jun 2007;17(3): Akhurst RJ, Hata A. Targeting the TGFbeta signalling pathway in disease. Nat Rev Drug Discov Oct 2012;11(10): Habashi JP, Doyle JJ, Holm TM, et al. Angiotensin II type 2 receptor signaling attenuates aortic aneurysm in mice through ERK antagonism. Science Apr ;332(6027): Neptune ER, Frischmeyer PA, Arking DE, et al. Dysregulation of TGF-beta activation contributes to pathogenesis in Marfan syndrome. Nat Genet Mar 2003;33(3): Ng CM, Cheng A, Myers LA, et al. TGF-beta-dependent pathogenesis of mitral valve prolapse in a mouse model of Marfan syndrome. J Clin Invest Dec 2004;114(11): Habashi JP, Judge DP, Holm TM, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science Apr ;312(5770): Matt P, Schoenhoff F, Habashi J, et al. Circulating transforming growth factor-beta in Marfan syndrome. Circulation Aug ;120(6): Radonic T, de Witte P, Baars MJ, Zwinderman AH, Mulder BJ, Groenink M. Losartan therapy in adults with Marfan syndrome: study protocol of the multi-center randomized controlled COM- PARE trial. Trials 2010;11: De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet Apr ;62(4): Hiratzka LF, Bakris GL, Beckman JA, et al ACCF/AHA/AATS/ACR/ ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with thoracic aortic disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovas-

63 Circulating transforming growth factor-β as a prognostic biomarker in Marfan syndrome 61 cular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation Apr ;121(13):e Vahanian A, Alfieri O, Andreotti F, et al. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J Oct 2012;33(19): Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr Dec 2005;18(12): Roman MJ, Devereux RB, Kramer-Fox R, O Loughlin J. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol Sep ;64(8): Murdoch JL, Walker BA, Halpern BL, Kuzma JW, McKusick VA. Life expectancy and causes of death in the Marfan syndrome. N Engl J Med Apr ;286(15): Groenink M, Rozendaal L, Naeff MS, et al. Marfan syndrome in children and adolescents: predictive and prognostic value of aortic root growth for screening for aortic complications. Heart Aug 1998;80(2): Groenink M, de Roos A, Mulder BJ, Spaan JA, van der Wall EE. Changes in aortic distensibility and pulse wave velocity assessed with magnetic resonance imaging following beta-blocker therapy in the Marfan syndrome. Am J Cardiol Jul ;82(2): Nollen GJ, Groenink M, Tijssen JG, van der Wall EE, Mulder BJ. Aortic stiffness and diameter predict progressive aortic dilatation in patients with Marfan syndrome. Eur Heart J Jul 2004;25(13): Silverman DI, Burton KJ, Gray J, et al. Life expectancy in the Marfan syndrome. Am J Cardiol Jan ;75(2): Mulder BJ. The distal aorta in the Marfan syndrome. Neth Heart J Nov 2008;16(11): Mimoun L, Detaint D, Hamroun D, et al. Dissection in Marfan syndrome: the importance of the descending aorta. Eur Heart J Feb 2011;32(4): Holm TM, Habashi JP, Doyle JJ, et al. Noncanonical TGFbeta signaling contributes to aortic aneurysm progression in Marfan syndrome mice. Science Apr ;332(6027): Haskett D, Doyle JJ, Gard C, et al. Altered tissue behavior of a non-aneurysmal descending thoracic aorta in the mouse model of Marfan syndrome. Cell Tissue Res Jan 2012;347(1): Matt P, Huso DL, Habashi J, et al. Murine model of surgically induced acute aortic dissection type A. J Thorac Cardiovasc Surg Apr 2010;139(4): Mu Y, Gudey SK, Landstrom M. Non-Smad signaling pathways. Cell Tissue Res Jan 2012;347(1): Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz III HC. Angiotensin II blockade and aorticroot dilation in Marfan s syndrome. N Engl J Med Jun ;358(26): Radonic T, de Witte P, Groenink M, et al. Inflammation aggravates disease severity in Marfan syndrome patients. PLoS One 2012;7(3):e

64

65 Chapter 3 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome Alexander W. den Hartog, Romy Franken, Piet de Witte, Teodora Radonic, Henk A. Marquering, Wessel E. van der Steen, Janneke Timmermans, Arthur J. Scholte, Maarten P. van den Berg, Aeilko H. Zwinderman, Barbara J.M. Mulder Maarten Groenink Adapted from: Radiology Nov;269(2):370-7

66 64 Chapter 3 abstract Purpose To assess reproducibility of aortic volume estimates and test its use serially in Marfan patients. materials and methods The study was approved by the medical ethics committee and all subjects gave written informed consent. In 81 Marfan patients, 7 healthy control subjects and in 22 Marfan patients at baseline and at 3 year follow-up, aortic volumes and diameters were estimated by contrast enhanced magnetic resonance imaging. Total aortic volume was defined as volume measurement starting at the level of the aortic annulus up to the aortic bifurcation. Intra- and interobserver agreement of aortic volume calculation was calculated by intraclass correlation coefficients (ICC). Differences in variables were analyzed with Student s t-test and logistic regression. Effect size was calculated. results Intra- and interobserver agreement of aortic volume calculation was respectively ICC=0.996 and ICC= Mean aortic volume was significantly greater in Marfan patients than in controls, respectively 104 ml/m² (95 % CI: ml/m²) versus 74 ml/m² (95 % CI: ml/m²), p< In 22 Marfan patients, mean aortic volume increased in 3 years with 17 ml (95 % CI: 8-26 ml, p=0.001, effect size 0.29), while mean aortic diameter did not increase significantly (0.4 mm, 95 % CI: mm, p=0.171, effect size 0.13). conclusion Assessment of aortic volume is highly reproducible and may be suited for the detection of aortic expansion in patients with Marfan syndrome.

67 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 65 introduction Marfan syndrome (MFS) is an autosomal dominant systemic connective-tissue disorder caused by mutations in the fibrillin-1 gene with an incidence of approximately 2-3 per individuals.(1) In patients with MFS the aorta gradually dilates, ultimately leading to aortic aneurysm formation and aortic dissection. Aortic complications are the leading cause of morbidity and mortality in patients with MFS.(2;3) With the introduction of prophylactic aortic root surgery at an aortic root diameter of cm, type A aortic dissection can be prevented effectively and life expectancy has increased significantly. (4-6) Prolonged longevity in MFS has shown that aortic complications are not limited to the aortic root and can occur throughout the entire aorta.(7) Therefore, it is recommended to visualize the total aorta by magnetic resonance imaging (MRI) or computed tomography (CT) at regular intervals.(8) However, complications in the aortic trajectory beyond the aortic root proof to be very difficult to foresee by measuring diameters.(9) Therefore, aortic diameter assessment in the aortic trajectory without localized aneurysms may not be a good marker for the extent of aortic disease. Volume measurement of the aorta could potentially reflect the extent of aortic disease more sensitively. The purpose of our study is to assess reproducibility of aortic volume estimates and test its use serially in Marfan patients. 3 materials and methods Analysis software for evaluation of functionality used in this study was 3Mensio Vascular (3Mensio Medical Imaging, Bilthoven, the Netherlands) and was provided without cost. Authors had full control of the data and information at all times study subjects The study was approved by the Academic Medical Center medical ethics committee and all participating subjects gave written informed consent. Between January 2008 and March 2010 total aortic volume and aortic diameters were measured in 81 patients with MFS by contrast enhanced three dimensional magnetic resonance imaging (3D-MRI). All patients were participants of the COMPARE study, which studies the effect of losartan on the aortic expansion rate in patients with MFS.(10) The inclusion criteria were diagnosis of MFS according to the Ghent criteria,(11) aortic root diameter < 50 mm, a maximum of 1 vascular prosthesis, absence of aortic dissection and an age of 18 years or older. Twenty patients had prophylactic aortic root replacement before the study started and eight patients underwent aortic root replacement during the study. All twenty-eight patients were operated due to progressive aortic root dilatation. Medical history and

68 66 Chapter 3 standardized measurements of height and weight were obtained from patients medical charts. See Table 1 for patient characteristics. Total aortic volumes and aortic diameters of a subpopulation of 22 MFS patients without a history of aortic surgery were compared with those of seven sex and age matched controls in whom aortic pathology was definitely ruled out. The 22 MFS patients underwent a second contrast enhanced 3D-MRI after 3 ± 0.2 years of follow up. In this follow up cohort aortic expansion rate measured by means of aortic diameter and aortic volume was calculated. magnetic resonance imaging All MRI scans were obtained in a single center (AMC, Amsterdam, The Netherlands) and acquired by a standardized commercial available protocol with a 1.5 Tesla MR system (Avanto, Siemens, Erlangen, Germany) using a phased array cardiac receiver coil. Contrast enhanced magnetic resonance angiography (MRA) of the total aorta was performed by first pass imaging of 0.2 ml/kg body weight contrast bolus of gadovist (Bayer Schering AG, Berlin, Germany) with a molarity of 1 mmol/l. Contrast was injected intravenously in the brachial vein at an infusion rate of 2 ml/s, and subsequently flushed by 20 ml saline at 2 ml/s, using contrast power injectors (Mallinckrodt Inc., St.Louis, MO, USA). Breath-holding at end-inspiration was performed. Contrast enhanced MRA was obtained by imaging the aorta continuously and subsequently triggered by scout imaging of the contrast bolus in the ascending aorta and aimed to visualize the total aorta during first pass of the contrast bolus in the aorta. The contrast enhanced MRA of the full aorta was acquired by means of a non-ecg gated 3D, T1-weighed, spoiled gradient-echo sequence (FLASH3D: Flip Angle 25 degrees, Field of View mm, matrix 348x227, 80 slice/slab, phase percentage 60-80%, 10% slice oversampling) with correction for gradient non-linearity and without zero-filling. This resulted in a near isotropic resolution of 1.4x1.3x1.4 mm/voxel. In order to perform valid volume assessment of the total aorta, (starting at the level of the aortic annulus and ending at the level of the aortic bifurcation) it was mandatory to visualize both the carotid vessels and iliac arteries on the contrast enhanced MRA. Aortic root size was measured by cine imaging sequences (Steady State Free Precession, SSFP) perpendicular to the long axis of the aortic root as shown by coronal and sagittal scouts (Slice thickness 6 mm, Flip Angle 80 degrees, Field of View 340 mm, matrix 256 x 192, frames per cardiac cycle, resulting in spatial resolution of 1.8x1.8x6.0 mm/voxel) during end-expiration. image processing The datasets were evaluated independently in a random order by two observers (AdH, PdW), both with two years of experience in cardiovascular imaging. Evaluation of data was supervised by cardiologist (MG) with over 15 years of experience in MFS and cardio-

69 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 67 vascular imaging. Vessel analysis software (3mensio vascular, 3mensio Medical Imaging BV, Bilthoven, the Netherlands) was used to calculated aortic volumes.(12) Images were loaded in the software with window and level settings acquired from the DICOM data. Total aortic volume was determined by the following technique; a central lumen line was created by manually placing a seeding point through the lumen of the aorta in the axial, the sagittal and the coronal plane.(figure 1) A complete set of multi-planar reformats were reconstructed by the computer perpendicular to this central lumen line, resulting in a stretched vessel view of the aorta, from the aortic valve to the aortic bifurcation. The aortic lumen was manually separated from the surrounding tissue by placing in four cross-sections a cut off line between the enhanced aortic lumen voxels and the surrounding voxels.(figure 1) The volume of the contrast-enhanced aortic lumen was reconstructed from the individually segmented axial slices. Aortic diameters were assessed at six landmark levels; the aortic root (greatest diameter of three cusp - cusp dimensions in diastole), the ascending and descending thoracic aorta at the level of the pulmonary bifurcation, the aortic arch, at the level of the diaphragm and the abdominal aorta. Aortic diameters beyond the aortic root were measured with electronic calipers perpendicular to the aortic lumen (inner wall to inner wall) in the long axis of a maximum intensity projecting of a contrast enhanced MR image. (Figure 2a and 2b) Where in doubt, multi-planar reconstructions were used to assess diameter. Mean aortic diameter was calculated as the average of these six measurements. Intra- and interobserver agreements of aortic volume and diameter were determined by analyzing 10 individual datasets twice by one observer (WvdS) and once by another observer (PdW). Repeated measurements were obtained in a random order without knowledge of the previous values. 3 statistical methods Statistical analysis was performed by using SPSS (SPSS, release 17.0 for Windows; SPSS, Chicago, IL). Normally distributed quantitative data are expressed as mean and 95 % CI. Intra- and interobserver agreements of aortic diameter and total aortic volume were determined with intraclass correlation coefficients (ICC) and processed in a Bland-Altman plot. Correlations in patients characteristics were sought with Pearson s correlation coefficient. Differences in variables within the group were compared with Student s t-test and logistic regression. Effect sizes of aortic diameter and aortic volume to detect aortic growth were calculated as the ratios of the mean difference between baseline and follow-up measurements divided by the pooled standard deviation. Effect sizes were compared using a t-test, which is described in Appendix I. P-values smaller than 0.05 were considered to be statistically significant.

70 68 Chapter 3 figure 1 View of the aorta during processing View of post processing software where seeding points are placed manually in the axial, the sagittal and the coronal plane of the contrast enhanced lumen of the aorta. Next to it a multi-planar reconstructed stretched view of an aorta with manually placed cut off lines.

71 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 69 figure 2a Maximum intensity projecting showing the aorta of a patient with Marfan syndrome At indicated levels aortic diameter was calculated. A = Ascending aorta B = Arch of the aorta C = Descending aorta D = Aorta at level of the diaphragm E = Abdominal aorta 3 figure 2b MR image showing an aortic root in the short axis of a patient with Marfan syndrome Greatest aortic root diameter was used of 3 measured distances. 1 = right coronary cusp left coronary cusp 2 = non coronary cusp left coronary cusp 3 = non coronary cusp right coronary cusp results Results are shown in Table 1. Mean aortic volume and mean aortic diameter of the 81 MFS patients were respectively 231 ml (95 % CI: ml) and 25.7 mm (95 % CI: mm). No aortic aneurysms between measured levels were detected. Intra-and interobserver variability of aortic volume determined in 10 MFS patients were respectively 4.63 ml (SD=4.82 ml, ICC=0.996) and 0.06 ml (SD=11.96 ml, ICC=0.980) and were as reproducible as mean aortic diameter measurements, respectively 0.2 mm (SD=0.38 mm, ICC=0.974) and 0.03 mm (SD=1.0 mm, ICC=0.901).(Table 2, Figure 3a and 3b)

72 70 Chapter 3 table 1 Patient characteristics marfan patients n 81 Age (years) 34 (32-37) Male (%) 43 (53) Length (cm) 188 ( ) Weight (kg) 77 (73 80) Body Surface Area (m2) 2.0 ( ) Prior aortic root replacement (%) 20 (25) aortic volume (av) - AV normalized for BSA (ml/m²) 116 ( ) - AV normalized for BSA in males (ml/m²) 119 ( ) - AV normalized for BSA in females (ml/m²) 111 ( ) aortic diameter normalized for bsa - Aortic root (mm/m²) 20.9 ( ) - Ascending aorta (mm/m²) 14.4 ( ) - Aortic arch (mm/m²) 12.0 ( ) - Proximal Descending aorta (mm/m²) 11.9 ( ) - Aorta at level of Diaphragm (mm/m²) 10.3 ( ) - Abdominal aorta (mm/m²) 8.4 ( ) - Mean aortic diameter (mm/m²) 12.9 ( ) Values are expressed as mean and (95 % CI) or n (%), n denotes number of patients table 2 Intra-observer and inter-observer variability of aortic volume and aortic diameter of 10 patients with Marfan syndrome intra-observer variability icc aortic: measurement 1 measurement 2 -Volume (ml) 219 ± ± Mean diameter (mm) 28.8 ± ± inter-observer variability observer 1 observer 2 -Volume (ml) 186 ± ± Mean diameter (mm) 28.8 ± ± Values are expressed as mean ± SD or n (%), ICC denotes intraclass correlation coefficient Aortic volume and mean aortic diameter were associated with Body Surface Area (BSA), respectively R=0.43, p<0.001 for both.(figure 4a and 4b) The association of aortic volume and age was comparable with the association of mean aortic diameter and age, respectively R=0.55 and R=0.56, p<0.001 for both.

73 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 71 figure 3a Intra-observer variability of aortic volume measurements in 10 patients with Marfan syndrome 3 figure 3b Inter-observer variability of aortic volume measurement in 10 patients with Marfan syndrome Aortic volume corrected for BSA and mean aortic diameter corrected for BSA in 22 Marfan patients were significantly greater than in seven sex and age-matched controls, respectively 104 ml/m² (95 % CI: ml/m²) versus 74 ml/m² (95 % CI: ml/ m²), p<0.001 and 12.5 mm/m² (95 % CI: mm/m²) versus 11.4 mm/m² (95 % CI: mm/m²), p=0.042.(table 3) Multiple logistic regression analysis showed that aortic volume corrected for BSA could discriminate between MFS patients and controls, while mean aortic diameter corrected for BSA could not, respectively OR (95 % CI: , p=0.003) vs OR (95 % CI: , p=0.167). In 22 patients with MFS, mean aortic volume increased in 3 years with 17 ml (95 % CI: 8-26 ml, p=0.001, effect size 0.29), while mean aortic diameter did not increase significantly, respectively 0.4 mm (95 % CI: mm, p=0.171, effect size 0.13).(Table 4) Mean aortic diameter corrected for BSA in Marfan patients with prior aortic root replacement was comparable with Marfan patients without aortic root replacement,

74 72 Chapter 3 figure 4a Correlation between body surface area and aortic volume in 81 patients with Marfan syndrome. figure 4b Correlation between body surface area and mean aortic diameter in 81 patients with Marfan syndrome. respectively 12.5 mm/m² (95 % CI: mm/m²) versus 13.1 mm/m² (95 % CI: mm/m²), p=0.238, while total aortic volume corrected for BSA was greater in MFS patients with prior aortic root replacement as compared to patients without aortic root replacement, respectively 127 ml/m² (95 % CI: ml/m²) versus 112 ml/m² (95 % CI: ml/m²), p=0.040.(table 5)

75 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 73 table 3 Aortic dimensions of 22 patients with Marfan syndrome and 7 controls marfan controls P- value n 22 7 Age (years) Male (%) 12 (55) 5 (71) Body Surface Area (m²) 1.95 ( ) 1.96 ( ) Aortic diameter normalized for BSA - Aortic root (mm/m²) 22.9 ( ) 16.5 ( ) < Ascending aorta (mm/m²) 13.2 ( ) 13.7 ( ) Aortic arch (mm/m²) 11.0 ( ) 11.7 ( ) Proximal Descending aorta (mm/m²) 11.2 ( ) 10.3 ( ) Aorta at level of Diaphragm (mm/m²) 9.3 ( ) 9.3 ( ) Abdominal aorta (mm/m²) 7.7 ( ) 6.9 ( ) Mean aortic diameter (mm/m²) 12.5 ( ) 11.4 ( ) Aortic volume normalized for BSA (ml/m²) 104 (95 114) 74 (62 87) <0.001 Values are expressed as mean and (95 % CI) or n (%), n denotes number of patients table 4 Aortic growth of 22 patients with Marfan syndrome in 3 years aortic dimensions (mm) baseline follow up growth effect size - Aortic root 44.3 ( ) 44.9 ( ) 0.7 ( ) Ascending aorta 25.6 ( ) 26.6 ( ) 0.9 ( ) Aortic arch 21.4 ( ) 21.4 ( ) 0.0 ( ) Proximal Descending aorta 21.7 ( ) 22.0 ( ) 0.3 ( ) Aorta at level of Diaphragm 18.0 ( ) 18.4 ( ) 0.4 ( ) Abdominal aorta 15.0 ( ) 15.4 ( ) 0.4 ( ) Mean aortic diameter 24.3 ( ) 24.6 ( ) 0.4 ( ) 0.13 Aortic Volume (ml) 203 ( ) 220 ( ) 17 (8 26) 0.29 Values are expressed as mean and (95 % CI) table 5 Aortic volume in 81 patients with and without aortic root replacement aortic root replacement P- value Yes (n=20) No (n=61) Age (years) Male (%) 13 (65) 30 (49) aortic: -Volume (ml) 257 ( ) 223 ( ) Volume normalized for BSA (ml/m²) 127 ( ) 112 ( ) Mean Aortic Diameter (mm) 25.2 ( ) 25.9 ( ) Mean Aortic Diameter normalized for BSA (mm/m²) 12.5 ( ) 13.1 ( ) Values are expressed as mean and (95 % CI) or n (%), n denotes number of patients

76 74 Chapter 3 discussion In this study we have shown good reproducibility of total aortic volume assessment. A significant difference in aortic volume between patients with global aortic pathology and without aortic pathology could be demonstrated. Furthermore, we demonstrated a superior effect size of aortic volume assessment over diameter assessment in total aortic expansion rate by serial imaging at three years follow-up. Reproducibility of aortic volume assessment in our study was comparable to reproducibility of volume measurements in abdominal aneurysms.(13-15) However, these studies used CT images for aortic volume rendering. MRI offers the advantage of lack of ionizing radiation exposure, important in relatively young MFS patients who will be receiving frequent examinations during their lifetime. In daily clinical practice, Marfan patients undergo MRI of the entire aorta at regular intervals, as advocated by current guidelines.(16;17) Generally, aortic diameter is measured at several aortic levels to quantify the extent of local aortic disease. Comparison of aortic diameter at several levels in MFS patients with a limited number of controls (Table 3) did only show significant differences at the level of the aortic root, which is a major clinical feature of MFS. Still, mean aortic diameter remained significantly larger in MFS, although the differences in aortic root size were somewhat averaged out (Table 3). In 3 years time, the only significant increase in aortic diameter in MFS patients could be shown at the level of the ascending aorta (Table 4). Although we do not advocate follow up of known aortic aneurysms by this method, we assessed the extent of total aortic disease in MFS patients by estimates of mean aortic diameter and total aortic volume. This was performed in MFS patients without localized aneurysms beyond the aortic root. Management of localized aortic aneurysms (as in the aortic root) is still based on the measurement of local aortic dimensions. However, a measure of total aortic size is relevant for estimation of extent and progression of aortic disease, both in clinical studies and clinical practice. Aortic volume measurement could limit follow-up time and sample size in clinical trials where the effect of medicinal therapy on aortic dilatation rate is assessed or to estimate the amount of aortic disease beyond the aortic root.(10) The mean aortic diameter is a mere approximation of total aortic volume, because of a very limited sample size (six in case of mean aortic diameter and almost unlimited in case of volume assessment). Hence, no statistically significant differences in mean aortic diameter between MFS patients and controls could be demonstrated. On the contrary, by means of aortic volume assessment, a statistically significant difference between MFS patients and controls was shown. Repeated assessment of aortic diameters, especially beyond the aortic root, do not often show an increase in aortic size. Accordingly, total aortic expansion rate measured by mean aortic diameter was very small (0.1 mm/year)

77 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 75 in our study and could easily be attributed to patient-, equipment- or technician dependent factors. Total aortic expansion rate became more obvious when aortic volume was assessed (5.6 ml/year). Aortic volume in operated MFS patients was significant different from non-operated patients, which could not be shown by aortic diameter measurement. MFS patients with prophylactic aortic root surgery have a more complicated clinical course.(7) However, complications in the aortic trajectory beyond the aortic root proof to be very difficult to predict by measuring aortic diameters.(18) Future studies are needed to assess the predictive value of aortic volume on aortic complications, such as aortic dissection beyond the aortic root. 3 study limitation Till date, reference values for total aortic volume have not previously been established in the literature. Therefore, the main limitation of our study is the absence of an independent reference standard for aortic volume. Furthermore, a control group of 7 individuals to calculate aortic volume must be regarded as very small. reference list 1. Judge DP, Dietz HC. Marfan s syndrome. Lancet 2005 Dec 3;366(9501): Chan YC, Ting CW, Ho P, Poon JT, Cheung GC, Cheng SW. Ten-year epidemiological review of inhospital patients with Marfan syndrome. Ann Vasc Surg 2008 Sep;22(5): Murdoch JL, Walker BA, Halpern BL, Kuzma JW, McKusick VA. Life expectancy and causes of death in the Marfan syndrome. N Engl J Med 1972 Apr 13;286(15): Jondeau G, Detaint D, Tubach F, Arnoult F, Milleron O, Raoux F, et al. Aortic event rate in the Marfan population: a cohort study. Circulation 2012 Jan 17;125(2): Gott VL, Greene PS, Alejo DE, Cameron DE, Naftel DC, Miller DC, et al. Replacement of the aortic root in patients with Marfan s syndrome. N Engl J Med 1999 Apr 29;340(17): Gott VL, Cameron DE, Alejo DE, Greene PS, Shake JG, Caparrelli DJ, et al. Aortic root replacement in 271 Marfan patients: a 24-year experience. Ann Thorac Surg 2002 Feb;73(2): Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart 2006 Sep;92(9): Kauffmann C, Tang A, Therasse E, Giroux MF, Elkouri S, Melanson P, et al. Measurements and detection of abdominal aortic aneurysm growth: Accuracy and reproducibility of a segmentation software. Eur J Radiol 2012 Aug;81(8): Trimarchi S, Jonker FH, Hutchison S, Isselbacher EM, Pape LA, Patel HJ, et al. Descending aortic diameter of 5.5 cm or greater is not an accurate predictor of acute type B aortic dissection. J Thorac Cardiovasc Surg 2011 Sep;142(3):e101-e107.

78 76 Chapter Radonic T, de Witte P., Baars MJ, Zwinderman AH, Mulder BJ, Groenink M. Losartan therapy in adults with Marfan syndrome: study protocol of the multi-center randomized controlled COM- PARE trial. Trials 2010;11: De Paepe A., Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996 Apr 24;62(4): van Prehn J., van der Wal MB, Vincken K, Bartels LW, Moll FL, van Herwaarden JA. Intra- and interobserver variability of aortic aneurysm volume measurement with fast CTA postprocessing software. J Endovasc Ther 2008 Oct;15(5): Kauffmann C, Tang A, Dugas A, Therasse E, Oliva V, Soulez G. Clinical validation of a software for quantitative follow-up of abdominal aortic aneurysm maximal diameter and growth by CT angiography. Eur J Radiol 2011 Mar;77(3): Wever JJ, Blankensteijn JD, Th MMW, Eikelboom BC. Maximal aneurysm diameter follow-up is inadequate after endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2000 Aug;20(2): Parr A, Jayaratne C, Buttner P, Golledge J. Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography. Eur J Radiol 2011 Jul;79(1): Baumgartner H, Bonhoeffer P, De Groot NM, de HF, Deanfield JE, Galie N, et al. ESC Guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J 2010 Dec;31(23): (17) Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE, Jr., et al ACCF/AHA/AATS/ ACR/ASA/SCA/SCAI/SIR/STS/SVM Guidelines for the diagnosis and management of patients with thoracic aortic disease. A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology,American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons,and Society for Vascular Medicine. J Am Coll Cardiol 2010 Apr 6;55(14):e27-e Trimarchi S, Jonker FH, Hutchison S, Isselbacher EM, Pape LA, Patel HJ, et al. Descending aortic diameter of 5.5 cm or greater is not an accurate predictor of acute type B aortic dissection. J Thorac Cardiovasc Surg 2011 Sep;142(3):e101-e107.

79 Aortic volume assessment for surveillance of aortic disease in Marfan syndrome 77 appendix i - definition and comparison of the effect sizes of aortic diameter and aortic volume growth. Because aortic diameters and aortic volumes are measured on different scales we calculated the extent of change as the ratio of the difference between the baseline and follow-up means over the pooled standard deviation of the volumes/diameters. Thus, the statistic then equals 3 Where is the mean at baseline, is the mean at follow up and S the (pooled) standard deviation. In that case we may write the variance of d as follows where n is the sample size, and r is the correlation between baseline and follow up measurements. To compare the effect sizes of the diameters, ddiameter, and of the volumes, dvolume, we calculated the test statistic where the variance of the difference, var(ddiameter-dvolume), equals var(ddiameter-dvolume)=2*{(1-rdiameter)+(1-rvolume)-(rd,v-r d,v)} / n and where rdiameter is the correlation between baseline and follow-up diameters, rvolume is the correlation between baseline and follow-up volumes, rd,v is the (pooled) correlation between diameter and volume at the same time (either at baseline or at follow-up) and r d,v is the correlation between volume and diameter measured at different times. We assumed that our t-statistic followed approximately a normal distribution.

80

81 Chapter 4 The risk for type B aortic dissection in Marfan syndrome Alexander W. den Hartog, Romy Franken, Aeilko H. Zwinderman, Janneke Timmermans, Arthur J. Scholte, Maarten P. van den Berg, Vivian de Waard, Gerard Pals, Barbara J.M. Mulder, Maarten Groenink Adapted from: Journal of the American College of Cardiology Jan;65(3):246-54

82 80 Chapter 4 abstract background Aortic complications beyond the ascending aorta have become a major clinical problem in patients with Marfan syndrome. objectives To identify clinical parameters associated with type B aortic dissection and to develop a risk model to predict type B aortic dissection in patients with Marfan syndrome. methods Marfan patients attending a Dutch Marfan clinic were recruited. Association of type B aortic dissection with several clinical parameters was assessed. A risk model was developed to predict type B dissection. results Between 1998 and 2013, 54 type B aortic dissections occurred in 600 Marfan patients (36±14 years, 52% male). Independent variables associated with type B aortic dissection were (A) prior prophylactic aortic surgery (HR:2.1; 95%CI: ; p=0.010), and (B) a proximal descending aorta diameter 27mm (HR:2.2; 95%CI: ; p=0.020). In our risk model, the 10 years occurrence of type B aortic dissection in the low, moderate, and high risk patients was 6%, 19%, and 34%, respectively. Angiotensin-II receptor blocker therapy was associated with less type B aortic dissections (HR:0.3; 95%CI: ; p=0.030). conclusions Marfan patients with prior prophylactic aortic surgery are at substantial risk for type B aortic dissection, even when the descending aorta is only slightly dilated. Angiotensin-II receptor blocker therapy may be protective in the prevention of type B aortic dissections.

83 The risk for type B aortic dissection in Marfan syndrome 81 introduction Life expectancy of patients with Marfan syndrome (MFS) has improved due to a more aggressive surgical approach in prophylactic aortic root and ascending aorta replacement (AoRR).(1) However, increased longevity of MFS patients has revealed an increased occurrence of complications beyond the ascending aorta (hereafter mentioned as distal aorta ).(2;3) Current surgical guidelines advocate prophylactic surgery of the distal aorta when the diameter exceeds 50mm.(4;5) However, type B aortic dissection, currently defined as aortic dissection not involving the ascending aorta, occurs frequently without significant aortic dilatation in MFS.(2;6) In the present study we investigated the association of type B aortic dissection in MFS patients with clinical parameters, including genetic analysis, aortic dimensions and elastic properties. Furthermore, we established a risk model to predict the occurrence of type B aortic dissection in adult MFS patients. 4 methods study design and participants We included all adult patients who were diagnosed with MFS by a multidisciplinary Marfan screening team according to the Revised Ghent criteria.(7) Included patients had to attend one of the four Dutch university Marfan screening clinics between 1998 and 2013, in which period one or more scans of the total aorta should have been acquired either by means of Magnetic Resonance Imaging (MRI) or Computed Tomography (CT). Patients with evidence of aortic dissection on the first available MRI or CT images were excluded. Starting point of the study was defined as the date of the first available aortic images (MRI or CT). Endpoints were the occurrence of type B aortic dissection or the combined endpoint, defined as the first of the following events: type B aortic dissection, distal aortic surgery or death. Type B aortic dissection was defined as a dissection of the descending aorta without involvement of the ascending aorta, which had to be confirmed by MRI or CT.(8;9) Patients in whom type A aortic dissection (with involvement of the ascending aorta) occurred during the study, were censored. Patient demographics, medicinal treatment, surgical history, the presence of a FBN1 mutation, date of type B aortic dissection and family history were obtained from patients medical charts. In patients negatively tested for a FBN1 mutation, the TGFBR1, TGFBR2, TGFB2, MYH11, MYLK1, SMAD3, ACTA2 genes were subsequently screened. Patients with a mutation leading to another connective tissue disease than MFS, such as Loeys-Dietz syndrome, were excluded from analysis. Aortic diameters were measured from the first and last available MRI and/or CT scan during follow-up and aortic dilatation rates were calculated. The

84 82 Chapter 4 study was conducted in accordance with all human research regulatory guidelines and the institutional conduct code for health related research. image analysis Standard, commercially available, non ECG gated imaging techniques were used for aortic imaging: three dimensional Gadolinium enhanced spoiled gradient echo for MRI and contrast enhanced imaging for CT. In a subset of patients, ECG gated MR CINE imaging was performed perpendicular to the descending aorta at the level of the pulmonary artery as described earlier.(10) Diameter measurements were performed by one analyst (AWdH) on multiplanar MRI/CT reconstructions from inner to inner edge. The largest diameter was measured in the following aortic segments: (1) from the origin of the brachiocephalic trunk to the origin of the subclavian artery (aortic arch), (2) from the origin of the subclavian artery to the diaphragm (descending thoracic aorta), and (3) from the diaphragm to the aortic bifurcation (abdominal aorta).(figure 1) The frequency of the following aortic characteristics was determined: (A) global dilatation of the descending aorta (defined as an aortic diameter in 2 segments above the normal mean(11;12)), (B) aortic diameter above the upper limit of normal ( 27mm(11)) and aortic diameter above the optimal cut-off point derived from the ROC curve, (C) presence of an aortic hump (defined as a local increase of 1 mm in aortic diameter within a segment).(figure 3) Aortic distensibility was calculated by one analyst (AWdH) from MRI CINE images and noninvasively measured blood pressure during MR imaging, as described earlier.(13) figure 1 This image shows a 3D MR image of the aorta in the long axis of a patient with Marfan syndrome statistical analysis Data analysis was performed using the SPSS statistical package (20.0 for windows; SPSS Inc., Chicago, Illinois, USA). Continuous variables were expressed as mean ± standard deviation (SD). Categorical variables were summarized as number with percentage. Cox proportional hazard analysis was used to identify determinants of type B aortic dissection or the combined outcome. The risks were expressed as hazard ratios (HR) with

85 The risk for type B aortic dissection in Marfan syndrome 83 corresponding 95% confidence intervals (95%CI). Quantitative risk factors were dichotomized. Relevant cut-off values were obtained by plotting a ROC curve and assessing the area under the curve (AUC) at fixed times points during follow-up. Aortic diameter was dichotomized depending on the upper limit of the normal range (<27mm or 27mm (11)). Aortic dilatation rate and distensibility were dichotomized using a threshold maximizing the concordance statistic of the Cox regression model. Two analyses were performed. In the first analysis patients were followed from the date of the first available MRI/CT images onwards until the occurrence of an event or end of follow-up. In the type B aortic dissection analysis, MFS patients who died without the occurrence of type B aortic dissection were censored at date of death. Prophylactic aortic surgery during follow-up and aortic dilatation rate was used as a time dependent covariate. In the second analysis, patients were followed from the date of the last available MRI/ CT imaging onwards. In this second analysis, prophylactic aortic surgery occurring before the date of the last available MRI/CT images, and aortic dilatation rate, were included as baseline predictor variables. Variables with a p-value 0.20 in univariable analyses were considered in multivariable models. Missing data were imputed five times and the regression parameters were estimated by the mean of the 5 imputation results. The multivariable Cox model on the selected determinants was analyzed with a forward conditional algorithm on each of the 5 imputed datasets. The final model included all predictor variables selected in at least one of the imputed datasets. 4 risk model A risk model to predict type B aortic dissection in MFS patients from the date of the first available aortic images onwards was developed. In the risk model, clinical variables that were independently associated with type B aortic dissection were included. We did not consider medicinal treatment as a prognostic variable in this risk model. To facilitate quick risk stratification we simplified the regression model by rounding the estimated regression parameters to the nearest integer values. Consequently, the risk score of an individual patient equals the number of risk factors in the patient, weighted with a simple integer value. To assess the discriminatory ability of the risk model we compared the Kaplan-Meier curves of the cumulative type B aortic dissection risk in the different risk strata (% with standard error). We also calculated the c-statistic of the risk model.(14)

86 84 Chapter 4 results Patients A total of 646 adult MFS patients visited one of the participating centres between 1998 and Of these 646 MFS patients, 46 patients were excluded due to a type A aortic dissection prior to first available aortic imaging by MRI or CT. No aortic arch dissections occurred. In the remaining 600 adult MFS patients (36±14 years, 52% male) aortic diameters, dilatation rate and distensibility were available in respectively 532 (89%), 423 (71%) and 140 (23%) patients. Aortic imaging by MRI was available in 500 patients (94%) and by CT in 32 patients (6%). FBN1 mutation analysis was available in 542 patients (90%) and a FBN1 mutation was found in 446 patients (83%). At baseline, 1 prophylactic aortic surgery had been performed in 143 MFS patients (24%; prophylactic AoRR n=134; distal prophylactic aortic surgery n=2; prophylactic AoRR and distal aortic surgery n=7). Baseline characteristics are summarized in Table 1. outcomes Type B aortic dissection During a median follow-up of 6.0 years from first available images onwards (3639 personyears), 2 type A aortic dissections (0.05%/year) and 54 type B aortic dissections (1.5%/ year) occurred. Sixty-six patients underwent prophylactic aortic surgery (1.8%/year; prophylactic AoRR n=52, prophylactic distal aortic surgery n=13; prophylactic AoRR and distal aortic surgery n=1). Out of a total of 9 deaths (0.25%/year), 5 were MFS related. Two patients died as a result of multiple distal aortic surgeries after type B aortic dissection, one patient died as a result of heart failure following multiple aortic surgeries after type A aortic dissection, one patient died as a result of heart failure with multi organ failure 9 months after re-operation of the AoRR, and 1 patient refused treatment before and after type B aortic dissection and subsequently died. The remaining 4 patients died from intestinal surgery, traumatic brain injury, neurological disorder, or an unknown cause. In 48 patients (89%) the origin of the type B aortic dissection was the proximal descending aorta and in 6 patients (11%) the origin was either the distal descending thoracic aorta or the abdominal aorta. In 30 MFS patients (56%), AoRR had been performed before type B aortic dissection occurred. In the remaining 24 MFS patients (44%), type B aortic dissection was the first aortic complication. In univariate Cox analysis, type B aortic dissection was associated with prior prophylactic aortic surgery, greater aortic diameter in all segments, decreased distensibility and global aortic dilatation (Table 1). Faster aortic dilatation rate was also associated with type B aortic dissection.(table 2) The optimal cut-off values to discriminate between high and low risk for type B aortic dissection were: a descending thoracic aortic diameter 27mm, an aortic dilatation rate of 0.5 mm/year and aortic distensibility of <

87 The risk for type B aortic dissection in Marfan syndrome 85 table 1 Baseline characteristics and associations with type B aortic dissection of 600 Marfan patients at the time of the first available aortic images total cohort b dissection no b dissection variable Number of patients univariable analysis multivariable analysis hr 95% ci p-value hr 95% ci p-value Age (years) 36 ± ± ± Male 312 (52) 30 (57) 282 (52) BSA (m 2 ) 2.0 ± ± ± FBN1 mutation 447 (83) 36 (86) 411 (82) aortic surgery # 143 (24) 27 (51) 116 (21) < AoRR 141 (24) 27 (51) 114 (21) <0.001 Age at AoRR 32 ± ± ± distal aortic surgery 9 (2) 4 (7) 5 (1) <0.001 Arch + prox desc 3 (1) 2 (4) 1 (<1) replacement Mid-thoracic 1 (<1) 1 (2) 0 replacement Bifurcation replacement 5 (1) 1 (2) 4 (1) aortic diameters (mm) and characteristics (n=532) Segment I 24 ± 4 26 ± 8 24 ± <0.001 Segment II 25 ± 5 28 ± 7 24 ± <0.001 Segment II/BSA 12 ± 3 14 ± 4 12 ± n (%) patients with 163 (31) 16 (50) 147 (29) diameter above normal* Global descending 172 (32) 17 (53) 155 (31) dilatation Presence of aortic 256 (48) 18 (56) 238 (48) hump Hump in dilated aorta 226 (42) 21 (66) 205 (41) Segment III 21 ± 5 25 ± ± <0.001 aortic distensibility (10-3 mm/hg) (n=140) Proximal descending 3.5 ± ± ± n (%) patients with 45 (32) 6 (75) 39 (30) distensibility < 2.5 Diaphragm 6.0 ± ± ± Abdominal 6.3 ± ± ± Abbreviations: BSA = body surface area; AoRR = aortic root replacement Plus-minus values are means ± SD. Categorical variables were summarized as number (%) # At baseline, in 7 patients distal aortic surgery had been performed after AoRR (not included in 143) * 27 (Reference 10) 4

88 86 Chapter 4 mmhg -1 with an AUC of 0.839, and 0.747, respectively. The optimal cut-off value for aortic diameter of 27 mm corresponded with aortic diameter above the upper limit of normal.(11) Kaplan-Meier curves (Figure 2a) illustrate the risk for type B aortic dissection between patients with a descending thoracic aortic diameter larger or smaller than the upper limit of normal (27mm). Kaplan-Meier curves for aortic distensibility are shown in Figure 2b. No Kaplan-Meier curves could be constructed for aortic dilatation rate, because of its time depending nature. In multivariable Cox analysis, prior prophylactic aortic surgery and an increased diameter of the descending thoracic aorta above the normal upper limit ( 27mm) were associated with type B aortic dissection, HR:2.1; 95%CI: ; p=0.010 and HR:2.2; 95%CI: ; p=0.020, respectively.(table 1) During follow-up, dilatation rate of the descending thoracic aorta 0.5mm/year was also associated with type B aortic dissection, HR:2.4; 95%CI: ; p=0.015, respectively.(table 2) In the second analysis, aortic characteristics of the last available MRI/CT images onwards were used. The mean diameter of the descending thoracic aorta of MFS patients with a type B aortic dissection during follow-up, measured median 1.4 years prior to the event, was 31±7 mm. In patients without a type B aortic dissection during follow-up, mean diameter of the descending thoracic aorta, measured median 1.9 years prior to the end of the study, was 26±5 mm (p<0.001). Overall, the second analysis rendered similar results as the first analysis; prior prophylactic aortic surgery (HR:2.3; 95%CI: ; p=0.007) and enlarged aortic diameter in all segments (proximal descending aorta diameter 27mm HR:3.4; 95%CI: ; p=0.006) were associated with risk of type B aortic dissection.(table 3) In multivariable Cox analysis, dilatation rate of the aorta was not independently associated with risk of type B aortic dissection. Furthermore, analysis of medicinal treatment during the last available scan showed that in 146 patients using angiotensin II receptor blockers (ARBs), type B aortic dissection was less frequent as compared to 436 patients not on ARB therapyhr:0.3; 95%CI: ; p= In addition, in univariate analysis, a trend is visible of the potential detrimental effect of ACEi (HR: 2.4, p=0.072), but in multivariate analysis this trend is no longer present (HR=1.7, p=0.188) (Table 3). No association with blood pressure could be demonstrated (Table 3) Combined endpoint A total of 51 type B aortic dissections, 14 distal aortic surgeries and 6 deaths occurred as first event. In multivariable Cox analysis the following clinical parameters were associated with the combined endpoint: (1) prior prophylactic aortic surgery (HR:1.9; 95%CI: ; p=0.018), (2) a descending thoracic aorta 27mm (HR:2.1; 95%CI: ; p=0.015), and dilatation rate of the proximal descending thoracic aorta 0.5mm/year (HR:2.4; 95%CI: ; p=0.004).

89 The risk for type B aortic dissection in Marfan syndrome 87 figure 2a Kaplan-Meier curves Kaplan Meier curve showing percentage freedom of type B aortic dissection in Marfan patients with an aortic diameter of 27 mm and in patients with an aortic diameter of <27 mm at the level of the proximal descending aorta from first available aortic images. Numbers of patients at risk of type B aortic dissection are demonstrated below the figure. 4 figure 2b Kaplan-Meier curves Kaplan Meier curve showing percentage freedom of type B aortic dissection in Marfan patients with an aortic distensibility of mmhg -1 and in patients with an aortic distensibility of < mmhg -1 at the level of the proximal descending aorta from first available aortic images. Numbers of patients at risk of type B aortic dissection are demonstrated below the figure. risk score for type b aortic dissection Independent risk factors for type B aortic dissection at the first available aortic images onwards were prior prophylactic aortic surgery and a descending thoracic aorta diameter 27mm. Therefore, three subgroups of patients were created: (1) patients without risk factors, (2) patients with one risk factor, and (3) patients with both risk factors. Kaplan- Meier curves of the cumulative risk of type B aortic dissection are given in the Figure 3. After 10 years of follow-up the cumulative risks of type B aortic dissection were 6% (se 2%), 19% (se 4%), and 34% (se 8%), respectively for low (n=345), moderate (n=176), and high risk patients (n=79) (p < 0.001). Optimal corrected c-statistic of this model was 0.53.

90 88 Chapter 4 table 2 Aortic dilatation rate during follow-up and association with type B aortic dissection follow-up variables total cohort b dissection no b dissection dilatation rate (mm/year) univariable analysis Segment I 0.2 ± ± ± Segment II 0.4 ± ± ± <0.001 hr 95% ci p-value hr multivariable analysis Dilatation rate (32) 11 (50) 124 (31) < Segment III 0.3 ± ± ± Plus-minus values are means ± SD. Categorical variables were summarized as number (%) 95% ci p-value survival after type b aortic dissection Median follow-up after type B aortic dissection was 7 years. In this period, 25 patients were treated conservatively (41%), 29 patients had 1 surgical procedure (53%), and 3 patients died after type B aortic dissection (6%). discussion Herewith we provide a contemporary overview of clinical outcomes in patients with known MFS without prior aortic dissection in one of the largest MFS studies till date, using 3-dimensional imaging techniques. We have shown a type B aortic dissection rate of 9% during a median follow-up of 6 years. Type B aortic dissections generally occurred in mildly dilated proximal descending aortas, especially in patients with prior prophylactic aortic surgery. Furthermore, we were able to develop a risk score to predict type B aortic dissection in MFS patients based on history of prophylactic aortic root surgery and proximal descending aortic diameter. Apparently, the occurrence of type A aortic dissection has become a rare event in patients with known MFS in the era of aggressive prophylactic surgery, most likely due to the strict adherence to guidelines for prophylactic intervention in specialized Marfan Centers in the Netherlands. Although AoRR has improved life expectancy considerably, the disease process is uninterrupted and will continue to affect the remainder of the aorta and great vessels in the majority of patients with MFS. Replacement of the aortic root/ascending aorta with a stiff vascular prosthesis may result in higher pulsatile forces acting onto the native aortic arch and proximal descending aorta, which become the main constituents of remaining windkessel function. The fact that age at the time of prophylactic aortic root surgery was similar in both groups (31±15 years vs.33 ± 13 years)

91 The risk for type B aortic dissection in Marfan syndrome 89 4 figure 3 Kaplan-Meier curve and four types of Marfan aortas At the top the kaplan-meier curve based on our Risk model. Below the examples of 4 types of Marfan Aortas. Kaplan Meier curve showing percentage freedom of type B aortic dissection in the 3 different patient groups. Starting point of this Kaplan-Meier was defined as the time of first available aortic images. Numbers of patients at risk for type B aortic dissection in the different categories are shown below the figure.

92 90 Chapter 4 table 3 Patient characteristics and clinical parameters associated with type B aortic dissection by using the last available aortic images as starting point variable univariable analysis multivariable analysis mean hr 95% ci p-value hr 95% ci p-value number of patients 600 Age (years) 40 ± Male 312 (52) BSA (m 2 ) 2.0 ± FBN1 mutation 448 (75) blood pressure (mmhg) Mean arterial pressure 90 ± Systolic blood pressure 123 ± Pulse pressure 49 ± medication β - blocker 426 (73) ARB 146 (25) ACE-I 39 (7) CA-Blockers 20 (3) Diuretics 33 (6) aortic surgery # 198 (33) < AoRR* 191 (32) <0.001 Age at AoRR 32 ± Distal aortic surgery ** 18 (3) <0.001

93 The risk for type B aortic dissection in Marfan syndrome 91 table 3 Patient characteristics and clinical parameters associated with type B aortic dissection by using the last available aortic images as starting point (continued) variable univariable analysis multivariable analysis mean hr 95% ci p-value hr 95% ci p-value aortic diameters (mm) and characteristics Segment I 25 ± <0.001 Segment II 26 ± <0.001 Segment II/BSA 13 ± <0.001 n (%) patients with diameter above normal*** 220 (41) < Dilatation rate 0.5 mm/year 135 (32) Global descending dilatation 382 (72) Presence of aortic hump 281 (54) Hump in dilated aorta 242 (46) Segment III 22 ± <0.001 Abbreviations: AoRR = aortic root replacement; ARB = angiotensin receptor blocker; ACE-I = angiotensin-converting-enzyme inhibitor CA = calcium channel Plus-minus values are means ± SD. Categorical variables were summarized as number (%) A total of 11 patients underwent distal aortic surgery after AoRR (not included in 198) * In 3 MFS patients, AoRR occured after last aortic imaging (not included in 191) ** 5 Distal aortic surgeries occured after last aortic imaging (not included in 18) *** 27 (Reference 10) # 4

94 92 Chapter 4 seems to underscore the possibility that the procedure itself may be a risk factor for subsequent type B dissection. However, another clarification may be that patients who undergo elective surgery are patients with a more advanced stage of the disease, for example caused by the effect of the FBN1 mutation on fibrillin-1 protein, and that these patients therefore are more at risk for aortic dissections. Both factors may play a role in the occurrence of subsequent dissection in the proximal descending aorta.(6;10;15) At the last available aortic images (approximately 1.4 years before the occurrence of type B aortic dissection) mean diameter of the proximal descending aorta was only 31±7 mm. We showed that a descending thoracic aorta diameter 27mm was associated with type B aortic dissection in patients with a mean age of 36 ± 14 years and a mean BSA of 2.0 ± 0.2 m 2. To discriminate between high and low risk we calculated optimal cut-off values for 3, 5, 7 and 10 years of follow-up, yielding 27, 25, 25 and 26 mm respectively. We chose 27 mm because this corresponded with the reported upper level of normal of the thoracic descending aorta and because of the small differences generated by variability in follow-up duration. Although aortic dilation rate was significantly associated with type B dissection in our univariate analysis, this was not the case in our multivariate analysis. The aortic dilatation rates were low in our study and probably outweighed by aortic diameter in the multivariate analysis. Currently, the surgical guidelines advocate replacement of an aneurysm of the distal aorta in MFS when the diameter exceeds 50 mm. 4 Notably, none of the MFS patients with a distal aortic dissection were approaching this threshold for distal aortic surgery in our study. Despite that open thoracoabdominal aortic replacement can be achieved relatively safely in experienced centres, 23 it is associated with lung damage, spinal cord injury and diffuse bleeding requiring surgical revision. Therefore, future studies are warranted that assess the risks of elective surgery against the risks of the natural course of the development of descending thoracic aortic aneurysms into distal dissections. treatment effects In patients with MFS, rigorous antihypertensive medical treatment, aiming at a systolic blood pressure <120 mmhg, is thought to be important in the prevention of type B aortic dissection.(4) We could not demonstrate an association of blood pressure with type B aortic dissection, probably because blood pressure was generally well regulated in the total cohort. No beneficial effect of β-blockers on prevention of type B aortic dissection in the cohort could be shown. This may be due to the design of this study (retrospective, non-randomized trial) and to the high percentage of patients already using β-blockers, large variety in β-blocking agents, dosages, treatment duration and unknown patient compliance. On the other hand, we demonstrated that the use of ARBs could be beneficial in the prevention of type B aortic dissection in MFS patients. Recently, beneficial effects of losartan on the aortic dilatation rate and aortic arch dilatation rate after AoRR

95 The risk for type B aortic dissection in Marfan syndrome 93 was shown in adult patients with MFS.(16) However, the effect of losartan on (distal) aortic dissections could not be determined due to the low incidence of events and relative short follow-up. Meta-analysis of several running trials may serve to address this issue prospectively.(17-19) study limitations The limitations of our study include its retrospective nature, irregular imaging schemes with non ECG gated scans, practice variation between centers, and the different modalities of aortic imaging (MRI/CT). Furthermore, reference values for normal aortic size were derived from measurements including the aortic wall (ref JACC Cardiovasc Imaging 2008 March;1(2):200-9.), whereas we used measurements from inner-edge to inner edge (thus excluding the aortic wall). Therefore our cut-off value of 27 mm is actually somewhat larger than reported upper limit of normal. However, aortic diameters measured by various imaging modes have generally been found to correlate well.(20;21) Furthermore, analysis performed with data derived only from MRI rendered similar results. We developed an easily applicable risk scoring system to predict the probability of a type B aortic dissection. However, the variables used in the scoring system are inherent characteristics of the population to which the model was applied and will need validation in a prospective study or validation cohort. Furthermore, this study represents MFS patients treated at specialized MFS centers. 4 conclusion Predictors for type B aortic dissection in MFS include prior prophylactic aortic surgery, and a slightly enlarged descending thoracic aorta. Angiotensin-II receptor blocker therapy appears to reduce the incidence of type B aortic dissections. Furthermore, our risk model strongly discriminates MFS patients with a low risk for type B aortic dissection from high risk MFS patients.

96 94 Chapter 4 reference list (1) Silverman DI, Burton KJ, Gray J, Bosner MS, Kouchoukos NT, Roman MJ, et al. Life expectancy in the Marfan syndrome. Am J Cardiol 1995 Jan 15;75(2): (2) Engelfriet P, Boersma E, Oechslin E, Tijssen J, Gatzoulis MA, Thilen U, et al. The spectrum of adult congenital heart disease in Europe: morbidity and mortality in a 5 year follow-up period. The Euro Heart Survey on adult congenital heart disease. Eur Heart J 2005 Nov;26(21): (3) Groenink M, Lohuis TA, Tijssen JG, Naeff MS, Hennekam RC, van der Wall EE, et al. Survival and complication free survival in Marfan s syndrome: implications of current guidelines. Heart 1999 Oct;82(4): (4) Baumgartner H, Bonhoeffer P, De Groot NM, de HF, Deanfield JE, Galie N, et al. ESC Guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J 2010 Dec;31(23): (5) Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE, Jr., et al ACCF/AHA/AATS/ ACR/ASA/SCA/SCAI/SIR/STS/SVM Guidelines for the diagnosis and management of patients with thoracic aortic disease. A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology,American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons,and Society for Vascular Medicine. J Am Coll Cardiol 2010 Apr 6;55(14):e27-e129. (6) Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart 2006 Sep;92(9): (7) Loeys BL, Dietz HC, Braverman AC, Callewaert BL, De BJ, Devereux RB, et al. The revised Ghent nosology for the Marfan syndrome. J Med Genet 2010 Jul;47(7): (8) DeBakey ME, Henly WS, Cooley DA, Morris GC, Crawford ES, Beall AC, Surgical management of dissecting aneurysms of the aorta. J Thorac Cardiovasc Surg 1965 Jan;49: (9) Daily PO, Trueblood HW, Stinson EB, Wuerflein RD, Shumway NE. Management of acute aortic dissections. Ann Thorac Surg 1970 Sep;10(3): (10) den Hartog AW, Franken R, de WitteP, Radonic T, Marquering HA, van der Steen WE, et al. Aortic disease in patients with Marfan syndrome: aortic volume assessment for surveillance. Radiology 2013 Nov;269(2): (11) Wolak A, Gransar H, Thomson LE, Friedman JD, Hachamovitch R, Gutstein A, et al. Aortic size assessment by noncontrast cardiac computed tomography: normal limits by age, gender, and body surface area. JACC Cardiovasc Imaging 2008 Mar;1(2): (12) Hager A, Kaemmerer H, Rapp-Bernhardt U, Blucher S, Rapp K, Bernhardt TM, et al. Diameters of the thoracic aorta throughout life as measured with helical computed tomography. J Thorac Cardiovasc Surg 2002 Jun;123(6): (13) Groenink M, de RA, Mulder BJ, Spaan JA, van der Wall EE. Changes in aortic distensibility and pulse wave velocity assessed with magnetic resonance imaging following beta-blocker therapy in the Marfan syndrome. Am J Cardiol 1998 Jul 15;82(2): (14) Pencina MJ, D Agostino RB. Overall C as a measure of discrimination in survival analysis: model specific population value and confidence interval estimation. Stat Med 2004 Jul 15;23(13): (15) Finkbohner R, Johnston D, Crawford ES, Coselli J, Milewicz DM. Marfan syndrome. Long-term survival and complications after aortic aneurysm repair. Circulation 1995 Feb 1;91(3):

97 The risk for type B aortic dissection in Marfan syndrome 95 (16) Groenink M, den Hartog AW, Franken R, Radonic T, de Waard V, Timmermans J, et al. Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J 2013 Dec;34(45): (17) Lacro RV, Dietz HC, Wruck LM, Bradley TJ, Colan SD, Devereux RB, et al. Rationale and design of a randomized clinical trial of beta-blocker therapy (atenolol) versus angiotensin II receptor blocker therapy (losartan) in individuals with Marfan syndrome. Am Heart J 2007 Oct;154(4): (18) Detaint D, Aegerter P, Tubach F, Hoffman I, Plauchu H, Dulac Y, et al. Rationale and design of a randomized clinical trial (Marfan Sartan) of angiotensin II receptor blocker therapy versus placebo in individuals with Marfan syndrome. Arch Cardiovasc Dis 2010 May;103(5): (19) Forteza A, Evangelista A, Sanchez V, Teixido G, Garcia D, Sanz P, et al. [Study of the efficacy and safety of losartan versus atenolol for aortic dilation in patients with Marfan syndrome]. Rev Esp Cardiol 2011 Jun;64(6): (20) Dinsmore RE, Liberthson RR, Wismer GL, Miller SW, Liu P, Thompson R, et al. Magnetic resonance imaging of thoracic aortic aneurysms: comparison with other imaging methods. AJR Am J Roentgenol 1986 Feb;146(2): (21) Meijboom LJ, Groenink M, van der Wall EE, Romkes H, Stoker J, Mulder BJ. Aortic root asymmetry in marfan patients; evaluation by magnetic resonance imaging and comparison with standard echocardiography. Int J Card Imaging 2000 Jun;16(3):

98

99 Chapter 5 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial Maarten Groenink, Alexander W. den Hartog, Romy Franken, Teodora Radonic, Vivian de Waard, Janneke Timmermans, Arthur J. Scholte, Maarten P. van den Berg, Anje M. Spijkerboer, Henk A. Marquering, Aeilko H. Zwinderman, Barbara J.M. Mulder Adapted from: European Heart Journal Dec;34(45):

100 98 Chapter 5 abstract aim Patients with Marfan syndrome have an increased risk of life-threatening aortic complications, mostly preceded by aortic dilatation. Treatment with losartan, an angiotensin-ii receptor-1 blocker, may reduce aortic dilatation rate in Marfan patients. methods and results In this multicenter, open-label, randomised controlled trial with blinded assessments, we compared losartan treatment with no additional treatment in operated and unoperated adults with Marfan syndrome. The primary endpoint was aortic dilatation rate at any predefined aortic level after three years of follow-up, as determined by magnetic resonance imaging. A total of 233 participants (47% female) underwent randomisation to either losartan (n=116) or no additional treatment (n=117). Aortic root dilatation rate after 3.1±0.4 years of follow-up, was significantly lower in the losartan group than in controls (0.77±1.36 mm vs. 1.35±1.55 mm, p=0.014). Aortic dilatation rate in the trajectory beyond the aortic root was not significantly reduced by losartan. In patients with prior aortic root replacement, aortic arch dilatation rate was significantly lower in the losartan group as compared to the control group (0.50±1.26 mm vs. 1.01±1.31 mm, p=0.033). No significant differences in separate clinical endpoints or the composite endpoint (aortic dissection, elective aortic surgery, cardiovascular death) between the groups could be demonstrated. conclusion In adult Marfan patients, losartan treatment reduces aortic root dilatation rate. After aortic root replacement, losartan treatment reduces dilatation rate of the aortic arch.

101 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 99 introduction Patients with Marfan syndrome (MFS) have an increased risk of sudden death due to aortic dissection, mostly preceded by aortic dilatation (1-3). Life expectancy has improved due to surgical techniques for prophylactic aortic root replacement (1) and possibly due to β-blocker therapy (2-4). However, cardiovascular complications remain a major problem (5-7). MFS is usually caused by mutations in the FBN1 gene, leading to deficiency or malformations of the fibrillin-1 protein (8). Abnormal or deficient fibrillin-1 probably affects structural integrity of the extracellular matrix and may thereby enhance release of active transforming growth factor-β (TGF-β) (9;10). These processes are assumed to contribute to development of aortic medial degeneration and subsequent aortic dilatation and/or dissection (11;12). Recently, losartan emerged as a potentially effective novel treatment strategy due to its ability to inhibit TGF-β signalling and thereby preventing progressive aortic root dilatation in a MFS mouse model (13). The apparent beneficial effect of losartan treatment may also be attributed to other mechanisms. Losartan selectively blocks the angiotensin-ii type 1 (AT1) receptor within the renin-angiotensin-aldosterone system (14) and attenuates canonical TGF-β signalling in the aorta. Furthermore, losartan inhibits TGF-β-mediated activation of extracellular signal regulated kinase (ERK), by allowing continued signalling thought the angiotensin-ii type 2 receptor. (15). Indeed, losartan was reported to slow the aortic root dilatation rate in a small retrospective cohort of paediatric patients with a severe MFS phenotype (16). The primary aim of the COMPARE (COzaar in Marfan PAtients Reduces aortic Enlargement) study was to determine whether Cozaar (losartan) reduces the aortic dilatation rate at any predefined aortic level in adults with MFS. Additional aims of the study were to determine whether losartan influences aortic volume and incidence of aortic dissection, elective aortic surgery or cardiovascular death (17). 5 methods study design and participants The design of the COMPARE study was a randomised, multicenter, open-label trial with blinded assessments of endpoints (17). Patients were enrolled from January 2008 to December Patients were identified by all four Dutch university hospitals with a specialized multidisciplinary Marfan screening clinic and by using the national database of adults with congenital heart disease (CONCOR) (18). Eligible patients were adults (>18 years) who were diagnosed with MFS according to the Ghent criteria of 1996 (19). Patients were

102 100 Chapter 5 ineligible if they 1) had a history of angioedema or other known intolerance for angiotensin converting enzyme inhibitors (ACEi) and/or angiotensin-ii receptor blockers (ARB), 2) were already using ACEi or ARB, 3) had renal dysfunction, 4) had a known intolerance for intravenous contrast agents, 5) had an aortic root diameter > 50 mm, 6) had a history of aortic dissection, 7) had more than one vascular prosthesis, 8) were planned for aortic surgery within six months of inclusion, or 9) had the intention to become pregnant in the following three years. All previously prescribed medication, including β-blockers and calcium channel blockers, was continued after inclusion. The trial complied with the Declaration of Helsinki and was conducted with approval of the Medical Ethical Committees of all participating hospitals. Written informed consent was obtained from all participants. This trial is registered at the Netherlands Trial Register (number NTR1423). medication Patients in the losartan group started on 50 mg daily, and the dosage was doubled after 14 days. When side effects, such as dizziness, syncope, angioedema or renal dysfunction occurred, losartan dosage was either reduced or treatment was terminated. Patients were randomly assigned into a 1:1 ratio to receive losartan daily (losartan group) or no additional treatment (control group). Randomisation was performed with a computer generated randomisation sequence using randomly permuted blocks of 10. We stratified for the four hospitals. assessment and outcomes Participating patients started losartan treatment after baseline examinations. At baseline and after three years of follow-up, we examined patients medical history and performed magnetic resonance imaging (MRI) of the entire aorta. When MRI was contraindicated, computed tomography (CT) was performed. Annually, patients were evaluated by transthoracic echocardiography (TTE) and interviewed for side-effects, changes in medication use and clinical events. Aortic measurements were evaluated independently by three observers (AWdH, RF and AMS) without knowledge of patients medical therapy. The primary endpoint of this study was aortic dilatation rate at the 6 predefined aortic levels, from the aortic root to the bifurcation, measured by means of MRI or CT after three years of follow-up. The secondary endpoints were 1) total aortic volume expansion rate and 2) the incidence of the combined endpoint: cardiovascular mortality / aortic dissection / prophylactic aortic surgery. The decision to perform prophylactic aortic surgery was completely at the discretion of the attending cardiologists, based on European and American guidelines (20;21). When the surgical threshold was reached, patients underwent either a Bentall or David procedure to replace the dilated aortic root. Anticoagulation therapy was initiated when appropriate.

103 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 101 cardiovascular imaging All MRI scans were performed at two centres (AMC Amsterdam and LUMC Leiden). Aortic diameters were measured at six landmark levels on the MRI and CT scans; the aortic root, the ascending and descending thoracic aorta at the level of the pulmonary bifurcation, the aortic arch, the descending thoracic aorta at the level of the diaphragm and the abdominal aorta just proximal to the aortic bifurcation. When aortic aneurysms were detected between these landmark levels, separate aneurysm measurements were performed. Aortic volume was measured from the aortic annulus to the aortic bifurcation. Additionally, the aortic root was measured by TTE. (See Appendix A for a detailed description of MRI, CT and TTE acquisitions). statistical methods Sample size calculation (330 patients) was based on the primary endpoint. We assumed that the mean aortic root dilatation rate in MFS patients would be 0.9 mm/year (22), and that losartan would reduce this to 0.5±1.5 mm/year (2-sided α=0.05; β=0.2) (17). The effect of losartan on aortic dilatation rate was evaluated by covariance analysis with baseline aortic dimension as covariate. As our primary outcome parameter consisted of the changes of aortic diameters at six levels that were possibly correlated, we performed multiple testing correction by using a permutation approach of 1,000 permutations that took the correlations between the diameters at the six levels into account. From the permutation distribution we derived that when using a significance level of the family wise error was maintained at The P-values of the aorta diameter changes at the six levels were obtained from the permutation Null-distribution. All analyses were performed on the basis of intention-to-treat. Additionally, per-protocol and sensitivity analyses were performed. Per protocol analyses were performed to evaluate change in diameter between the two groups of MFS patients who continued their losartan treatment throughout the entire study and in whom losartan treatment was not started during the study, respectively. For the sensitivity analyses, patients who experienced a clinical endpoint were also included. Data shown are mean ± standard deviation. The combined secondary endpoint (aortic dissection, elective aortic surgery or cardiovascular death) was evaluated by means of the X 2 test. The proportions of patients with a stable aortic root diameter during three years of follow-up (dilatation rate 0 mm/3year) were compared using Fisher s exact test. Covariance analysis was also used to evaluate the losartan effect on aortic dilatation rate in subgroups of patients: males vs. females, with or without a known FBN1 mutation or b-blocker therapy, mean arterial pressure or > 90 mmhg, baseline aortic root diameter or > 45 mm and age or > 40 years. The mean differences in aortic root dilatation rate between losartan treated patients and control patients were plotted in a forest plot (23) and tested for significance using the interaction test between treatment-indicator (losartan or no losartan) and subgroup. 5

104 102 Chapter 5 Data analysis was performed using the SPSS statistical package (19.0 for windows; SPSS Inc., Chicago, Illinois, USA). results Patients From January 2008 until December 2009, 233 patients (41±13 years, 47% females) were enrolled; 116 were randomly assigned to treatment with losartan and 117 to no additional treatment (Figure 1). Patient characteristics at baseline are shown in Table 1. Follow-up was 3.1±0.4 years, similar in both arms. A losartan dosage of 100 mg daily was achieved in 63 patients (54%). In 34 patients (29%) losartan dosage was 50 mg daily, in 2 patients (2%) losartan dosage was reduced to 25 mg and in 17 patients (15%) losartan treatment was ceased due to side effects, including dizziness caused by low blood pressure (n=14), renal dysfunction (n=1), extreme fatigue (n=1) or angioedema (n=1). In one patient randomised to the control group, losartan was initiated after 2 years (Figure 1). Other cardiovascular medicinal treatment regimes did not change during the study between baseline and follow-up. Five patients underwent a contrast-enhanced ECG triggered CT scan instead of MRI. Primary endpoint Aortic root dilatation rate could be evaluated in 145 patients with a native aortic root at the time of exclusion (Figure 1). Baseline characterises were comparable between patients with and without a native aortic root, with exception of the distal aortic dimensions (aortic volume; 222±56 ml vs. 271±70 ml respectively, p<0.001). There were no statistical significant differences between the losartan treated and control group in these 145 MFS patients with a native aortic root (aortic root diameter; 43.8±5.0 mm vs. 43.2±4.4 mm, p=0.436). The aortic root dilatation rate was significantly lower in the losartan group than in the control group, 0.77±1.36 mm/3 years vs. 1.35±1.55 mm/3 years, respectively, p=0.014 (Table 2, Figure 2 and 3). Aortic root dilatation rate in patients on only losartan therapy was 0.91±1.25 mm/3 years (n=17) and in patients without losartan or any other form of cardiovascular medical therapy was 1.34±1.12 mm/3 years (n=21, p=0.268). The per protocol and sensitivity analyses rendered similar results. Losartan was also significantly associated with reduced aortic root dilatation rate as measured by TTE in the intention-to-treat analysis, respectively 1.34±1.51 mm/3 years vs. 1.93±1.39 mm/3 years, p=0.021 (Table 2). As expected, losartan significantly reduced mean arterial blood pressure by 6±11 mmhg compared to baseline (p<0.001) and differed significantly from blood pressure

105 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 103 Patients from all four participating Marfan screening clinics and CONCOR assessed 797 patients for screening 564 Were not eligible 259 Did not meet protocol criteria 89 More than 1 aortic graft 56 Prior aortic dissection 72 Already using ACEi or ARB 22 Mental retardation 20 Pregnancy 253 Declined participation 142 Miscellaneous reasons 69 Refused medical treatment 42 Logistic reasons 52 Did not respond 233 patients underwent randomisation 116 patients were assigned to Losartan group 117 patients were assigned to Control group Follow up not performed due to 1 Non cardiovascular death 2 Refused participation Follow up not performed due to 2 Type B aortic dissection 10 Refused participation patients in the intention-to-treat analysis 105 patients in the intention-to-treat analysis Aortic root replacement* 27 replacements prior to study 10 replacements during study Aortic root replacement** 36 replacements prior to study 8 replacements during study 78 patients in the intention-to-treat aortic root analysis 67 patients in the intention-to-treat aortic root analysis 17 patients prematurely discontinued losartan treatment due to 14 dizziness or low blood pressure 1 renal dysfunction 1 extreme fatigue 1 angioedema 1 patient started losartan therapy during study figure 1 Randomisation and Follow-up for Aortic Diameter Analysis Patients were excluded from aortic diameter analysis due to refusal of participation in follow-up, noncardiovascular death in the losartan group or type B aortic dissection in the control group. ACEi denotes angiotensin converting enzyme inhibitors and ARB denotes angiotensin-ii receptor blockers. *After three years, a total of 37 patients had an aortic root graft (27 prior to the study and 10 during the study) in the losartan group. Of these 37 patients, 2 are not included in the box aortic root replacement due to non-cardiovascular death and refusal to participate in follow-up. **After three years, a total of 44 patients had an aortic root graft (36 prior to the study and 8 during the study) in the control group. Six of the 44 patients are not included in the box aortic root replacement due to refusal to participate in follow-up.

106 104 Chapter 5 table 1. Baseline Demographic and Clinical Characteristics of the Patients. * Variables control losartan general features N=117 N=116 Gender (female) 62 (53.0) 47 (40.5) Body Surface Area (m²) 2.0 ± ± 0.2 Age (years) 37.8 ± ± years 69 (59.0) 70 (60.3) > 40 years 48 (41.0) 46 (39.7) Cardiovascular medication usage β-blocker 82 (70.1) 87 (75.0) Calcium Channel Blocker 3 (2.6) 2 (1.7) Blood pressure Systolic (mmhg) 125 ± ± 14 Diastolic (mmhg) 74 ± ± 11 Mean Aortic Pressure 90 mmhg 65 (55.6) 63 (54.3) > 90 mmhg 52 (44.4) 53 (45.7) Aortic root 45 mm 50 (61.7) 44 (49.4) > 45 mm 31 (38.3) 45 (50.6) FBN1 mutation 97 (88.2) 86 (74.8) Aortic root surgery 36 (30.8) 27 (23.3) Distal aorta surgery 5 (4.3) 2 (1.7) Mitral valve prolapse 65 (55.6) 63 (54.3) Mitral valve surgery 5 (4.3) 4 (3.4) aortic dimensions by mri Aortic root (mm) 43.7 ± ± 5.6 Z-Score Aortic root 3.8 ± ± 1.5 Ascending aorta (mm) 28.1 ± ± 3.6 Aortic arch (mm) 24.4 ± ± 2.8 Descending aorta pulmonary artery (mm) 23.9 ± ± 3.7 diaphragm (mm) 21.2 ± ± 2.5 abdominal (mm) 16.2 ± ± 3.9 Aortic volume (ml) 244 ± ± 55 aortic dimensions by tte Aortic root (mm) 42.7 ± ± 5.0 * Plus-minus values are means ± SD. FBN1 analyses were not performed in 1 patient in the losartan group and in 3 patients in the control group. In 5 patients from the losartan group mutations were found in TGFB2 and MYH11 (n=1) gene. In 4 patients from the control group mutations were found in TGFB2 (n=1), MYLK (n=1), MYH11 (n=1) and TGFBR1 (n=1) gene. Abbreviations: TTE = transthoracic echocardiography, MRI = magnetic resonance imaging

107 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 105 table 2. Primary Outcomes in the Intention-to-Treat Population during the Study Period. * Outcome control losartan P-value N=105 N=113 aortic dilatation rate by mri Aortic root # 1.35 ± ± Ascending aorta 0.85 ± ± Aortic arch 0.61 ± ± Descending aorta pulmonary artery 0.72 ± ± diaphragm 0.43 ± ± abdominal 0.37 ± ± Aortic volume 12 ± ± aortic dilatation rate by tte Aortic root 1.93 ± ± * Data are change in milimeter per 3 years, with the exception of aortic volume (mililiter per 3 years) * Plus-minus values are means ± SD P-value after multiple testing correction # Aortic root assessed in 145 patients (67 in control group, 78 in losartan group) Abbreviations: TTE = transthoracic echocardiography, MRI = magnetic resonance imaging 5 figure 2 Aortic Root Dimensions at Baseline and after 3 Years of Follow up of Both Groups BL-L denotes Baseline aortic root diameter of patients treated with losartan, FU-L denotes Follow up aortic root diameter of patients treated with losartan, BL-C denotes Baseline aortic root diameter of patients without losartan therapy, FU-C denotes Follow up aortic root diameter of patients without losartan therapy. Data shown are mean ± standard error of the mean. changes in the control group after 3 years (3±9 mmhg, p=0.032). No correlation was found between mean arterial blood pressure or systolic blood pressure with aortic root dimension (p=0.855 and p=0.819, respectively) or aortic root dilatation rate (p=0.716 and p=0.967, respectively). Furthermore, regression analysis showed that change in mean arterial blood pressure or change in systolic blood pressure was not correlated with aortic root dilatation rate in patients treated with losartan or controls (respectively, r=0.058; p=0.630 and r=0.001; p=0.993, Figure 4). The percentage of participants with a stable aortic root (defined as a dilatation rate 0 mm/3 years) was 50% in the losartan group and 31% in the control group (p=0.022), with a number-needed-to-treat of 5.3 patients.

108 106 Chapter 5 figure 3 MR Images Showing the Aortic Root in Short Axis of a COMPARE Patient with Marfan at Baseline and after 3 Years of Follow-up Greatest aortic root diameter of 3 measured distances was used; A and D: non coronary cusp to right coronary cusp increased from 41 to 43 mm; B and E: right coronary cusp to left coronary cusp increased from 42 to 44 mm; C and F: non coronary cusp to left coronary cusp increased from 41 to 42 mm. Aortic dilatation rate beyond the aortic root was evaluated in 218 patients (Figure 1). Aortic dilatation rate in the trajectory beyond the aortic root was not significantly reduced by losartan (Table 2). figure 4 Absence of Correlation between Systolic Blood Pressure Change and Aortic Root Dilatation Rate in Both Groups Correlation between change in systolic blood pressure and aortic root dilatation rate in patients treated with Losartan (r=0.058, p=0.630) and in Controls (r=0.001, p=0.993).

109 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 107 secondary outcomes Aortic volume increase was assessed in 168 MFS patients with a native aortic root or aortic root replacement prior to study inclusion (excluded due to technical issues: 33, clinical endpoints: 20, refusal: 12, see Figure 1). In the intention-to-treat-analysis, the total aortic volume increase was similar in both groups (Table 2). The per protocol and sensitivity analyses rendered similar results. A total of 19 patients underwent prophylactic aortic surgery due to progressive aortic dilatation. No difference in separate clinical endpoints or the composite endpoint was found between the groups (prophylactic aortic root surgery: 10 vs. 8, distal aortic surgical intervention: 0 vs. 1, type B aortic dissection: 0 vs. 2, respectively for the losartan and control groups). No cardiovascular deaths occurred during the study. losartan treatment and aortic root replacement A history of aortic root replacement prior to inclusion was present in 63 patients (27 in the losartan group). At baseline, patients with aortic root replacement demonstrated greater aortic dimensions in the remaining aortic trajectory as compared to the total patient cohort. Furthermore, patients randomised to losartan demonstrated smaller dimensions of the aortic arch and the descending thoracic aorta at the level of the diaphragm as compared to the control group at baseline (respectively 24±3 mm vs. 26±4 mm, p=0.029 and 21±2 mm vs. 23±4 mm, p=0.009). Patients with prior aortic root replacement demonstrated greater distal aortic dilatation rates as compared to unoperated patients (Table 3 and Table 4). After aortic root replacement, aortic arch dilatation rate was significantly lower in the losartan group than in the control group (0.50±1.26 mm/3 years vs. 1.01±1.31 mm/3 years, respectively p=0.033). Aortic dilatation rate in the descending aorta at the level of the pulmonary artery and diaphragm was comparable between the groups (Table 3). No significant difference in aortic volume increase between groups could be demonstrated (Table 3). However, operated patients in the control group showed a significantly larger increase in aortic volume during follow-up than unoperated patients (20±18 ml/3 years vs. 8±13 ml/3 years, respectively, p=0.004). Losartan-treated patients in the operated and unoperated subgroups did not show this disparity (15±10/3 years vs. 11±15 ml/3 years, respectively, p=0.488). 5 subgroup analysis No subgroups (FBN1 mutation, mean arterial pressure, aortic root diameter, concomitant β-blocker usage, gender and age) could be identified in whom losartan therapy was more beneficial in reducing aortic root dilatation rate (Figure 5). No interaction between treatment-indicator and subgroups could be demonstrated (p>0.467).

110 108 Chapter 5 table 3. Aortic Dilatation Rate by MRI in Patients with Aortic Root Replacement at Baseline. * Outcome control losartan P-value N=36 N=27 Aortic arch 1.01 ± ± Descending aorta pulmonary artery 1.00 ± ± diaphragm 0.48 ± ± abdominal 0.16 ± ± Aortic volume 20 ± ± * Data are change in milimeter per 3 years, with the exception of aortic volume (mililiter per 3 years). * Plus-minus values are means ± SD. Abbreviations: MRI = magnetic resonance imaging table 4. Aortic Dilatation Rate by MRI in Patients with a Native Aortic Root at Baseline. * Outcome control losartan P-value N=73 N=82 Aortic arch 0.44 ± ± Descending aorta pulmonary artery 0.60 ± ± diaphragm 0.40 ± ± abdominal 0.46 ± ± Aortic volume 8 ± ± * Data are change in milimeter per 3 years, with the exception of aortic volume (mililiter per 3 years). * Plus-minus values are means ± SD. Abbreviations: MRI = magnetic resonance imaging figure 5 Effect of Losartan Treatment on Aortic Root Dilatation in Subgroups of Marfan Patients Among subgroups of patients, the mean differences in aortic root dilatation rate between losartan treated patients and control patients are indicated by solid squares. Horizontal lines represent 95 % Confidence Intervals (95% CI). (n) denotes number of patients in subgroup-analysis, MAP (mean arterial pressure, mmhg), Aortic root is presented in mm, Age is presented in years.

111 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 109 discussion This is the first prospective, randomised, controlled trial indicating a beneficial effect of losartan treatment on aortic root dilatation rate in adults with MFS. The reduction of mean aortic root dilatation rate in the losartan group was present, irrespective of age, sex, blood pressure, aortic root size, presence of a FBN1 mutation and concomitant β-blocker use (Figure 5). As subgroup analyses were performed on relatively small groups of patients, these results should be interpreted with some prudence. Subgroup analysis on aortic root dilatation rate between patients on only losartan and those without medical therapy did not show a significant difference, due to limited sample size. The effects of losartan monotherapy on aortic root dilatation rate will have to be awaited from other studies. Although we could not demonstrate a significant association of losartan treatment with reduced aortic dilatation rate beyond the aortic root or with clinical events, losartan was significantly associated with reduced dilatation rate of the aortic arch in the subgroup of patients with a history of aortic root surgery. However, this result should be interpreted with some caution as baseline aortic dimensions of patients with prior aortic root replacement were not completely comparable between the groups. We found large variability in individual aortic root dilatation rates in the losartan group and losartan treatment did not normalize the dilatation rate to that of the healthy population (aortic root dilatation rate of mm for each advancing decade of life) (24). The large interindividual differences in response to losartan treatment may be partly explained by genetic factors, such as different types of FBN1 mutations (25) and genetic modifiers, especially those involved in other inflammatory pathways (26) and partly by interindividual variation in forces acting on the aortic tissue. However, subgroup analyses could not identify any patient group with larger or smaller reduction in aortic root dilatation rate, most likely due to lack of power. Furthermore, aortic root diameters measured by MRI were larger than measured by TTE (mean difference 0.9±1.6mm); however, aortic root dilatation rate reduction measured by MRI was comparable to TTE findings (0.58 mm and 0.59 mm). This phenomenon has been observed and explained previously (27). The incidence of clinical events was low in our study. Therefore, the clinical relevance of losartan treatment on aortic surgery and aortic dissection could not be determined by this trial and requires a prospective study with longer follow-up and a much larger sample size. The low incidence of aortic dissections and absence of death in our study may have been caused by the low threshold for prophylactic aortic root surgery in MFS at mm according to current guidelines (20;21). In the current era of aggressive surgical prophylactic treatment, ascending aortic dissection has become a rare event in patients with known MFS. As a corollary, the fate of 5

112 110 Chapter 5 the aortic trajectory beyond the aortic root has become a major clinical issue. We could not demonstrate a reduction of aortic dilatation rate beyond the aortic root associated with losartan treatment in the entire cohort, most likely due to lack of power. Another possible explanation may relate to the different developmental origin of aortic root (neural crest cells) as compared to the remaining aorta. Neural crest cells showed to have a different response to TGF-β signalling (28;29). By blocking the AT1-receptor, losartan reduces arterial blood pressure and wall stress by vaso-active mechanisms (14). In our study, mean arterial blood pressure was significantly reduced in the losartan treatment group as compared to the controls. However, no correlation was found between change in mean arterial pressure or systolic blood pressure with aortic root dilatation rate. A possible explanation for the beneficial effect of losartan might be due to the ability of losartan to inhibit TGF-β signalling. TGF-β antagonists, other than losartan, that have no effect on blood pressure provide overt vascular protection in mouse models (15). However, wall stress appears to be an essential precondition for the development of aortic media degeneration, as shown by the completely normal architecture of the aortic wall in newborn MFS mice. Therefore, the beneficial effects of losartan may be caused by both signalling and blood pressure lowering effects. Targeted treatment dosage of 100 mg losartan daily was reached in only 54% of the patients, mainly due to side effects, such as hypotension by concomitant β-blocker use. However, in the intention to treat analysis, which includes the lower losartan treatment dosage as well, a beneficial effect of losartan on aortic root dilatation rate could be demonstrated. A possible explanation could be that a low dosage of losartan daily might already be enough to inhibit TGF-β signalling cascade. We observed that the trajectory beyond the aortic root dilates more progressively in patients with a history of aortic root replacement, as previously reported (6;30;31). The aortic dilatation rate of the aortic trajectory beyond the aortic root may be enhanced by hemodynamic factors, altered wall mechanics, loss of windkessel effect with higher pulsatile forces acting onto the descending aorta, or clamping of the aorta during the operation (6). In summary, losartan treatment reduces aortic root dilatation rate in adults with MFS with a number-needed-to-treat of 5.3 patients when comparing the percentage of patients with stable aortic root between both groups. Following prophylactic aortic root replacement, losartan treatment also has a beneficial effect on dilatation rate in the aortic arch.

113 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 111 study limitations We were unable to enrol the original defined total sample size of 330 patients, mainly due to our strict inclusion and exclusion criteria. Second, inclusion was limited to the patients known at the designated four Dutch university hospitals with a specialized multidisciplinary Marfan screening clinic and by using the national database of adults with congenital heart disease (CONCOR). Aortic root dilatation rate was overestimated in our original sample size analysis (dilatation rate: 1.35 mm/3 years as opposed to the expected 2.7 mm/3 years). Although this could be interpreted as a result of a less severely affected study cohort, similar aortic root dilatation rates have been reported previously (32). Another limitation of our study design is that aortic root dilatation rate could only be assessed in 145 MFS patients with a native aortic root. The decision to include MFS patients with prior aortic root replacement in our study was based on the hypothesis that losartan might also have a beneficial effect on the aortic trajectory beyond the aortic root. Furthermore, this is clinically highly relevant as a large proportion of the current MFS population already underwent aortic root replacement or most likely will undergo in the near future. Furthermore, the open label character of this study is a limitation. We were persuaded by the clinical urgency to rapidly assess the effect of losartan after the positive results in mice. A double blinded study would have delayed this process. More and more patients were already being treated with losartan before any evidence of a beneficial effect in humans (See Figure 1). Therefore we decided to perform an open label study in collaboration with the Dutch Marfan Organisation. Nevertheless, the endpoints were evaluated without knowledge of patients medical therapy. 5

114 112 Chapter 5 reference list (1) Silverman DI, Burton KJ, Gray J, Bosner MS, Kouchoukos NT, Roman MJ, et al. Life expectancy in the Marfan syndrome. Am J Cardiol 1995 Jan 15;75(2): (2) Engelfriet P, Mulder B. Is there benefit of beta-blocking agents in the treatment of patients with the Marfan syndrome? Int J Cardiol 2007 Jan 18;114(3): (3) Hartog AW, Franken R, Zwinderman AH, Groenink M, Mulder BJ. Current and future pharmacological treatment strategies with regard to aortic disease in Marfan syndrome. Expert Opin Pharmacother 2012 Apr;13(5): (4) Shores J, Berger KR, Murphy EA, Pyeritz RE. Progression of aortic dilatation and the benefit of longterm beta-adrenergic blockade in Marfan s syndrome. N Engl J Med 1994 May 12;330(19): (5) Cameron DE, Alejo DE, Patel ND, Nwakanma LU, Weiss ES, Vricella LA, et al. Aortic root replacement in 372 Marfan patients: evolution of operative repair over 30 years. Ann Thorac Surg 2009 May;87(5): ) Engelfriet PM, Boersma E, Tijssen JG, Bouma BJ, Mulder BJ. Beyond the root: dilatation of the distal aorta in Marfan s syndrome. Heart 2006 Sep;92(9): (7) Groenink M, Lohuis TA, Tijssen JG, Naeff MS, Hennekam RC, Van Der Wall EE, et al. Survival and complication free survival in Marfan s syndrome: implications of current guidelines. Heart 1999 Oct;82(4): (8) Robinson PN, Arteaga-Solis E, Baldock C, Collod-Beroud G, Booms P, De PA, et al. The molecular genetics of Marfan syndrome and related disorders. J Med Genet 2006 Oct;43(10): (9) Ramirez F, Dietz HC. Marfan syndrome: from molecular pathogenesis to clinical treatment. Curr Opin Genet Dev 2007 Jun;17(3): (10) Franken R, den Hartog AW, de Waard V, Engele L, Radonic T, Lutter R, et al. Circulating transforming growth factor-beta as a prognostic biomarker in Marfan syndrome. Int J Cardiol 2013 Apr 10. (11) Carlson RG, Lillehei CW, Edwards JE. Cystic medial necrosis of the ascending aorta in relation to age and hypertension. Am J Cardiol 1970 Apr;25(4): (12) Nagashima H, Sakomura Y, Aoka Y, Uto K, Kameyama K, Ogawa M, et al. Angiotensin II type 2 receptor mediates vascular smooth muscle cell apoptosis in cystic medial degeneration associated with Marfan s syndrome. Circulation 2001 Sep 18;104(12 Suppl 1):I282-I287. (13) Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science 2006 Apr 7;312(5770): (14) Burnier M. Angiotensin II type 1 receptor blockers. Circulation 2001 Feb 13;103(6): (15) Habashi JP, Doyle JJ, Holm TM, Aziz H, Schoenhoff F, Bedja D, et al. Angiotensin II type 2 receptor signaling attenuates aortic aneurysm in mice through ERK antagonism. Science 2011 Apr 15;332(6027): (16) Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz HC, III. Angiotensin II blockade and aorticroot dilation in Marfan s syndrome. N Engl J Med 2008 Jun 26;358(26): (17) Radonic T, de Witte P, Baars MJ, Zwinderman AH, Mulder BJ, Groenink M. Losartan therapy in adults with Marfan syndrome: study protocol of the multi-center randomized controlled COM- PARE trial. Trials 2010;11:3. (18) van der Velde ET, Vriend JW, Mannens MM, Uiterwaal CS, Brand R, Mulder BJ. CONCOR, an initiative towards a national registry and DNA-bank of patients with congenital heart disease in the Netherlands: rationale, design, and first results. Eur J Epidemiol 2005;20(6):

115 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 113 (19) De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996 Apr 24;62(4): (20) Baumgartner H, Bonhoeffer P, De Groot NM, de HF, Deanfield JE, Galie N, et al. ESC Guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J 2010 Dec;31(23): (21) Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE, Jr., et al ACCF/AHA/AATS/ ACR/ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation 2010 Apr 6;121(13):e266-e369. (22) Roman MJ, Rosen SE, Kramer-Fox R, Devereux RB. Prognostic significance of the pattern of aortic root dilation in the Marfan syndrome. J Am Coll Cardiol 1993 Nov 1;22(5): (23) Lewis S, Clarke M. Forest plots: trying to see the wood and the trees. BMJ 2001 Jun 16;322(7300): (24) Vasan RS, Larson MG, Levy D. Determinants of echocardiographic aortic root size. The Framingham Heart Study. Circulation 1995 Feb 1;91(3): (25) Judge DP, Biery NJ, Keene DR, Geubtner J, Myers L, Huso DL, et al. Evidence for a critical contribution of haploinsufficiency in the complex pathogenesis of Marfan syndrome. J Clin Invest 2004 Jul;114(2): (26) Radonic T, de Witte P, Groenink M, de Waard V, Lutter R, van EM, et al. Inflammation aggravates disease severity in Marfan syndrome patients. PLoS One 2012;7(3):e (27) Meijboom LJ, Groenink M, Van Der Wall EE, Romkes H, Stoker J, Mulder BJ. Aortic root asymmetry in marfan patients; evaluation by magnetic resonance imaging and comparison with standard echocardiography. Int J Card Imaging 2000 Jun;16(3): (28) ten Dijke P, Arthur HM. Extracellular control of TGFbeta signalling in vascular development and disease. Nat Rev Mol Cell Biol 2007 Nov;8(11): (29) Topouzis S, Majesky MW. Smooth muscle lineage diversity in the chick embryo. Two types of aortic smooth muscle cell differ in growth and receptor-mediated transcriptional responses to transforming growth factor-beta. Dev Biol 1996 Sep 15;178(2): (30) Mulder BJ. The distal aorta in the Marfan syndrome. Neth Heart J 2008 Nov;16(11): (31) Nollen GJ, Meijboom LJ, Groenink M, Timmermans J, Barentsz JO, Merchant N, et al. Comparison of aortic elasticity in patients with the marfan syndrome with and without aortic root replacement. Am J Cardiol 2003 Mar 1;91(5): (32) Meijboom LJ, Timmermans J, Zwinderman AH, Engelfriet PM, Mulder BJ. Aortic root growth in men and women with the Marfan s syndrome. Am J Cardiol 2005 Nov 15;96(10): (33) Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005 Dec;18(12): (34) den Hartog AW, Franken R, de Witte P, Radonic T, Marquering HA, van der Steen WE, et al. Aortic Disease in Patients with Marfan Syndrome: Aortic Volume Assessment for Surveillance. Radiology 2013 Jun 25. 5

116 114 Chapter 5 appendix a - image acquisition MRI acquisition was performed by either an Avanto (Siemens, Erlangen, Germany) or a Philips (Intera, release 11 and 12; Philips Medical Systems, Best, the Netherlands) 1.5 Tesla MRI scanner using a phased array cardiac receiver coil. Contrast-enhanced magnetic resonance angiography (MRA) of the total aorta was performed by first pass imaging of 0.2 ml/kg body weight contrast bolus of gadovist (Bayer Schering AG, Berlin, Germany) with a molarity of 1 mmol/l. Contrast agent was injected intravenously in the brachial vein at an infusion rate of 2 ml/s, and subsequently flushed by 20 ml saline at 2 ml/s, using contrast power injectors (Mallinckrodt Inc., St.Louis, MO, USA or Medrad Spectris Solaris EP MR Injection System, Warrendale, USA). Contrast enhanced MRA image acquisition was triggered by scout imaging of the contrast bolus and aimed to visualize the total aorta during first pass of the contrast bolus in the aorta. Imaging occurred during breath-holding at end-inspiration. The contrast enhanced MRA of the full aorta was acquired by means of a standard, commercially available non-ecg gated 3D, T1-weighed, spoiled gradient-echo sequence (either 3DFLASH on the Siemens system or 3DFFE on the Philips system). This resulted in a 3D presentation of the entire aorta with a near-isotropic resolution of 1.4x1.3x1.4 mm/voxel. In patients without aortic root replacement, aortic root size was assessed by cine imaging sequences (Steady State Free Precession, SSFP) perpendicular to the long axis of the aortic root as shown by coronal and sagittal scouts (either TrueFisp on the Siemens system or Balanced TFE on the Philips system) during end-expiration. Typical SSFP characteristics were: Slice thickness 6 mm, Flip Angle degrees, Field of View mm, matrix size 256x192, frames per cardiac cycle. These acquisitions resulted in a CINE short axis representation of the aortic root at the level of the sinus of Valsalva with an in-plane spatial resolution of x mm/pixel.and a temporal resolution of approximately ms. TTE was performed with a Vivid 7 (GE, Vingmed Ultrasound, Horton, Norway) ultrasound system by experienced ultrasound technicians. Aortic root diameters were measured in end-diastole at the level of the sinus of Valsalva, by using the leading edge to leading edge technique in parasternal long axis, consistent with the current American Society of Echocardiography guidelines (33). CT was performed by use of a Philips 64 slice CT scanner, using generally available iodine based contrast agents in a small number of patients.

117 Losartan Reduces Aortic Dilatation Rate in Adults with Marfan Syndrome: a Randomised Controlled Trial 115 image Processing Aortic root diameter was assessed by greatest end-diastolic diameter of three cusp-cusp dimensions from outer to inner wall on the SSFP images. All measurements beyond the aortic root were performed on multiplanar MRA reconstructions from inner to inner edge. Vessel analysis software (3mensio vascular, 3mensio Medical Imaging BV, Bilthoven, the Netherlands) was used to calculated aortic volumes. Intra- and inter-observer variability of aortic volume assessment showed excellent reproducibility (34). Images were loaded in the software with window and level settings acquired from the DICOM data. Total aortic volume was determined by the following technique; a central lumen line was created by manually placing a seeding point through the lumen of the aorta in the axial, the sagittal and the coronal plane. A complete set of multi-planar reformats was reconstructed by the computer perpendicular to this central lumen line, resulting in a stretched vessel view of the aorta, from the aortic valve to the aortic bifurcation. The aortic lumen was manually separated from the surrounding tissue by placing a cutoff line between the enhanced aortic lumen voxels and the surrounding voxels in four cross-sections. The volume of the contrast-enhanced aortic lumen was reconstructed from the individually segmented axial slices, starting at the level of the aortic root and ending at the level of the aortic bifurcation. 5

118

119 Chapter 6 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations Alexander W. den Hartog, Romy Franken, Maarten P. van den Berg, Aeilko H. Zwinderman, Janneke Timmermans, Arthur J. Scholte, Vivian de Waard, Anje M. Spijkerboer, Gerard Pals, Barbara J.M. Mulder, Maarten Groenink Adapted from: Netherlands Heart Journal 2016 Nov;24(11):

120 118 Chapter 6 abstract background Mild biventricular dysfunction is often present in patients with Marfan syndrome. Losartan has been shown to reduce aortic dilatation in patients with Marfan syndrome. This study assesses the effect of losartan on ventricular volume and function in genetically classified subgroups of asymptomatic Marfan patients without significant valvular regurgitation. methods In this predefined substudy of the COMPARE study, Marfan patients were classified based on the effect of their FBN1 mutation on fibrillin-1 protein, categorised as haploinsufficient or dominant negative. Patients were randomised to a daily dose of losartan 100 mg or no additional treatment. Ventricular volumes and function were measured by magnetic resonance imaging at baseline and after 3 years of follow-up. results Changes in biventricular dimensions were assessed in 163 Marfan patients (48% female; mean age 38±13 years). In patients with a haploinsufficient FBN1 mutation (n=43), losartan therapy (n=19) increased both biventricular end diastolic volume (EDV) and stroke volume (SV) when compared with no additional losartan (n=24) : left ventricular EDV: 9±26 ml vs. -8±24 ml, p=0.035 and right ventricular EDV 12±23 ml vs. -18±24 ml; p<0.001 and for left ventricle SV: 6±16 ml vs. -8±17 ml; p=0.009 and right ventricle SV: 8±16 ml vs. -7±19 ml; p=0.009, respectively. No effect was observed in patients with a dominant negative FBN1 mutation (n=92), or without an FBN1 mutation (n=28). conclusion Losartan therapy in haploinsufficient Marfan patients increases biventricular end diastolic volume and stroke volume, furthermore, losartan also appears to ameliorate biventricular filling properties.

121 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations 119 introduction Marfan syndrome (MFS) is caused by mutations in the FBN1 gene, leading to deficiency (haploinsufficient {HI} mutations) or malformation (dominant-negative {DN} mutations) of the fibrillin-1 protein,(1) ultimately leading to aortic dilatation and dissection.(2-4) Treatment with losartan, an angiotensin-ii receptor-1 blocker (ARB), has been shown to have a beneficial effect on aortic pathology in MFS murine models. Human MFS studies on aortic dilatation rate have demonstrated conflicting results,(5-7) however, losartan may have a beneficial effect in patients with an HI mutation. Although a mild biventricular dysfunction is often present in MFS patients without significant valvular pathology,(8-12) the effect of losartan on cardiac dimensions and function in MFS patients is still unknown. Here, we assess the effect of losartan on left and right ventricular dimensions and function in both HI and DN asymptomatic MFS patients without significant valvular regurgitation by cardiac magnetic resonance imaging (CMR). methods Patients All study subjects were participants in the multicentre, randomised COMPARE trial. (5;13) In short, the COMPARE trial was designed to investigate the effect of losartan on the aortic dilatation rate in adult MFS patients. Patients ( 18 years) fulfilling the Ghent criteria of 1996 were included.(14) Exclusion criteria were: (A) treatment with angiotensin-converting-enzyme inhibitor (ACEi) or ARB before study entry, (B) an aortic root diameter >50 mm, (C) a history of aortic dissection, and (D) 1 vascular prosthesis. For the current predefined study, (E) only patients fulfilling the revised Ghent criteria of 2010 (15) were included, and (F) patients with moderate to severe valvular regurgitation were excluded.(16) All previously prescribed medication, including β-blockers and calcium channel blockers, was continued during the study. Written informed consent was obtained from all participants. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the institution s human research committee. All patients gave their informed consent. 6 mutation classification To detect mutations, sanger sequencing of the 65 coding FBN1 exons in genomic DNA was used. Mutations were classified into two groups according to the effect of the mutation on the fibrillin-1 protein.(17) Patients with a DN effect were defined as having mutations leading to production of malfunctional fibrillin-1 protein. DN mutations included:

122 120 Chapter 6 1) missense mutations leading to stable mutant fibrillin-1 protein; 2) mutations leading to exon skipping or deletion resulting in in-frame events (18); and 3) premature termination codon mutations leading to a shorter but stable fibrillin-1 protein (> 10 exons). An HI effect of the FBN1 mutation was defined as a mutation leading to a reduced amount of normal fibrillin-1 protein (19-21) as predicted by the Alamut software (Interactive Biosoftware, Rouen, France).(21) HI mutations included: 1) deletion of the whole FBN1 gene; 2) deletion of at least the first or last exon; 3) premature termination codon mutations leading to nonsense mediated decay or to a very short truncated protein (translation < 10 exons of the FBN1 gene) (22); and 4) missense mutations leading to degradation of the protein.(23) medication Participating patients started on losartan treatment after baseline examinations. Patients were randomly assigned to a 1:1 ratio to receive losartan daily (losartan group), or to no additional treatment (control group). Target dosage was 100 mg losartan daily. magnetic resonance imaging At baseline and after 3 years of follow-up, CMR imaging of the right and left ventricle was performed. All CMR scans were performed at two centres (Academic Medical Center Amsterdam and Leiden University Medical Center). To obtain ventricular volumes, magnetic resonance imaging was performed with either an Avanto (Siemens, Erlangen, Germany), or a Philips (Intera, release 11 and 12; Philips Medical Systems, Best, the Netherlands) 1.5 Tesla MRI scanner using a phased array cardiac receiver coil. ECG-gated cine images were acquired during breath-hold using a segmented, steady-state, free-precession sequence. Short-axis views were obtained every 10 mm, starting from the base up to the apex and covering the entirety of both ventricles. Additionally, flow (m/s) through the aortic root and ascending aorta was assessed by phase contrast velocity mapping to rule out significant valvular regurgitation, unnoticed at prior echocardiography. image analysis The primary endpoint of this study was change in left and right ventricular volumes and systolic function after 3 years of follow-up measured by CMR. All measurements were performed using MASS and FLOW image analysis software (Medis, Leiden, the Netherlands) by a single observer (AWdH) under the supervision of an experienced cardiologist (MG) and radiologist (AMS) with no prior knowledge of patients medical therapy. Endocardial contours of the left and right ventricle were manually traced in end systole and end diastole on all short-axis images. End diastole was defined as the phase with the largest ventricular area and end systole as the phase with the smallest ventricular area.

123 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations 121 Left ventricular (LV) and right ventricular (RV) end diastolic volume (EDV), end systolic volume (ESV) and stroke volume (SV=EDV-ESV) were determined. Left and right ventricular ejection fraction (EF) was calculated (EF=SV/EDV). Flow through the aorta was calculated using manually drawn areas on the modulus images and the velocity values of the corresponding velocity encoded images by using flow image analysis software (Flow, Medis, Leiden, the Netherlands). Severity of valvular regurgitation was assessed quantitatively by CMR in accordance with ACC/AHH guidelines for the management of patients with valvular heart disease.(16) Moderate to severe valvular regurgitation was defined as regurgitant volume of 30 ml per beat or a regurgitant fraction of 30%.(16) A dilated ventricle was assumed to exceed an EDV of 97 ml/m 2.(24) statistical analysis Data are presented as mean value ± standard deviation and categorical variables as numbers and percentages. Between-group comparisons of quantitative and categorical variables were made by Students t-test and Fisher s exact tests, respectively. Associations between characteristics were explored with regression analysis. The effect of losartan on change in ventricular volumes and function was evaluated by the two-tailed Student`s t-tests. All analyses were per protocol analyses in order to evaluate the change in ventricular volumes between the two groups of MFS patients who continued their losartan treatment throughout the entire study, and those in whom losartan treatment was not started during the study. A two-tailed P < 0.05 was considered statistically significant. For statistical analyses, PASW 19.0 (SPSS Inc., Chicago, IL) for Windows (Microsoft, Redmond, WA) was used. 6 results Patients Patient recruitment and characteristics have been described in detail previously.(5) (17) In the COMPARE trial, a total of 233 MFS patients were included between January 2008 and December 2009.(5) Change in cardiac function was assessed in 163 patients (48% female, mean age 38 ± 13 years, Figure 1). A pathogenic FBN1 mutation was found in 135 patients (83%). In patients with a known FBN1 mutation, 92 patients (68%) had a DN FBN1 mutation and 43 patients (32%) had an HI FBN1 mutation. No FBN1 mutation could be found in 28 patients. A total of 77 MFS patients were randomised to losartan treatment, while 86 MFS patients received no additional treatment. The remaining cardiovascular medicinal treatment did not change during follow-up. Baseline characteristics between the treatment groups were comparable with the exception of body surface area (BSA) and heart rate in HI patients. (Table 1)

124 122 Chapter patients in Compare trial 29 patients excluded due to 6 mutations other than FBN1 23 Baseline CMR incomplete 204 patients with adequate baseline ventricular CMR 104 patients were assigned to Losartan group 100 patients were assigned to Control group 27 patients exluded due to: 17 Discontinued losartan treatment 2 Valve regurgitation 8 FU CMR not performed or incomplete 14 patients excluded due to: 1 Started on losartan treatment 13 FU CMR not performed or incomplete 77 patients in ventricular analysis 86 patients in ventricular analysis figure 1 Randomization and follow-up for ventricular function analysis CMR denotes cardiac magnetic resonance, FU denotes follow-up Left ventricular EF in DN patients was lower than the EF in HI patients, 51 ± 7% vs. 53 ± 4%, p= The remaining ventricular characteristics were comparable between the groups, including the frequency of a dilated ventricle 26% vs. 23%, p=0.724, respectively. losartan and ventricular function in mutational subgroups After a mean of 3.1 ± 0.4 years of follow-up, the biventricular EDV increased in HI patients on losartan therapy (n=19) compared with HI controls (n=24) LV-EDV: 9 ± 26 ml vs. -8 ± 24 ml; p=0.035 and RV-EDV 12 ± 23 ml vs. -18 ± 24 ml; p<0.001, respectively. Similar results were seen in RV-ESV, however, not in LV-ESV (Table 2). In HI patients on losartan therapy, the increase in biventricular EDV was accompanied by an increase in SV, for LV-SV: 6 ± 16 ml vs. -8 ± 17 ml; p=0.009 and RV-SV: 8 ± 16 ml vs. -7 ± 19 ml; p=0.009, respectively (Table 2, Figure 2). No effect of losartan was observed in patients with a DN FBN1 mutation or without an FBN1 mutation. No effect of losartan on mean arterial blood pressure (MAP) was observed in HI patients. Changes in left and right ventricular volumes and function in the entire cohort were comparable between the losartan group and the control group. Furthermore, analysis of losartan only therapy or a combination of losartan and β-blocker therapy on right or left ventricular volumes, heart rate or EF in this subgroup rendered similar results.

125 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations 123 table 1 Baseline characteristics of 163 patients with Marfan syndrome characteristics losartan no losartan no FBN1 dn hi no FBN1 dn hi Number of patients Age (years) 39 ± ± ± ± ± ± 13 Male (%) 9 (56) 23 (55) 10 (53) 6 (50) 24 (48) 13 (54) BSA (m 2 ) 2.0 ± ± ± 0.2* 1.9 ± ± ± 0.2* MAP (mmhg) 93 ± ± ± ± ± 9 90 ± 8 Heartrate (bpm) 67 ± ± ± 14* 68 ± ± ± 6* Aortic root diameter (mm) 45 ± 6 44 ± 5 44 ± 6 42 ± 3 45 ± 5 43 ± 5 Aortic root replacement (%) 4 (25) 6 (14) 5 (26) 7 (58) 16 (32) 5 (21) Mitral valve surgery (%) 1 (6) (8) 0 Valve dysfunction moderate AR (%) (3) moderate MR (%) (3) moderate TR (%) moderate PR (%) Medication β-blocker use (%) 13 (81) 31 (74) 14 (74) 9 (75) 36 (72) 14 (58) Left Ventricular ESV/BSA (ml/m 2 ) 39 ± 9 42 ± ± ± ± ± 9 EDV/BSA (ml/m 2 ) 81 ± ± ± ± ± ± 15 SV/BSA (ml/m 2 ) 43 ± 8 44 ± ± 9 45 ± 8 44 ± 7 46 ± 8 EF (%) 53 ± 7 51 ± 6 52 ± 4 52 ± 6 50 ± 7 54 ± 5 Right Ventricular ESV/BSA (ml/m 2 ) 45 ± ± ± ± ± ± 10 EDV/BSA (ml/m 2 ) 86 ± ± ± ± ± ± 13 SV/BSA (ml/m 2 ) 41 ± 9 42 ± 8 42 ± 7 43 ± 9 42 ± 7 45 ± 7 EF (%) 48 ± 5 52 ± 7 51 ± 5 50 ± 6 51 ± 8 53 ± 7 BSA = Body Surface Area, MAP = Mean Arterial Pressure, AR = Aortic Regurgitation MR = Mitral Regurgitation, TR = Tricuspid Regurgitation, PR = Pulmonary Regurgitation ESV = End Systolic Volume, EDV = End Diastolic Volume, SV = Stroke Volume EF = Ejection Fraction, CI = Cardiac Index DN = Dominant Negative FBN1 mutation HI = Haploinsufficient FBN1 mutation Plus minus values are means±sd. * p<0.005 for BSA and Heartrate in HI patients between treatment groups 6

126 124 Chapter 6 table 2 The effect of losartan on left and right ventricle volume and function in Marfan patients with a FBN1 mutation leading to haploinsufficiency or a dominant-negative effect during 3 years of follow up haploinsufficiency dominant negative losartan no losartan P-value losartan no losartan P-value Number of patients Δ MAP (mmhg) -3 ± 8-4 ± ± 10-2 ± Δ HR (bpm) 3 ± 16 4 ± ± 21 3 ± Δ Left Ventricular ESV (ml) 3 ± 16 0 ± ± 22 2 ± EDV (ml) 9 ± 26-8 ± ± 33 0 ± SV (ml) 6 ± 16-8 ± ± 19-2 ± EF (%) 1 ± 5-2 ± ± 6-1 ± Δ Right Ventricular ESV (ml) 4 ± ± ± 22-8 ± EDV (ml) 12 ± ± 24 < ± 31-8 ± SV (ml) 8 ± 16-7 ± ± 17 0 ± EF (%) 1 ± 7 1 ± ± 7 2 ± MAP = Mean Arterial Pressure HR = Heartrate ESV = End Systolic Volume EDV = End Diastolic Volume SV = Stroke Volume EF = Ejection Fraction Plus minus values are means±sd. blood pressure and ventricular function In the entire cohort, after 3 year s treatment losartan significantly reduced MAP by 7±10 mmhg when compared with baseline (p<0.001), and differed significantly from change in MAP compared with the control group. At baseline, no significant correlation was found between MAP and ESV, EDV or SV (r=0.064, p=0.426; r=-0.014, p=0.860; r=-0.108, p=0.176; respectively). In losartan treated patients, there was also no significant correlation between change in MAP and change in ESV, EDV or SV (r=-0.117, p=0.326; r=-0.060, p=0.615; r=0.023, p=0.847; respectively) discussion To our knowledge, this is the first large follow-up study to assess the effect of losartan on biventricular function in both HI and DN asymptomatic MFS patients using CMR. In this study, LV-EF was lower in MFS patients with a DN FBN1 mutation than in those with an HI FBN1 mutation. Losartan therapy increased EDV and SV, independent of change

127 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations 125 figure 2 Boxplot showing change in left and right ventricular stroke volume between control and losartan group in Marfan patients with a FBN1 mutation leading to haploinsufficiency (n=43) or to a dominantnegative effect (n=92) SV denotes stroke volume, LV denotes left ventricle, RV denotes right ventricle. Boxes indicate median with interquartile range. Error bars indicate upper 90 percentile and lower 10 percentile limits. 6 in blood pressure only in MFS patients with an HI FBN1 mutation. Furthermore, no beneficial effect of losartan on cardiac systolic function in the study group as a whole could be demonstrated. MFS is caused by mutations in the FBN1 gene, leading to HI or DN mutations of the fibrillin-1 protein.(1) Both mutations indirectly lead to increased levels of TGF-β, which is correlated to aortic disease in MFS.(25;26) However, TGF-β most likely also plays a role in myocardial fibrosis.(27;28) Furthermore, blocking of TGF-β in both rat studies involving pressure-overloaded hearts and mice studies involving hypertrophic hearts, prevented myocardial fibrosis and diastolic dysfunction.(29-31) Losartan, an angiotensin II receptor type 1 inhibitor, is known to reduce both TGF-β signalling and blood pressure. However, we demonstrated that (change of) mean arterial blood pressure did not correlate with (change in) ventricular dimensions. Therefore, the effect of losartan on ventricular dimensions is not merely blood pressure related. A possible explanation for the significant difference in baseline ventricular function and response to losartan between HI and DN might therefore be due to a difference in TGF-β levels. This is in line with an earlier study by Franken et al., who demonstrated that losartan therapy significantly reduced the aortic root dilatation rate in the HI, whereas only a modest, insignificant reduction was found in DN MFS patients.(25) Furthermore, HI

128 126 Chapter 6 patients have a reduced but functionally normal fibrillin-1 protein, while the DN patient group is more heterogeneous since they display a broad spectrum of dysfunctional fibrillin-1 proteins, possibly explaining the variable response to losartan in the DN group. In addition, reduced ventricular function in DN patients might also be the result of a disorganised ECM due to the dysfunctional fibrillin-1 proteins.(32;33) In conclusion, systolic ventricular function in MFS patients with a DN FBN1 mutation was slightly, but significantly reduced compared with patients with an HI FBN1 mutation. In MFS patients with an FBN1 mutation leading to HI, losartan increased both EDV and SV, independent of change in blood pressure, suggesting a beneficial effect on left and right ventricular filling properties. No beneficial effect of losartan on ventricular function in the entire MFS cohort could be demonstrated. limitations Our patients were still relatively young and mainly asymptomatic. Furthermore, the clinical implications of our findings are limited. However, the occurrence of ventricular dysfunction will most likely increase due prolonged longevity in MFS as a result of prophylactic aortic root replacement and improved medicinal therapy. Other limitations are the limited sample size, lack of TGF-β levels and relatively small subgroups. Although losartan appeared to increase EDV in HI patients, thereby suggesting a beneficial effect on left and right ventricular filling properties, absolute diastolic filling properties were not assessed by CMR.

129 The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations 127 reference list (1) Robinson PN, Arteaga-Solis E, Baldock C, et al. The molecular genetics of Marfan syndrome and related disorders. J Med Genet 2006 Oct;43(10): (2) Habashi JP, Doyle JJ, Holm TM, et al. Angiotensin II type 2 receptor signaling attenuates aortic aneurysm in mice through ERK antagonism. Science 2011 Apr 15;332(6027): (3) den Hartog AW, Franken R, Zwinderman AH, et al. The risk for type B aortic dissection in Marfan syndrome. J Am Coll Cardiol 2015 Jan 27;65(3): (4) Nagashima H, Sakomura Y, Aoka Y, et al. Angiotensin II type 2 receptor mediates vascular smooth muscle cell apoptosis in cystic medial degeneration associated with Marfan s syndrome. Circulation 2001 Sep 18;104(12 Suppl 1):I282-I287. (5) Groenink M, den Hartog AW, Franken R, et al. Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J 2013 Dec;34(45): (6) Milleron O, Arnoult F, Ropers J, et al. Marfan Sartan: a randomized, double-blind, placebocontrolled trial. Eur Heart J 2015 Aug 21;36(32): (7) Lacro RV, Dietz HC, Sleeper LA, et al. Atenolol versus losartan in children and young adults with Marfan s syndrome. N Engl J Med 2014 Nov 27;371(22): (8) de Witte P, Aalberts JJ, Radonic T, et al. Intrinsic biventricular dysfunction in Marfan syndrome. Heart 2011 Dec;97(24): (9) Aalberts JJ, van Tintelen JP, Meijboom LJ, et al. Relation between genotype and left-ventricular dilatation in patients with Marfan syndrome. Gene 2014 Jan 15;534(1):40-3. (10) De Backer JF, Devos D, Segers P, et al. Primary impairment of left ventricular function in Marfan syndrome. Int J Cardiol 2006 Oct 10;112(3): (11) Meijboom LJ, Timmermans J, van Tintelen JP, et al. Evaluation of left ventricular dimensions and function in Marfan s syndrome without significant valvular regurgitation. Am J Cardiol 2005 Mar 15;95(6): (12) Alpendurada F, Wong J, Kiotsekoglou A, et al. Evidence for Marfan cardiomyopathy. Eur J Heart Fail 2010 Oct;12(10): (13) Radonic T, de Witte P, Baars MJ, Zwinderman AH, Mulder BJ, Groenink M. Losartan therapy in adults with Marfan syndrome: study protocol of the multi-center randomized controlled COM- PARE trial. Trials 2010;11:3. (14) De Peape A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996 Apr 24;62(4): (15) Loeys BL, Dietz HC, Braverman AC, et al. The revised Ghent nosology for the Marfan syndrome. J Med Genet 2010 Jul;47(7): (16) Bonow RO, Carabello BA, Chatterjee K, et al Focused update incorporated into the ACC/ AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation 2008 Oct 7;118(15):e523-e661. (17) Franken R, den Hartog AW, Radonic T, et al. Enhanced beneficial outcome of losartan therapy in Marfan patients with FBN1 haploinsufficiency mutations. Eur Heart J (18) Liu W, Schrijver I, Brenn T, Furthmayr H, Francke U. Multi-exon deletions of the FBN1 gene in Marfan syndrome. BMC Med Genet 2001;2:11. 6

130 128 Chapter 6 (19) Dietz HC, McIntosh I, Sakai LY, et al. Four novel FBN1 mutations: significance for mutant transcript level and EGF-like domain calcium binding in the pathogenesis of Marfan syndrome. Genomics 1993 Aug;17(2): (20) Milewicz DM, Grossfield J, Cao SN, Kielty C, Covitz W, Jewett T. A mutation in FBN1 disrupts profibrillin processing and results in isolated skeletal features of the Marfan syndrome. J Clin Invest 1995 May;95(5): (21) Hollister DW, Godfrey M, Sakai LY, Pyeritz RE. Immunohistologic abnormalities of the microfibrillarfiber system in the Marfan syndrome. N Engl J Med 1990 Jul 19;323(3): (22) Schrijver I, Liu W, Odom R, et al. Premature termination mutations in FBN1: distinct effects on differential allelic expression and on protein and clinical phenotypes. Am J Hum Genet 2002 Aug;71(2): (23) Schrijver I, Liu W, Brenn T, Furthmayr H, Francke U. Cysteine substitutions in epidermal growth factor-like domains of fibrillin-1: distinct effects on biochemical and clinical phenotypes. Am J Hum Genet 1999 Oct;65(4): (24) Hudsmith LE, Petersen SE, Francis JM, Robson MD, Neubauer S. Normal human left and right ventricular and left atrial dimensions using steady state free precession magnetic resonance imaging. J Cardiovasc Magn Reson 2005;7(5): (25) Franken R, den Hartog A, Radonic T, et al. Beneficial Outcome of Losartan Therapy Depends on Type of FBN1 Mutation in Marfan Syndrome. Circ Cardiovasc Genet 2015 Jan 22. (26) Franken R, Radonic T, den Hartog AW, et al. The revised role of TGF-beta in aortic aneurysms in Marfan syndrome. Neth Heart J 2015 Feb;23(2): (27) Lijnen PJ, Petrov VV, Fagard RH. Induction of cardiac fibrosis by transforming growth factorbeta(1). Mol Genet Metab 2000 Sep;71(1-2): (28) Dobaczewski M, Chen W, Frangogiannis NG. Transforming growth factor (TGF)-beta signaling in cardiac remodeling. J Mol Cell Cardiol 2011 Oct;51(4): (29) Kuwahara F, Kai H, Tokuda K, et al. Transforming growth factor-beta function blocking prevents myocardial fibrosis and diastolic dysfunction in pressure-overloaded rats. Circulation 2002 Jul 2;106(1): (30) Blyszczuk P, Muller-Edenborn B, Valenta T, et al. Transforming growth factor-beta-dependent Wnt secretion controls myofibroblast formation and myocardial fibrosis progression in experimental autoimmune myocarditis. Eur Heart J 2016 Apr 20. (31) Rikitake Y, Oyama N, Wang CY, et al. Decreased perivascular fibrosis but not cardiac hypertrophy in ROCK1+/- haploinsufficient mice. Circulation 2005 Nov 8;112(19): (32) Cook JR, Carta L, Benard L, et al. Abnormal muscle mechanosignaling triggers cardiomyopathy in mice with Marfan syndrome. J Clin Invest 2014 Mar 3;124(3): (33) Pinto YM, Pinto-Sietsma SJ, Philipp T, et al. Reduction in left ventricular messenger RNA for transforming growth factor beta(1) attenuates left ventricular fibrosis and improves survival without lowering blood pressure in the hypertensive TGR(mRen2)27 Rat. Hypertension 2000 Nov;36(5):

131

132

133 Chapter 7 Summary

134 132 Chapter 7 summary Marfan syndrome (MFS) is a multisystemic disorder of the connective tissue with a prevalence of 1 per 5000 and is caused by mutations in the FBN1 gene.(1) Cardiovascular characteristics of MFS include aortic dilatation, aortic dissection and a diminished ventricular function. Transforming growth factor β (TGF-β) signaling is upregulated in mice with a FBN1 mutation and treatment of these mice with losartan attenuates TGF-β signaling, prevents aortic media degeneration and aortic dilatation.(2) This thesis focuses on aortic dilatation, ventricular function and type B aortic dissection in adults with MFS and the effect of losartan on these cardinal features. chapter 1 introduces the main themes of the thesis by providing information regarding the cardiovascular manifestations of MFS, survival since the introduction of prophylactic aortic root replacement and relevant histological and biochemical pathways leading to aortic degeneration. Furthermore, current and possible future pharmacological interventions used to slow down aortic dilatation in MFS are addressed. In chapter 2 we demonstrate an elevated level of plasma TGF-β in MFS patients as compared to healthy controls, a correlation of plasma TGF-β with aortic root size, aortic root dilatation rate and cardiovascular events including aortic dissection and prophylactic aortic root surgery. Therefore, TGF-β might be a prognostic biomarker or is the endresult of severe wall damage in MFS. The latter implies an even more comprehensive pathway leading to the MFS cardiovascular features. With the introduction of prophylactic aortic root replacement, type A aortic dissection can be prevented effectively and life expectancy of MFS patients has increased. (3) This increased longevity has resulted in an increased incidence of complications in the aortic trajectory beyond the aortic root. However, identification of patients at risk for these distal complications has shown to be difficult to foresee by measuring aortic diameters, distensibility or dilatation rate and might therefore not reflect the extent of aortic disease. In chapter 3 we introduce aortic volume measurement as a novel method to measure the extent of aortic disease. We demonstrate a significant difference in aortic volume between MFS patients and controls as well as between operated and non-operated MFS patients. This novel method is highly reproducible and detects total aortic dilation rate more accurately than mean aortic diameter assessment. Although the correlation between end-points and total aortic volume measurements has not been assessed yet, total aortic size can be relevant in both clinic and future studies for estimation of the extent or progression of aortic disease. In chapter 4 we demonstrate a type B aortic dissection rate of 9% during a median follow-up period of 6 years (54 type B aortic dissection in 646 adult MFS patients). Predictors for type B aortic dissection are prior prophylactic aortic surgery and a slightly enlarged descending thoracic aorta. With our developed risk model, we can discriminate

135 Summary 133 MFS patients with a low risk for type B aortic dissection from MFS patients at high risk. The risk assessment tool may be used to intensify monitoring or treatment of patients with at least one of the risk factors. Furthermore, treatment with angiotensin receptor blockers appears to reduce the incidence of type B aortic dissections. In order to reduce aortic dilatation rate and prevent aortic root surgery or aortic dissection in MFS patients, pharmacological treatment with β-blockers is considered as standard care. However, β-blocker therapy does not address the TGF-β mediated aortic media degeneration. Losartan shows to prevent aortic media degeneration and aortic dilatation in MFS mice.(2) In chapter 5 we show a beneficial effect of losartan on aortic root dilatation rate in adults with MFS in a prospective, randomized controlled trial. Furthermore, losartan reduces aortic dilatation rate of the aortic arch in the subgroup of patients with a prior aortic root replacement. However, a large variability in individual aortic root dilatation rates upon losartan treatment is noticed. In chapter 6 we demonstrate that left ventricular ejection fraction was lower in MFS patients with a dominant negative FBN1 mutation (a mutation leading to malformed fibrillin-1 protein) than in those with a haploinsufficient FBN1 mutation (a mutation leading to a decreased amount of normal functional fibrillin-1 protein). Only in MFS patients with a haploinsufficient FBN1 mutation, losartan therapy increases end diastolic volume and stroke volume. As the occurrence of ventricular dysfunction will most likely increase due prolonged longevity as a result of prophylactic aortic (root) replacement and improved medicinal therapy, these results will be important for future MFS patients. reference list 7 1. Judge DP, Dietz HC. Marfan s syndrome. Lancet Dec 3;366( ): Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz HC III. Angiotensin II blockade and aorticroot dilation in Marfan s syndrome. NEnglJMed Jun 26;358( ): Jondeau G, Detaint D, Tubach F, Arnoult F, Milleron O, Raoux F, et al. Aortic Event Rate in the Marfan Population: A Cohort Study. Circulation Jan 17;125(2):

136

137 Future Perspectives

138 future PersPectives In his thesis we have assessed the effect of losartan on both aortic dilatation rate and ventricular function in patients with Marfan syndrome (MFS) and have tried to gain more knowledge regarding type B aortic dissection in this patient group. Improved understanding of the histological abnormalities causing MFS has identified several potential pharmacotherapies directly targeting aortic medial degeneration.(1) Losartan showed to be a promising candidate in MFS mice and showed to decreased aortic dilatation rate in a retrospective observational study in children.(2,3) Therefore, several randomized clinical trials were initiated to investigate whether losartan therapy on top of or instead of β-blocker therapy would be beneficial in reducing aortic root dilatation rate in MFS patients.(4 8) The COMPARE trial in this thesis showed for the first time a beneficial effect of losartan on top of β-blocker treatment on aortic dilation rate in MFS.(5) However, between the publication of the COMPARE trial and finishing this thesis, several other trials have been published. A small study by Chiu et al. demonstrated similar results as the COMPARE trial.(4) However, the Marfan Sartan trial and a study by Forteza et al. demonstrated no beneficial effect of losartan on top of baseline therapy or compared with atenolol treatment on aortic dilatation rate or on clinical events.(7,8) A large trial by Lacro et al. demonstrated that losartan and β-blocker therapy were both effective in reducing the aortic z-score.(6) Although these losartan trials have demonstrated conflicting results, we can conclude that losartan is both a well tolerated and a safe treatment option in MFS. Therefore, losartan could be a good alternative for patients with intolerance for β-blockers. However, we have to await publication of ongoing losartan trials and future meta-analyses to be able to answer the question if losartan is (more) effective in reducing aortic dilatation rate in MFS (than treatment with β-blockers) and effective in preventing aortic dissection or surgery.(9) Furthermore, the losartan trials demonstrated that losartan attenuates, but does not normalize the aortic dilatation rate to that of the healthy population. Therefore, research in the MFS field should further explore the pathways contributing to aortic medial degeneration development and elucidation of these pathways most likely will result into new potential pharmacotherapies targeting the pathogenic basis for MFS. The pathway that is responsible for aortic dilatation is most likely also responsible for an impaired ventricular function in MFS,(10) especially in patients with an DN FBN1 mutation. We showed that losartan therapy only in patients with an HI FBN1 mutation increased EDV and SV. Although the differences we found were relatively small, our data might be relevant for future MFS patients as the prolonged longevity after prophylactic aortic root replacement will most likely increase the occurrence of ventricular dysfunction. Future studies will have to show us if this is the case.

139 Future Perspectives 137 In this thesis we also have obtained more insight in aortic dilatation in MFS. Total aortic volume showed to be a highly reproducible new method to detect aortic dilatation and thereby assess the extent of aortic disease. However, association of total aortic volume with major adverse cardiac events, such as type B aortic dissection or prophylactic aortic surgery, have not been explored in this thesis. A follow-up study could be performed to assess this. Type B aortic dissection in MFS remains a major problem and is responsible for a great cause of both morbidity and mortality. Dissection of the distal aorta, especially after profylactic aortic root replacement, showed to be very difficult to prevent with standard measuring tools. With our developed risk model, we were able to discriminate patients at high risk for type B aortic dissection from low risk patients. The risk model could therefore be used to intensify monitoring or treatment for these patients. However, these data will need validation in future prospective studies. reference list 1. Hartog AW, Franken R, Zwinderman AH, Groenink M, Mulder BJ. Current and future pharmacological treatment strategies with regard to aortic disease in Marfan syndrome. ExpertOpinPharmacother Apr;13( ): Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science Apr 7;312( ): Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz HC III. Angiotensin II blockade and aorticroot dilation in Marfan s syndrome. NEnglJMed Jun 26;358( (Electronic)): Chiu H-H, Wu M-H, Wang J-K, Lu C-W, Chiu S-N, Chen C-A, et al. Losartan added to β-blockade therapy for aortic root dilation in Marfan syndrome: a randomized, open-label pilot study. Mayo Clin Proc Mar;88(3): Groenink M, den Hartog AW, Franken R, Radonic T, de Waard V, Timmermans J, et al. Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J Dec;34(45): Lacro RV, Dietz HC, Sleeper LA, Yetman AT, Bradley TJ, Colan SD, et al. Atenolol versus Losartan in Children and Young Adults with Marfan s Syndrome. N Engl J Med Nov 27;371(22): Forteza A, Evangelista A, Sánchez V, Teixidó-Turà G, Sanz P, Gutiérrez L, et al. Efficacy of losartan vs. atenolol for the prevention of aortic dilation in Marfan syndrome: a randomized clinical trial. Eur Heart J Mar 21;37(12): Milleron O, Arnoult F, Ropers J, Aegerter P, Detaint D, Delorme G, et al. Marfan Sartan: a randomized, double-blind, placebo-controlled trial. Eur Heart J Aug 21;36(32): Pitcher A, Emberson J, Lacro RV, Sleeper LA, Stylianou M, Mahony L, et al. Design and rationale of a prospective, collaborative meta-analysis of all randomized controlled trials of angiotensin receptor antagonists in Marfan syndrome, based on individual patient data: A report from the Marfan Treatment Trialists Collaboration. Am Heart J May;169(5): de Witte P, Aalberts JJ, Radonic T, Timmermans J, Scholte AJ, Zwinderman AH, et al. Intrinsic biventricular dysfunction in Marfan syndrome. Heart Dec;97( ):

140

141 Nederlandse Samenvatting

142 140 Nederlandse samenvatting samenvatting Marfan syndroom (MFS) is een aangeboren bindweefselaandoening met een prevalentie van 1 op 5000 dat wordt veroorzaakt door een fout (mutatie) in het FBN1 gen en overmatige expressie van transforming growth factor-β (TGF-β). Dit zorgt voor afwijkingen in de longen, de ogen (ectopia lentis), het skelet en beïnvloedt het gehele cardiovasculaire systeem. Typerende cardiovasculaire afwijkingen voor MFS zijn aorta (wortel) dilatatie, aorta dissectie, verdikking van de atrioventriculaire kleppen en een verminderde ventrikelfunctie. Met behulp van de Gent criteria uit 2010 wordt de diagnose MFS gesteld wat onder andere de familie geschiedenis, aorta dilatatie of dissectie, ectopia lentis en multipele systemische afwijkingen omvat. De huidige richtlijnen zijn gericht op het voorkomen van een aortadissectie en adviseren profylactische chirurgie bij een aorta(wortel) diameter van mm. Verder wordt een behandeling met bètablokkers geadviseerd om de bloeddruk, de chronotropie en inotropie van het hart te verlagen. De achterliggende gedachte is dat hiermee de krachten die op de aortawand worden uitgeoefend verminderen en zo aorta dilatatie en aorta dissectie kan worden voorkomen. Bètablokker therapie richt zicht echter niet op aorta degeneratie, wat de onderliggende oorzaak is van aorta pathologie bij MFS. In muizen met een FBN1 mutatie leidt een verhoogd serum van TGF-β tot het MFS fenotype. Behandeling van deze muizen met losartan, een angiotensine receptor blokker, doet het TGF-β niveau in het bloed dalen en voorkomt daarmee de degeneratie en dilatatie van de aorta. Het doel van dit proefschrift is meer kennis te vergaren over aorta dilatatie, ventrikel functie en type B aorta dissectie in volwassen patiënten met MFS en het effect van losartan hierop. hoofdstuk 1 is te beschouwen als een algemene introductie over MFS. Dit hoofdstuk begint met een beschrijving van de cardiovasculaire manifestaties van MFS, de verbeterde overleving sinds patiënten profylactisch aortawortelvervanging ondergaan en de relevante histologische en biochemische cascades die uiteindelijk verantwoordelijk zijn voor degeneratie en dilatatie van de aorta. Verder beschrijft dit hoofdstuk de huidige en mogelijke toekomstige medicinale behandelingen om de aorta dilatatie een halt toe te roepen. In 1991 werd ontdekt dat een fout (mutatie) in het FBN1 gen de veroorzaker is van MFS. Het FBN1 gen zorgt voor de vorming van het fibrilline-1 eiwit, dat op zijn beurt weer zorgt dat het bindweefsel in de aorta goed functioneert. Tevens zorgt fibrilline -1 voor binding van TGF-β. Een mutatie in fibrilline-1 leidt tot een minder goed functionerend bindweefsel en voor hogere concentratie vrij en actief TGF-β. In hoofdstuk 2 laten we zien dat MFS patiënten een verhoogde TGF-β concentratie hebben in vergelijking met

143 Nederlandse samenvatting 141 mensen zonder MFS. Tevens tonen we aan dat een hogere TGF-β waarde correleert met een grotere aortawortel diameter, snellere groei van de aortawortel en een hogere kans op een aorta dissectie. TGF-β zou daarom een prognostische biomarker kunnen zijn, echter kan het ook een marker zijn van de reeds opgetreden schade in de aortawand. In het laatste geval zal er een nog complexere cascade ten grondslag liggen aan de MFS pathologie. Met de introductie van preventieve aortawortel chirurgie kan type A aorta dissectie effectief worden voorkomen en is de levensverwachting van MFS patiënten enorm verbeterd. Deze verbeterde overleving heeft echter wel geleid tot een verhoogde incidentie van complicaties in de distale aorta. Het discrimineren van MFS patiënten met een verhoogd risico op distale complicaties van laag risico MFS patiënten blijkt lastig middels het meten van aorta diameter, aorta stijfheid of aorta dilatatie snelheid. Deze parameters reflecteren dan waarschijnlijk ook niet de ernst van de aorta ziekte in MFS. In hoofdstuk 3 introduceren we de aorta volume meting als een nieuwe methode om de omvang van de aorta ziekte te evalueren. In dit hoofdstuk tonen we een significant verschil aan in aorta volume tussen patiënten met en zonder MFS. Tevens was er een verschil in aorta volume tussen geopereerde en niet-geopereerde MFS patiënten. Verder blijkt dat deze nieuwe methode zeer goed reproduceerbaar is en de totale aorta dilatatie accurater meet dan gemiddelde diameter metingen. In dit proefschrift is de correlatie tussen cardiovasculaire eindpunten en aorta volume nog niet geëxploreerd. Aorta volume kan echter relevant zijn voor gebruik in de kliniek en in klinische studies om de omvang of progressie van aortaziekte te monitoren. hoofdstuk 4 toont in een retrospectieve MFS studie een type B aorta dissectie incidentie van 9% aan gedurende een follow up periode van 6 jaar (54 type B dissecties in 646 MFS patiënten). Voorspellers van type B aorta dissectie zijn een eerdere profylactische aorta-chirurgie en een vergrootte descenderende thoracale aorta. Deze factoren hebben we meegenomen in het ontwikkelen van een risico model, waarmee MFS patiënten met een laag risico kunnen worden onderscheiden van MFS patiënten met een hoog risico. Het risicomodel zou gebruikt kunnen worden om patiënten met tenminste 1 risicofactor intensiever te monitoren. Daarnaast laten we zien dat het gebruik van een angiotensine receptor blokker geassocieerd is met een verlaagde kans op een type B aorta dissectie. De huidige MFS richtlijnen adviseren een behandeling met bètablokkers om aorta dilatatie snelheid te verminderen. Therapie met bètablokkers is echter niet specifiek gericht op de TGF-β geïnduceerde aorta degeneratie. Losartan behandeling in MFS muizen studies voorkomt de ontwikkeling van aorta degeneratie en aorta dilatatie. In hoofdstuk 5 tonen wij de resultaten van de prospectieve gerandomiseerde COMPARE studie, waarin wij onderzoeken of losartan in combinatie met de standaard bètablokker behandeling de aorta groei in MFS patiënten doet vertragen. In de studie zijn 233 MFS

144 142 Nederlandse samenvatting patiënten gerandomiseerd naar standaard behandeling of standaard behandeling met losartan. Aan het begin en aan het einde van de studie werd er een MRI scan gemaakt om de groei van de aorta te kunnen meten. Na 3 jaar follow-up was de aortagroei minder groot in de losartan groep. In hoofdstuk 6 tonen we een verminderde linker ventrikel ejectiefractie aan in MFS patiënten met een dominant negatieve FBN1 mutatie (een mutatie wat leidt tot fibrilline-1 eiwit van slechte kwaliteit) ten opzicht van MFS patiënten met een haploinsufficiënte FBN1 mutatie (een mutatie wat leidt tot een verminderde hoeveelheid van goed functionerend fibrilline-1 eiwit). In MFS patiënten met een haploinsufficiënte FBN1 mutatie leidt behandeling met losartan tot een toename in eind diastolische volume en slagvolume. Deze bevindingen zullen belangrijk zijn voor de toekomst aangezien het optreden van ventrikel dysfunctie waarschijnlijk zal toenemen vanwege de toegenomen levensverwachting voor MFS patiënten door de preventieve aortawortel chirurgie.

145

146

147 Publication List & Author Contribution

148 146 Publication list Publication list included in this thesis 1. hartog aw, Franken R, Zwinderman AH, Groenink M, Mulder BJ: Current and future pharmacological treatment strategies with regard to aortic disease in Marfan syndrome. Expert Opin Pharmacother 2012 April; 13: Franken R, den hartog aw, de Waard V, Engele L, Radonic T, Lutter R, Timmermans J, Scholte AJ, van den Berg MP, Zwinderman AH, Groenink M, Mulder BJ: Circulating transforming growth factor-beta as a prognostic biomarker in Marfan syndrome. Int J Cardiol 2013 Oct; 168: den hartog aw, Franken R, de Witte P, Radonic T, Marquering HA, van der Steen WE, Timmermans J, Scholte AJ, van den Berg MP, Zwinderman AH, Mulder BJ, Groenink M: Aortic disease in patients with Marfan syndrome: aortic volume assessment for surveillance. Radiology 2013 Nov; 269: den hartog aw, Franken R, Zwinderman AH, Timmermans J, Scholte AJ, van den Berg MP, de Waard V, Pals G, Mulder BJM, Groenink M: The risk for type B aortic dissection in Marfan syndrome. J Am Coll Cardiol 2015 Jan; 65: Groenink M, den hartog aw, Franken R, Radonic T, de Waard V, Timmermans J, Scholte AJ, van den Berg MP, Spijkerboer AM, Marquering HA, Zwinderman AH, Mulder BJ: Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J 2013 Dec; 34: den hartog aw, Franken R, van den Berg MP, Zwinderman AH, Timmermans J, Scholte AJ, de Waard V, Spijkerboer AM, Pals G, Mulder BJM, Groenink M: The effect of losartan therapy on ventricular function in Marfan patients with haploinsufficient or dominant negative FBN1 mutations. Neth Heart J 2016; 24: not included in this thesis 7. Franken R, hartog aw den, Singh M, Pals G, Zwinderman AH, Groenink M, Mulder BJM: Marfan syndrome: Progress report. Prog Pediatr Cardiol 2012; 34: Franken R, den hartog aw, van de Riet L, Timmermans J, Scholte AJ, van den Berg MP, de Waard V, Zwinderman AH, Groenink M, Yip JW, Mulder BJM: Clinical features differ substantially between Caucasian and Asian populations of Marfan syndrome. Circ J 2013; 77: Franken R, Hibender S, den hartog aw, Radonic T, de Vries CJM, Zwinderman AH, Groenink M, Mulder BJM, de Waard V: No beneficial effect of general and specific anti-inflammatory therapies on aortic dilatation in Marfan mice. PloS One 2014; 9:e Franken R, den hartog aw, Radonic T, Micha D, Maugeri A, van Dijk FS, Meijers- Heijboer HE, Timmermans J, Scholte AJ, van den Berg MP, Groenink M, Mulder BJ,

149 Publication list 147 Zwinderman AH, de Waard V, Pals G: Beneficial Outcome of Losartan Therapy Depends on Type of FBN1 Mutation in Marfan Syndrome. Circ Cardiovasc Genet 2015; 8: Franken R, Radonic T, den hartog aw, Groenink M, Pals G, van Eijk M, Lutter R, Mulder BJM, Zwinderman AH, de Waard V, COMPARE study group: The revised role of TGF-β in aortic aneurysms in Marfan syndrome. Neth Heart J 2015; 23: Franken R, den hartog aw, Franken R, Radonic T, de Waard V, Timmermans J, Scholte AJ, van den Berg MP, Spijkerboer AM, Marquering HA, Zwinderman AH, Mulder BJ, Groenink M: Effect van losartan op aortadilatatie in volwassenen met het Marfan syndroom. Ned Tijdschr Geneeskd (2):A6845 author contribution 1. current and future pharmacological treatment strategies with regard to aortic disease in marfan syndrome. AWdH, RF, AHZ, MG, BJM conception and design of research; AWdH prepared figures; AWdH drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript 2. circulating transforming growth factor-beta as a prognostic biomarker in marfan syndrome. RF, AWdH, VdW, LE, TR, RL, JT, AJS, MPvdB, AHZ, MG, BJM conception and design of research; RF, LE, TR performed experiments; RF, TR, LE analyzed data; RF, AWdH, AHZ, MG, BJM interpreted results of experiments; RF prepared figures; RF drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript 3. aortic disease in patients with marfan syndrome: aortic volume assessment for surveillance. AWdH, RF, PdW, TR, HAM, WEvdS, JT, AJS, MPvdB, AHZ, MG, BJM conception and design of research; AWdH, RF, PdW, WEvdS, TR performed experiments; AWdH, RF, PdW, WEvdS, AHZ analyzed data; AWdH, RF, AHZ, MG, BJM interpreted results of experiments; AWdH, PdW, WEvdS prepared figures; AWdH, PdW, WEvdS drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript 4. the risk for type b aortic dissection in marfan syndrome. AWdH, RF, JT, AJS, MPvdB, VdW, AHZ, MG, BJM conception and design of research; AWdH, RF performed experiments; AWdH, RF, AHZ analyzed data; AWdH, AHZ, MG,

150 148 Publication list BJM interpreted results of experiments; AWdH prepared figures; AWdH drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript 5. losartan reduces aortic dilatation rate in adults with marfan syndrome: a randomized controlled trial. JT, AJS, MPvdB, AHZ, MG, BJM conception and design of research; AWdH, RF performed experiments; AWdH, RF, AHZ analyzed data; AWdH, AHZ, MG, BJM interpreted results of experiments; AWdH, RF prepared figures; AWdH, RF drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript 6. the effect of losartan therapy on ventricular function in marfan patients with haploinsufficient or dominant negative fbn1 mutations. AWdH, RF, JT, AJS, MPvdB, VdW, AMS, GP, AHZ, MG, BJM conception and design of research; AWdH, RF performed experiments; AWdH, RF, AHZ analyzed data; AWdH, AHZ, MG, BJM interpreted results of experiments; AWdH prepared figures; AWdH drafted manuscript. All authors edited and revised manuscript for important intellectual content; All authors approved final version of manuscript

151 PhD Portfolio 149 Phd Portfolio PhD student: A.W. den Hartog PhD period: Supervisors: Prof. dr. B.J.M. Mulder and Prof. dr. A.H. Zwinderman Co-supervisors: dr. M. Groenink and dr. V. de Waard Phd training courses Practical Biostatistics 2012 Good clinical practice BROK 2012 Laboratory animals (art. 9 functionary) 2012 Presentations European Society of Cardiology congress, Paris (poster) 2011 American College of Cardiology, Chicago (poster) 2012 American Heart Association scientific sessions, Los Angeles (poster) 2012 Adult Congenital Heart Disease Scientific session, Toronto (poster) 2012 European Society of Cardiology congress, Munich (oral) 2012 European Society of Cardiology congress, Munich (poster) 2012 Adult Congenital Heart Disease scientific session, Skamania (poster) 2013 European Society of Cardiology congress, Amsterdam (poster) 2013 NVVC spring conference oral presentation 2012, 2014 NVVC fall conference oral presentation teaching Mentoring and supervising Bachelor student and thesis - 3 Master students and thesis awards and prices European Society of Cardiology Congress, Amsterdam (Hotline) 2012 NVVC best oral presentation 2012,2013

152

153 Curriculum Vitae 151 curriculum vitae Alexander Willem den Hartog werd op 19 augustus 1984 geboren in Haarlem. Na het VWO op het Christelijk Lyceum Zeist begon hij zijn studie geneeskunde in 2004 aan de Universiteit van Amsterdam. Ter verbreding van zijn horizon onderbrak hij zijn studie voor een extra wetenschappelijke stage in Canada. Na het behalen van zijn artsendiploma in 2011 is hij begonnen met zijn promotieonderzoek op de afdeling cardiologie van het Academisch Medisch Centrum in Amsterdam, met dit proefschrift als resultaat. Het promotieonderzoek werd begeleid door prof. dr. B.J.M. Mulder, prof. dr. A.H. Zwinderman, dr. M. Groenink en dr. V. de Waard. In 2014 begon hij zijn specialisatie tot cardioloog in het Onze Lieve Vrouwe Gasthuis onder supervisie van drs T. Slagboom en dr. J.P.R. Herrman. Momenteel doet hij zijn vooropleiding bij de interne geneeskunde in het Onze Lieve Vrouwe Gasthuis onder supervisie van dr. Y. F.C. Smets en drs. W.E.M. Schouten. Sinds 2014 is hij getrouwd met Nazanin Hakimzadeh en samen hebben zij 1 dochter, Ariana.

154

155 Dankwoord

156 154 Dankwoord Voor het tot stand komen van dit proefschrift heb ik 2,5 jaar doorgebracht als promovendus op de afdeling Cardiologie in het AMC. Drie jaar na het beëindigen van mijn tijd aldaar is het proefschrift dan eindelijk klaar. Zoals mij voorgangers kunnen beamen is promoveren alleen mogelijk met de hulp van vele anderen. Ik wil iedereen bedanken die een bijdrage heeft geleverd aan de totstandkoming van dit proefschrift en in het speciaal alle deelnemers van de COMPARE studie. Echter zonder de begeleiders van mijn studentenstage A.C.J.M. de Pont, J. Horn en M. Schultz, allen intensivisten, was het waarschijnlijk nooit tot een promotietraject gekomen. Zij waren degenen die mij als jonge student lieten zien wat onderzoek doen inhield en gaven mij de kans gaven om mijzelf verder te ontwikkelen in Toronto. Veel dank ben ik verschuldigd aan mijn promotor, prof. dr. Mulder. Beste Barbara, ik heb groot respect voor jouw organiserend vermogen en de efficiëntie waarmee je te werk gaat. Je weet mensen te motiveren en bent daarnaast een zeer succesvolle wetenschappen en dokter. Ik heb veel van je kunnen leren. Prof. dr. Zwinderman, beste Koos, bedankt voor al de keren dat ik bij je kon aankloppen. Middels je beruchte white board of simpelweg met een goed gesprek wist je me dan weer in de juiste richting te sturen. Daarnaast wil ik je bedanken voor je blijvende vertrouwen. Dr. Groenink, beste Maarten, bedankt voor al de tijd die jij in het COMPARE project en mij hebt geïnvesteerd. De uren naast je op F3, het MRI workstation, maar ook bij je thuis, terwijl jij de artikelen totaal herschreef zal ik niet vergeten. Drie andere belangrijke lessen die ik van jou heb geleerd: 1) Neem nooit genoegen met de eerste-de-beste optie voor alle smaken te hebben geprobeerd 2) Assumption is the mother of all fuck-ups 3) Er is niks mis met een goede opsomming, ook al loopt de zin door. Dr. de Waard, beste Vivian, dank voor jouw persoonlijke touch binnen het project. Met jouw hulp kwamen niet alleen alle lastig biochemische vraagstukken tot een goed eind, maar ook de onderlinge samenwerking. Beste COMPARE partners, prof. dr. Maarten van den Berg, dr. Arthur Scholte en dr. Janneke Timmermans, hartelijke dank voor jullie hulp als ik me in jullie centra begaf, het beschikbaar stellen van de patiënten, het meedenken over potentiele artikelen en jullie correcties van de manuscripten.

157 Dankwoord 155 Tevens wil ik alle overige medeauteurs bedanken voor hun waardevolle bijdragen. De leden van de promotiecommissie, dr. W. Stooker, prof. dr. R. Balm, dr. D.R. Koolbergen, dr. R.N. Planken, prof. dr. J. De Backer, prof. dr. AC. van Rossum, en prof. Dr. M.M.A.M Mannens ben ik zeer erkentelijk voor het beschikbaar stellen van hun expertise voor de beoordeling van de inhoud van mijn proefschrift en voor hun bereidheid om zitting te nemen in mijn promotiecommissie. Anita en Regina wil ik bedanken voor al hun hulp en ondersteuning op B2. Margreet bedankt voor de hulp in de laatste fase. Dank gaat ook uit aan Peggy voor het inplannen van alle MRIs, Rachel en alle MRI laboranten voor hun begeleiding op de MRI. Speciale dank gaat uit naar Soha voor het inwerken op het MRI-station en hulp bij het analyseren van de beelden. Mijn collega van het eerste uur, Romy. Dank voor de samenwerking, het vele leed, zweet, zondagochtenden en de vele mooie momenten die we samen hebben beleefd. Speciale dank gaat uit naar Teodora, voorgangster op de COMPARE studie, die ons in de begintijd van de promotie wegwijst wist te maken in het project. Mijn roommates, Zeliha en Paul, bedankt voor de alle mooie tijden op B Zonder jullie zou er überhaupt geen thesis zijn, aangezien ik altijd bij jullie heb moeten aankloppen voor schrijf-, statistiek- en abstract advies. Ook voor mijn promotiedag waardeer ik jullie ondersteuning. Tevens dank aan al mijn collega`s bij de congenitale cardiologie (Jouke, Annelieke, Teun, Mark, Maayke, Klaartje, Carla, Ilja en Josephine) voor de mooie tijden die we hebben beleefd op B2, congressen en alle andere activiteiten. Hetzelfde geldt natuurlijk voor alle andere onderzoekers binnen de cardiologie uit het AMC. Beste Sylvia en Lia, jullie deur stond altijd open. Dank daarvoor. Ook wil ik de medische studenten die hun stage bij mij hebben gelopen bedanken; Miram Said, Floor Louwerenburg, Kristine Beumer, Catherine Mary-Kaye Eriëlle de Cuba. Ik hoop dat de stage voor jullie hetzelfde heeft betekend als wat het ooit voor mij heeft betekend. Lieve ouders, hartelijk dank voor jullie eeuwige en onvoorwaardelijke steun, de aansporingen en de kansen die jullie me hebben gegeven om onder andere naar het buitenland

158 156 Dankwoord te gaan. Het heeft m`n leven veranderd en dankzij jullie ben ik waar ik nu ben. Hiervoor ben ik jullie eeuwig dankbaar voor. Mijn twee lieve kleine zusjes, Stephanie en Rachelle. Inmiddels zijn jullie 2 fantastische volwassen vrouwen en zijn jullie beide hard op weg om ontzettend succesvol te worden in jullie eigen field. Ik ben ontzettend trots op jullie. Dear Hakimzadeh Family, thank you for the very warm welcome you have given me into the family. It`s always a great pleasure to be with you. Lieve Nazanin, dr. Hakimzadeh, jij bent de meest bijzondere, slimme en lieve persoon die ik ken. Je hebt je PhD inmiddels afgerond en hebt er een prachtige dochter bij. Ik ben ontzettend trots op jou en weet zeker dat je succesvol zal zijn met je nieuwe business. Ik geniet van elke dag met jou en Ariana, houd ontzettend veel van jullie beide en zie uit naar onze gezamenlijke toekomst. The end.

159