Étude randomisée «MarfanSartan» comparant un antagoniste de l angiotensine II au placebo chez des patients présentant un syndrome de Marfan

Transcription

1 Archives of Cardiovascular Disease (2010) 103, CLINICAL RESEARCH Rationale and design of a randomized clinical trial (Marfan Sartan) of angiotensin II receptor blocker therapy versus placebo in individuals with Marfan syndrome Étude randomisée «MarfanSartan» comparant un antagoniste de l angiotensine II au placebo chez des patients présentant un syndrome de Marfan Delphine Detaint a, Philippe Aegerter b, Florence Tubach c, Isabelle Hoffman c, Henri Plauchu d, Yves Dulac e, Laurence Olivier Faivre f, Marie-Ange Delrue g, Patrick Collignon h, Sylvie Odent i, Maria Tchitchinadze a, Catherine Bouffard a, Florence Arnoult a,j, Mathieu Gautier a, Catherine Boileau a,k, Guillaume Jondeau a,,l a Service de cardiologie, centre de référence pour le syndrome de Marfan et apparentés, hôpital Bichat, Paris, France b Unité de recherche clinique, hôpital Ambroise-Paré, AP HP, université de Versailles St-Quentin-en-Yvelines, Paris, France c Inserm CIE 801, unité de recherche clinique, hôpital Bichat, AP-HP, université Paris-7, Paris, France d Service de génétique, hôpital Hôtel-Dieu, université Claude-Bernard, Lyon, France e Pédiatrie Cardiologie, pôle Enfants, hôpital des Enfants, Toulouse, France f Centre de génétique, hôpital d Enfants, université de Dijon, Dijon, France g Service de génétique médicale, hôpital Pellegrin Enfants, Bordeaux, France h Département de génétique médicale, hôpital d Enfants de la Timone, Marseille, France i Unité de génétique médicale, hôpital Sud, CHU, Rennes, France j Service d explorations fonctionnelles, hôpital Bichat, Paris, France k Laboratoire central de biochimie d hormonologie et de génétique moléculaire, hôpital Ambroise-Paré, AP-HP, Paris, France l Inserm, U 698, université Paris-7 Denis-Diderot, Paris, France Received 25 February 2010; received in revised form 23 April 2010; accepted 30 April 2010 Corresponding author. Fax: address: guillaume.jondeau@bch.aphp.fr (G. Jondeau) /$ see front matter 2010 Elsevier Masson SAS. All rights reserved. doi: /j.acvd

2 318 D. Detaint et al. KEYWORDS Aorta; Aortic aneurysm; Thoracic; Angiotensin II receptor blocker; Transforming growth factor beta Summary Background. Recent studies have demonstrated that blockade of the angiotensin II type 1 receptor with losartan decreases aortic damage in an animal model of Marfan syndrome (a KI mouse model with a pathogenic mutation in the gene coding for fibrillin-1). Aims. To demonstrate a beneficial effect of losartan on aortic dilatation when added to optimal therapy in patients with Marfan syndrome. Methods. This is a multicentre, randomized, placebo-controlled, double-blind, clinical trial with a 2-year inclusion period and a 3-year follow-up period. Aortic root diameter will be measured using two-dimensional echocardiography. Secondary endpoints will include incidence of aortic dissection, aortic root surgery, death, quality of life, tolerance and compliance with treatments. We aim to enrol a total of 300 patients aged 10 years who fulfil the Ghent criteria for Marfan syndrome. Analyses will be based on intention to treat. Conclusion. The results of this clinical trial could lead to profound modification of the management of aortic risk and complications in patients with Marfan syndrome and possibly in patients with thoracic aortic aneurysms of other aetiologies Elsevier Masson SAS. All rights reserved. MOTS CLÉS Syndrome de Marfan ; Aorte ; Anévrisme de l aorte thoracique ; Antagoniste des récepteurs de l angiotensine II ; TGF-beta Résumé Contexte. Le losartan limite la dilatation aortique sur un modèle de souris KI, porteuse d une mutation pathogénique du gène FBN1, codant pour la fibrilline de type 1. But de l étude. Montrer le bénéfice du losartan ajouté au traitement optimal chez des patients présentant un syndrome de Marfan. Méthodes. Étude randomisée, contre placebo, en double insu, avec une période d inclusion de deux ans et de trois ans de suivi. Le diamètre aortique est mesuré par échographie bidimensionnelle. Les critères secondaires sont l incidence des décès, de la dissection aortique, de la chirurgie aortique, la qualité de vie, la tolérance et la compliance au traitement. Trois cent patients devraient être inclus (patients de plus de dix ans remplissant les critères de Ghent). L analyse sera effectuée en intention de traiter. Conclusion. Les résultats de cette étude pourraient conduire à modifier la prise en charge de la pathologie aortique et ses complications chez les patients présentant un syndrome de Marfan mais aussi possiblement dans la prise en charge des anévrismes aortiques thoraciques d autres étiologies Elsevier Masson SAS. Tous droits réservés. Background Marfan syndrome: definition Marfan syndrome (MFS) is an autosomal dominant disorder with pleiotropic features, including skeletal abnormalities, ectopia lentis and aortic root dilatation. The main causal gene for MFS is FBN1, encoding fibrillin-1, a large glycoprotein that is a main component of extracellular microfibrils. Prognosis is determined mainly by aortic complications (dissection or death), after progressive dilatation of the aortic root. Diagnosis criteria have changed over time; international criteria were proposed in 1988 and refined in 1996, to increase specificity and to integrate genetic testing [1,2]. The current nosology is based on the Ghent criteria, which define major and minor manifestations in different systems (Table [not provided]) [2]. In this setting, the diagnosis of MFS requires at least two major criteria and the involvement of at least one other body system (i.e., three criteria in total). In the presence of an FBN1 mutation or when MFS is diagnosed in a first-degree relative, only one major criterion and the involvement of another body system is required [2]. These Ghent criteria have excellent specificity for FBN1 mutation recognition, because its detection is possible in 95% of patients who fulfil these criteria [3]. However, a mutation in the FBN1 gene is not pathognomonic of MFS and may generate a large array of phenotypes that overlap with MFS (familial ectopia lentis, Shprintzen-Goldberg syndrome, other fibrillinopathies) [4]. On the other hand, some features of MFS can also be present in patients with mutations in the gene coding for transforming growth factor beta (TGF ) receptor 2 (TGFBR2), who present with MFS type 2 [5 7]. Lastly, mutations in the gene coding for TGF receptor 1 (TGFBR1) may also lead to overlapping syndromes [3]. Therefore, defining clear frontiers for MFS, fulfilling the Ghent criteria and differentiating from other clinical conditions that overlap with MFS (MFS type 2, Loeys-Dietz syndrome, familial thoracic aortic aneurysm, Ehlers-Danlos vascular syndrome [8 10]) can sometimes be challenging. In the near future, a revised nosology will emerge, which will focus on the features and criteria that distinguish MFS from other disorders; in the meantime, however, the Ghent criteria remain the reference and will be used for classification of patients in the current study.

3 Marfan Sartan study 319 Current medical management The aortic root diameter at the sinuses of Valsalva is considered to be the best predictor of the occurrence of an aortic event [11]. Patients with an absolute aortic diameter > 50 mm, an aortic ratio > 1.3 (observed/expected diameters) or a z score > 3 (the z score is the number of standard deviations above the mean) are considered at high risk for catastrophic complications, based on data obtained more than 10 years ago, before the publication of the Ghent criteria [12]. In addition, rapid growth of the aortic root (> 0.5 cm/year) and a family history of dissection are also predictors of poor outcome in patients with MFS. Therefore, routine monitoring of the aortic diameter is necessary for determining the best time for surgery. Associated medical therapy is mandatory and includes avoidance of isometric exercises, sports limitation and preventive medical therapy with beta-blockers or calcium channel blockers, inducing bradycardia [13]. Indeed, the use of beta-adrenergic blockade to decrease the haemodynamic stress on the ascending aorta has been suggested since the 1970s [14]. The first randomized, open-label trial was published in 1994 [15] (Fig. 1). In this study, the rate of change of aortic root diameter and clinical outcomes (aortic regurgitation, aortic dissection, surgery, heart failure and death) were compared between 32 patients assigned randomly to receive propanolol and 38 controls. At baseline, the absolute aortic diameters were larger in the treated group than in the non-treated group (34.6 vs 30.2 mm) but the aortic ratios (observed/expected diameters) were similar (1.4 vs 1.3, not significant). After a decade of follow-up, the mean change in aortic ratios over time was significantly lower in the treatment group (0.023/year) than in the control group (0.084/year). No statistically significant difference was observed between groups in event-free survival, but the rate of clinical events was higher in the control group than in the treatment group. Retrospective studies in children have demonstrated the beneficial role of beta-blockers [16,17], with a reduction in the rate of change in aortic diameter of 0.16 mm/year and a decrease in the number of aortic complications [16]. From a theoretical point of view, the efficacy of betablockers relies on their haemodynamic properties, reducing the force of left ventricular ejection (i.e., dp/dt) by negative inotropy and the number of impulses due to bradycardia, particularly during stress and exercise. Most series also demonstrate an increase in indexes of arterial wall compliance (which is basically decreased in patients with MFS [18]) with beta-blockers [19 21]. Nevertheless, the absence of an increase in aortic compliance with betablocker therapy has been observed in patients with marked aortic enlargement or increased weight, stressing the importance of starting treatment early in the course of the disease, with dose optimization [22 24]. Marfan syndrome in 2010: physiopathology In MFS, histological observations of the aortic wall classically show medial degeneration, with smooth muscle cell disappearance, disorganization of elastic fibres and accumulation of mucopolysaccharides. These abnormalities are not specific to MFS, as they can be observed in the aortic wall Figure 1. Beneficial effect of beta-blocker therapy on the aortic root diameter (from Shores et al. [15]). The Y-axis represents the ratio of observed aortic diameter/normal aortic diameter. Arrows correspond to patients not taking the drug. of aneurysms with other aetiologies, such as those related to bicuspid aortic valves or even degenerative aneurysms [25 27]. These histological abnormalities led to hypotheses being generated on the pathogenesis of MFS aortic disease. Mutations in the FBN1 gene encoding fibrillin-1 result in an abnormal protein and enhanced proteolytic degradation of fibrillin-1. This protein is an essential component of the microfibrils that play a role in extracellular matrix structure and regulation, elastic fibre organization and cell adhesion [28]. Hence, the presence of structurally abnormal microfibrils would weaken the extracellular matrix, allowing progressive aortic dilatation. In this hypothesis, all features associated with MFS are secondary to the weakening of the extracellular matrix in different tissues, leading

4 320 D. Detaint et al. Figure 2. Relationship between transforming growth factor beta (TGF )-binding protein and extracellular microfibrils (from Isogai et al. [31]). TGF is synthesized as a precursor molecule containing a propeptide region. After it is synthesized, the TGF homodimer interacts with a latency-associated peptide (LAP; a protein derived from the N-terminal region of the TGF gene product), forming a complex called small latent complex. This complex remains in the cell until it is bound by another protein called latent TGF -binding protein (LTBP), forming a larger complex called large latent complex. This complex is secreted to the extracellular matrix. to increased growth of bones, hernia, cutaneous striaes, etc. Alternative hypothesis have been proposed, however, based on observations made in a mouse model of MFS, which carries a pathogenic mutation in FBN1 [29]. In this model, markers for activation of the TGF pathway (i.e., p-smad-2) were present in smooth muscle cells, suggesting activation. TGF molecules are cytokines, synthesized and secreted by smooth muscle cells as inactive precursors in the form of a large latent complex (that includes a pro-tgf molecule and the large TGF -binding protein ), which is stored in the extracellular matrix [30]. One proposed hypothesis is that abnormal fibrillin causes failure of the latent complex sequestration and excessive activation [31,32] (Fig. 2). Excessive TGF signalling would then cause increased smad-2 phosphorylation and nuclear localization, resulting in altered gene expression (Fig. 3). Experimental therapy Experimental studies in the KI mouse have shown an absence of development of aortic aneurysm or myxomatous mitral valve in animals treated with TGF antibodies [33,34]. Interestingly, a similar effect has been observed with the use of losartan, a blocker of the angiotensin II type 1 receptor (AT1) [33]. Indeed, AT1 stimulation with angiotensin II activates the process of fibrosis and cell proliferation. These effects are mediated by TGF activation. In the experimental study by Habashi et al. [33], comparison of aortic diameter growth between Marfan mice treated with placebo, propanolol or losartan showed that both beta-blockers and AT1 blockers (angiotensin II type 1 receptor blockers [ARBs]) reduced the rate of change in aortic diameter compared with placebo. However, regarding the histological abnormalities, only treatment with losartan reduced the disarray of the extracellular matrix significantly. The primary results of non-randomized and small studies in humans are encouraging. Brooke et al. [35] compared the progression of aortic root diameter in 18 children with MFS, before and after treatment with ARBs. After treatment with ARBs in addition to beta-blockers over 2 years, the rate of change of the Valsalva and sinotubular junction decreased significantly from 3.5 mm/year to 0.46 mm/year. However, this study had many limitations, including its retrospective nature, the selection of children at the time of maximal aortic growth, the absence of a control group of any kind and the normalization of the aortic diameter using unusual rules. Although this hypothesis is very appealing, some unexplained results have been observed. Firstly, blocking only AT1, ARBs allow beneficial effects (fibrosis and cell proliferation inhibition) via signalling through the angiotensin II type 2 receptor (AT2), whereas angiotensin-converting enzyme inhibitors (ACEIs) reduce both AT1 and AT2 signalling. Ahismatos et al. [36] randomized 17 adults with MFS to receive either an ACEI (perindopril) or placebo in association with beta-blocker therapy. The study showed that, after 6 months, the stiffness of the aorta and the aortic root diameters had decreased significantly in patients receiving the ACEI and the aortic diameter became smaller compared with placebo. Levels of TGF were reduced by

5 Marfan Sartan study 321 Rationale for this trial We have initiated this trial to evaluate the safety and benefit of losartan on aortic root growth in MFS when added to accepted standard therapy. Beta-blocker therapy has limitations (merely a haemodynamic effect, lack of target in the underlying tissue pathology, side effects), but remains the standard of medical care in MFS. Therefore, considering a trial without allowing beta-blocker therapy might be perceived to be unethical. Besides, while beta-blockers are exerting their haemodynamic effects, losartan should modify the physiology within the aortic wall; in other words, the pathophysiology indicates that their effects should be additive. Methods Study design Hypothesis and overview This trial is designed to test the hypothesis that the addition of ARB therapy (losartan) to optimal standard therapy will reduce the rate of aortic dilatation compared with placebo in MFS patients aged 10 years. The decision to include younger patients rather than just adults derives from the more rapid aortic root dilatation progression during growth and the fact that demonstration of benefit in a mouse model was obtained in young individuals. Population Figure 3. The transforming growth factor beta (TGF ) signalling pathway (modified from ten Dijke and Arthur [42]). ALK5: transforming growth factor beta receptor 1; TGFBR2: transforming growth factor beta receptor 2. ACEI therapy, suggesting that ACEIs, like ARBs, target the underlying tissue pathology, in addition to reducing haemodynamic stress. These results, obtained in only one centre in a very small number of patients (10 receiving placebo, seven receiving perindopril) will have to be reproduced by others before being accepted widely. However, they are supported by non-randomized data [37]. Secondly, the presence of p-samd-2 in smooth muscle cells has been observed in the aortic wall of patients with a TGFBR2 mutation blocking transmission of the signal and in aneurysms of various aetiologies; this is compatible with the release of TGF by the matrix when it is destroyed, regardless of the aetiology of the extracellular matrix alteration [27]. Hence the benefit of blockade of the renin-angiotensin system seems promising in MFS, but has not been demonstrated in humans. Multicentre trials are needed to address this question. A trial designed to test the efficacy of losartan vs atenolol is ongoing in the USA [38] and aims to include 604 patients (children and adults). Another three-arm Italian trial is ongoing to evaluate the effect of losartan vs nebivolol vs a combination of both on the progression of aortic root dilatation [39]. Inclusion criteria The inclusion criteria for this study are as follows: age 10 years; diagnosis of MFS according to Ghent criteria, with or without known FBN1 mutation; informed consent and assent of participant, and parent(s) or legal guardian, as applicable. Patients will be excluded from the study for the following reasons: prior or planned aortic root surgery; inability to obtain accurate measurement of aortic root due to poor acoustic windows with transthoracic echocardiography; contraindication to losartan (bilateral renal artery stenosis, history of angio-oedema while taking ARB therapy); lactose intolerance, galactosaemia, or glucose or galactose malabsorption; pregnancy or planned pregnancy within 36 months of enrolment; absence of medical insurance (Securité sociale or Couverture maladie universelle). Randomization and stratification Eligible subjects will continue their standard prophylactic therapy (beta-blocker [or calcium channel blocker if beta-blocker therapy is not tolerated]) and will be assigned randomly to receive either losartan or placebo. Randomization will assign losartan or placebo in a 1:1 ratio and will be stratified according to centre, age at inclusion (< or 18 years) and existence (or not) of ongoing prophylactic therapy (beta-blocker or calcium channel blocker). The randomization code will not be available to the investigators by any means, ensuring double-blind assignment.

6 322 D. Detaint et al. Table 1 Scheduled examinations at each visit. Examination Timing Day 0 Day 15 Month 6 Month 12 Month 18 Month 24 Month 30 Month 36 Informed consent signed Randomization Clinical examination Complete medical history SF-36 quality of life questionnaire Laboratory testing Creatinine concentration, uraemia, kalaemia Pregnancy test Echocardiography Treatment distribution aortic complications (aortic root surgery, aortic dissection, cardiac death, death); compliance with treatment (incidence of adverse drug reactions reported during routine surveillance [hypotension, syncope]; compliance with losartan [proportion of patients taking at least 80% of pills given]; compliance with beta-blockers based upon patient s declaration); evaluation of quality of life based upon the SF-36 questionnaire. Statistical considerations Figure 4. Methods of aortic root measurement using twodimensional echocardiography (parasternal view) (from Roman et al. [40]). Ao: aorta; OG: left atrium; VG: left ventricle. Follow-up Follow-up visits are scheduled every 6 months (Table 1) and include: clinical examination; questions about adverse drug reactions; SF-36 quality of life questionnaire; echocardiography; laboratory testing, including creatinine plasma concentration, uraemia and kalaemia (annually). The aortic root diameter will be measured using transthoracic echocardiographic pictures or sequences recorded by trained echocardiographers according to recommendations [40] (Fig. 4). Endpoints The primary endpoint is the rate of change in aortic root diameter (sinuses of Valsalva), normalized to its theoretical value (expressed as z-score per year). The secondary endpoints include: rate of change in aortic root diameter with absolute dimension (expressed in mm per year); Sample size The sample necessary to verify the hypothesis has been calculated using data obtained from Shores et al. [15]. In this study, which compared the rate of aortic root dilatation in patients receiving beta-blockers vs no treatment, the rate of aortic root dilatation was /year without treatment vs /year with beta-blockers (around four times lower) and the maximum standard deviation was The sample size calculation was based on the assumption that losartan efficacy in patients already receiving prophylactic therapy will be half of the beta-blocker efficacy (0.01), and on the comparison of losartan with placebo with 0.80 power and a two-sided of To account for potential patient dropout, a total of 150 patients will be required in each group. Analyses Analyses will be performed on an intention-to-treat basis. Baseline characteristics will be summarized as means ± standard deviations and percentages, and compared between the two arms of the trial using the chi-square test or analysis of variance. As a supplementary analysis, the mortality rate or event-free survival will be compared using the log-rank test. The significance level will be For the primary endpoint, the final analysis will compare the mean slope of aortic dilatation between groups with covariate-adjusted analysis and after 3 years of follow-up. Trial organization and timeline The Assistance publique Hôpitaux de Paris and the département de recherche clinique et du développement will be

7 Marfan Sartan study 323 responsible for all aspects of the study. The protocol has been approved by the Agence française de sécurité sanitaire des produits de Santé and by a safety committee (Comité de protection des personnes). Enrolment began in September 2008; by mid-april 2010 more than 200 subjects had been enrolled. All centres will follow the same study procedure. The trial has been declared on clinicaltrial.gov (NCT ). Discussion Choice of primary endpoint Our primary endpoint is the rate of change in aortic root diameter normalized to its theoretical value. This choice is based on the relationship between aortic root dimensions and outcome, namely aortic complication, and therefore prophylactic intervention. Aortic root diameters are also easily measured by transthoracic echocardiography, with good reproducibility. Furthermore, from a statistical point of view, the primary endpoint is a continuous variable, which is more powerful than a quantitative variable for identifying a potential difference between groups. Moreover, it was used previously for the evaluation of beta-blocker therapy in 1994 [15] and, more recently, for the evaluation of losartan [33,35]. Our decision to compare normalized diameters and not absolute diameters was due to the relationship between aortic diameter and body surface area, which is very variable. Using the normalized diameter should decrease variability and therefore increase the power of the study to detect a difference. Randomization should result in two similar groups, reducing differences in body size between groups and allowing comparison of the changes in mean absolute diameter (our second endpoint). We recognize that different normalizations exist in the literature, and we planned to use, like most teams, standardization according to the nomogram publication by Roman et al. [40]. However, we recently published new nomograms for children [41]. Implication for future management of Marfan syndrome If demonstration of safety and efficacy of losartan in association with beta-blocker therapy can be obtained in this double-blind, randomized trial, standard care of these patients will be modified. Until the results are obtained, however, the standard care remains beta-blocker therapy, as a few differences can be identified between a KI mouse model with one FBN1 mutation and humans with MFS. Firstly, in the mouse model, only one mutation has been evaluated, and this mutation has been associated with only one genetic background (all the mice have the same genotype). In contrast, in humans, each family carries its own mutation or almost so (private mutations), and the genetic background differs from one individual to another, which is responsible for the unique aspect of each human. It is not clear that all mutations will act through the same mechanism (negative dominance vs haploinsufficiency), and the effect of a given mutation is probably highly dependent on the genetic background, as suggested by the great clinical variability of MFS severity within one family (modifier genes). Besides, compliance may be an issue for a drug given long-term; while tap water is an absolute necessity for mice, taking an additional pill is optional for patients. Side effects may occur that may further limit the compliance of patients (including hypotension, which gives an increased perception of fatigue). These factors are crucial for drugs that have to be taken for life. It is impossible, therefore, to conclude before the results of the ongoing randomized trials are available, whether the prescription of losartan is useful in patients with mutations in the FBN1 gene. Comparison of ongoing trials The USA trial is comparing beta-blocker therapy (atenolol) directly with losartan in an open-label, randomized trial [38]. This study evaluates the advantages of two different first-line therapies but not the benefit of combining the two drugs compared with up-to-date standard therapy. Besides, in this protocol, the criteria chosen for optimizing beta-blocker therapy are not in keeping with usual care in European countries, including France (up-titration based on 24-hour electrocardiogram and mean dose of beta-blockade higher than that used in France). The Italian trial is comparing three different approaches directly: beta-blocker or losartan or both [39]. The beta-blocker being used (nebivolol) carries theoretical advantages over the non-selective propanolol used in the landmark study of Shores et al. [15], and over the betablocker used in the USA trial (atenolol): its vasodilatory properties could decrease the rebound wave and therefore the stress applied on the proximal aorta and enhance the haemodynamic benefit of the drug; its beta-1 selectivity should increase its tolerance and therefore compliance. The ARB being used also allows 1-day administration and optimal receptor blockade. Lastly, the relative benefits of the two classes of drug and their combination are ideal. The drawback of having three groups is the necessity for a high number of patients to obtain the statistical power to be able to recognize differences between groups. Lastly, the University of Ghent has also started a randomized trial with a design similar to ours, but also evaluating the evolution of aortic stiffness over time. The similar design may allow secondary combination of the populations to increase statistical power, which is obviously an issue when the protocol aims to include such a selected population. Which population may benefit from the results? In current medical practice, beta-blocker therapies are proposed for patients with ascending aorta aneurysms of various aetiologies, based on haemodynamic concept and the randomized trial completed in the population with MFS. We reported recently that similarities in the pathogenesis of aortic aneurysms of varying aetiologies, including increased p-smad-2 (a marker for TGF activation), have been observed [27]. Therefore, if slowing of aortic dilatation can be demonstrated in patients with MFS with AT1 blockade, the question of the applicability of the findings to a

8 324 D. Detaint et al. larger population (i.e., all aortic aneurysms) will obviously arise. Similarly, AT1 blockade may become the first-line medical therapy in hypertension associated with conditions known to be at risk of development of aneurysm of the ascending aorta (e.g. bicuspid aortic valve, coarctation of the aorta). Limitations Our study will not be able to evaluate the superiority of one therapy over the other. However beta-blocker therapy is the standard of care for patients with MFS and the current study will evaluate the efficacy of losartan in addition to this prophylactic standard therapy. The study results may not be extrapolated to children aged younger than 10 years and to variants of MFS who do not meet the Ghent criteria. Conclusions The rationale behind a trial of losartan therapy in addition to standard beta-blocker therapy in MFS is strong and is based on recent advances in the understanding of the disease pathogenesis. The results should establish if the benefits demonstrated in the mouse model are applicable to humans or if other strategies should be looked for. If strong evidence of a treatment benefit for ARBs in patients with MFS can be obtained, the natural history of MFS and aortic aneurysms with other aetiologies may change dramatically in the future. Conflict of interest statement No conflict of interest to report. Acknowledgement Funding: Ministère de la Santé, projet hospitalier de recherche clinique (PHRC). References [1] Beighton P, de Paepe A, Danks D, et al. International nosology of heritable disorders of connective tissue, Berlin, Am J Med Genet 1988;29: [2] De Paepe A, Devereux RB, Dietz HC, et al. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996;62: [3] Loeys BL, Schwarze U, Holm T, et al. Aneurysm syndromes caused by mutations in the TGF-beta receptor. N Engl J Med 2006;355: [4] Faivre L, Collod-Beroud G, Loeys BL, 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;81: [5] Attias D, Stheneur C, Roy C, et al. Comparison of clinical presentations and outcomes between patients with TGFBR2 and FBN1 mutations in Marfan syndrome and related disorders. Circulation 2009;120: [6] Boileau C, Jondeau G, Babron MC, et al. Autosomal dominant Marfan-like connective-tissue disorder with aortic dilation and skeletal anomalies not linked to the fibrillin genes. Am J Hum Genet 1993;53: [7] Boileau C, Jondeau G, Mizuguchi T, et al. Molecular genetics of Marfan syndrome. Curr Opin Cardiol 2005;20: [8] Loeys BL, Chen J, Neptune ER, et al. A syndrome of altered cardiovascular, craniofacial, neurocognitive and skeletal development caused by mutations in TGFBR1 or TGFBR2. Nat Genet 2005;37: [9] Pepin M, Schwarze U, Superti-Furga A, et al. Clinical and genetic features of Ehlers-Danlos syndrome type IV, the vascular type. N Engl J Med 2000;342: [10] Vaughan CJ, Casey M, He J, et al. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation 2001;103: [11] Gott VL, Greene PS, Alejo DE, et al. Replacement of the aortic root in patients with Marfan s syndrome. N Engl J Med 1999;340: [12] Legget ME, Unger TA, O Sullivan CK, et al. Aortic root complications in Marfan s syndrome: identification of a lower risk group. Heart 1996;75: [13] Keane MG, Pyeritz RE. Medical management of Marfan syndrome. Circulation 2008;117: [14] Halpern BL, Char F, Murdoch JL, et al. A prospectus on the prevention of aortic rupture in the Marfan syndrome with data on survivorship without treatment. Johns Hopkins Med J 1971;129: [15] Shores J, Berger KR, Murphy EA, et al. Progression of aortic dilatation and the benefit of long-term beta-adrenergic blockade in Marfan s syndrome. N Engl J Med 1994;330: [16] Ladouceur M, Fermanian C, Lupoglazoff JM, et al. Effect of beta-blockade on ascending aortic dilatation in children with the Marfan syndrome. Am J Cardiol 2007;99: [17] Rossi-Foulkes R, Roman MJ, Rosen SE, 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;83: [18] Jondeau G, Boutouyrie P, Lacolley P, et al. Central pulse pressure is a major determinant of ascending aorta dilation in Marfan syndrome. Circulation 1999;99: [19] Groenink M, de Roos A, Mulder BJ, et al. 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;82: [20] Jeremy RW, Huang H, Hwa J, et al. Relation between age, arterial distensibility, and aortic dilatation in the Marfan syndrome. Am J Cardiol 1994;74: [21] Nollen GJ, van Schijndel KE, Timmermans J, et al. Magnetic resonance imaging of the main pulmonary artery: reliable assessment of dimensions in Marfan patients on a simple axial spin echo image. Int J Cardiovasc Imaging 2003;19:141 7 [discussion 49 50]. [22] Haouzi A, Berglund H, Pelikan PC, et al. Heterogeneous aortic response to acute beta-adrenergic blockade in Marfan syndrome. Am Heart J 1997;133:60 3. [23] Rios AS, Silber EN, Bavishi N, et al. Effect of long-term betablockade on aortic root compliance in patients with Marfan syndrome. Am Heart J 1999;137: [24] Yin FC, Brin KP, Ting CT, et al. Arterial hemodynamic indexes in Marfan s syndrome. Circulation 1989;79: [25] Borges LF, Touat Z, Leclercq A, et al. Tissue diffusion and retention of metalloproteinases in ascending aortic aneurysms and dissections. Hum Pathol 2009;40: [26] Fedak PW, Verma S, David TE, et al. Clinical and pathophysiological implications of a bicuspid aortic valve. Circulation 2002;106:900 4.

9 Marfan Sartan study 325 [27] Gomez D, Al Haj Zen A, Borges LF, et al. Syndromic and non-syndromic aneurysms of the human ascending aorta share activation of the Smad2 pathway. J Pathol 2009;218: [28] Chung AW, Au Yeung K, Sandor GG, et al. 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;101: [29] Judge DP, Biery NJ, Keene DR, et al. Evidence for a critical contribution of haploinsufficiency in the complex pathogenesis of Marfan syndrome. J Clin Invest 2004;114: [30] Chaudhry SS, Cain SA, Morgan A, et al. Fibrillin-1 regulates the bioavailability of TGFbeta1. J Cell Biol 2007;176: [31] Isogai Z, Ono RN, Ushiro S, et al. Latent transforming growth factor beta-binding protein 1 interacts with fibrillin and is a microfibril-associated protein. J Biol Chem 2003;278: [32] Neptune ER, Frischmeyer PA, Arking DE, et al. Dysregulation of TGF-beta activation contributes to pathogenesis in Marfan syndrome. Nat Genet 2003;33: [33] Habashi JP, Judge DP, Holm TM, et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science 2006;312: [34] 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 2004;114: [35] Brooke BS, Habashi JP, Judge DP, et al. Angiotensin II blockade and aortic-root dilation in Marfan s syndrome. N Engl J Med 2008;358: [36] Ahimastos AA, Aggarwal A, D Orsa KM, et al. Effect of perindopril on large artery stiffness and aortic root diameter in patients with Marfan syndrome: a randomized controlled trial. JAMA 2007;298: [37] 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;95: [38] Lacro RV, Dietz HC, Wruck LM, 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;154: [39] Gambarin FI, Favalli V, Serio A, et al. Rationale and design of a trial evaluating the effects of losartan vs. nebivolol vs. the association of both on the progression of aortic root dilation in Marfan syndrome with FBN1 gene mutations. J Cardiovasc Med (Hagerstown) 2009;10: [40] Roman MJ, Devereux RB, Kramer-Fox R, et al. Twodimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol 1989;64: [41] Gautier M, Detaint D, Fermanian C, et al. Nomograms for aortic root diameters in children using two-dimensional echocardiography. Am J Cardiol 2010;105: [42] ten Dijke P, Arthur HM. Extracellular control of TGFbeta signalling in vascular development and disease. Nat Rev Mol Cell Biol 2007;8: