Marfan Sartan: A Randomized Placebo-Controlled Trial
Marfan Sartan: A Randomized Placebo-Controlled Trial
Marfan Sartan was a prospective, randomized, double-blind, multi-centre, placebo-controlled, parallel group, add-on trial comparing Losartan with placebo in patients with MFS, in addition to standard preventive therapy. The study background and design have been published. The protocol was reviewed and authorized by the Agence Française de Sécurité Sanitaire des Produits de Santé and the IRB 'Comité de Protection des Personnes Île-de-France XI, St Germain en Laye'. All patients provided written informed consent before study entry. This study complies with the declaration of Helsinki. This study was registered on ClinicalTrials.gov under the identifier NCT00763893.
To be included, patients had to be at least 10 years old, fulfil diagnosis of MFS according to Original Ghent criteria and have signed informed consent. They were not included if they had prior or planned aortic root surgery, contraindication to Losartan, pregnancy or planned pregnancy. Aortic diameter was not an entry criterion into the study.
Patients were enrolled at seven sites in France (Centre National de Référence in Paris, and Centres de Compétence in Lyon, Bordeaux, Rennes, Toulouse, Dijon, Marseille).
Eligible patients were randomly assigned (1 : 1) to Losartan (Losartan 50 mg once daily if <50 kg, 100 mg once daily if ≥50 kg as recommended for treatment of hypertension or heart failure) or placebo. After randomization, patients were followed up every 6 months with echocardiographic recordings. Measurement of potassium and creatinine plasma levels was performed every year. Other drugs and treatments were left to the treating physician's discretion.
Treatment and follow-up were to be maintained 3 years, and follow-up visits were scheduled every 6 months. After 3 years into the study, the patients had the opportunity to prolong their participation until the last included patient in the study was followed up for 3 years.
Echocardiographic measurements were made in a central lab (hospital Bichat, Paris), by experienced cardiologists (O.M., F.A., G.J.) blinded to the patient's identity and the treatment received. Measures were obtained at the level of the annulus, the sinuses of Valsalva (aortic root), the sino-tubular junction, the ascending aorta, the aortic arch immediately prior to the left subclavian artery, descending thoracic aorta (maximal diameter), and abdominal aorta (at the level of the celiac artery). As recommended by the ASE and EAE, measures were made using 2D imaging at end-diastole, in a strictly perpendicular plane to that of the long axis of the aorta using the leading edge to leading edge convention for aortic root, sino-tubular junction, and the ascending aorta; inner edge to inner edge convention for the annulus, aortic arch, descending aorta, and abdominal aorta.
A two-step sequencing strategy was performed to identify the molecular defect in patients in the course of diagnosis. In a first step, the FBN1 gene was investigated by combining Sanger sequencing and multiplex ligation-dependent probe amplification analysis. If no mutation was identified, other genes associated with the disease (TGFBR2, TGFBR1, SMAD3, TGFB2, ACTA2, and FBN2) were sequenced. To assess the deleterious effect of identified sequence variants, a series of widely accepted prediction algorithms was used (for details on molecular methods and algorithms, see Ref. 11).
The primary endpoint was the rate of change in aortic root diameter (sinuses of Valsalva), normalized to its theoretical value and expressed as mean change in z-score per year.
The secondary endpoints included: mean rate of change in raw unnormalized aortic root diameter (expressed in millimeter per year) and aortic complications (aortic root surgery, aortic dissection, cardiac death, and death).
Sample size calculation was based on the assumption that Losartan efficacy in patients already receiving prophylactic therapy would be half of the β-blocker efficacy reported in the study by Shores, i.e. from 0.02 to 0.01 with a SD of 0.03. Choosing a 0.80 power and a two-sided type I error of 0.05, the number of patients required by group (of similar size) would be 142. One hundred and fifty patients were required in each group.
Randomization. Randomization was made using an internet-based computerized randomization system stratified on study centre, age (< or ≥18 years old), and treatment with baseline preventive therapy. To ensure allocation concealment, Losartan and matching placebo were supplied to study sites in masked identical packages. Drug packages were given to the patients during follow-up visits.
Blinding. All study personnel involved in the operations of the study or with any potential site contact, such as medical monitors, remained blinded to treatment assignments from the time of randomization until after completion of statistical analysis.
Statistical Methods. Statistical analyses were performed on an intention-to-treat basis. For the primary endpoint, standard linear regressions were first used to estimate individual slopes for the change of aorta diameters over time, using all available measurements of each subject. A linear model (ANOVA) with adjustment on the stratification factors (centre, protective treatment at baseline, age <18 years old at baseline), was used to compare the mean slope of aortic dilatation between groups. All calculations were performed using R software version 3.1.1.
Methods
Marfan Sartan was a prospective, randomized, double-blind, multi-centre, placebo-controlled, parallel group, add-on trial comparing Losartan with placebo in patients with MFS, in addition to standard preventive therapy. The study background and design have been published. The protocol was reviewed and authorized by the Agence Française de Sécurité Sanitaire des Produits de Santé and the IRB 'Comité de Protection des Personnes Île-de-France XI, St Germain en Laye'. All patients provided written informed consent before study entry. This study complies with the declaration of Helsinki. This study was registered on ClinicalTrials.gov under the identifier NCT00763893.
Population
To be included, patients had to be at least 10 years old, fulfil diagnosis of MFS according to Original Ghent criteria and have signed informed consent. They were not included if they had prior or planned aortic root surgery, contraindication to Losartan, pregnancy or planned pregnancy. Aortic diameter was not an entry criterion into the study.
Patients were enrolled at seven sites in France (Centre National de Référence in Paris, and Centres de Compétence in Lyon, Bordeaux, Rennes, Toulouse, Dijon, Marseille).
Eligible patients were randomly assigned (1 : 1) to Losartan (Losartan 50 mg once daily if <50 kg, 100 mg once daily if ≥50 kg as recommended for treatment of hypertension or heart failure) or placebo. After randomization, patients were followed up every 6 months with echocardiographic recordings. Measurement of potassium and creatinine plasma levels was performed every year. Other drugs and treatments were left to the treating physician's discretion.
Treatment and follow-up were to be maintained 3 years, and follow-up visits were scheduled every 6 months. After 3 years into the study, the patients had the opportunity to prolong their participation until the last included patient in the study was followed up for 3 years.
Echocardiographic measurements were made in a central lab (hospital Bichat, Paris), by experienced cardiologists (O.M., F.A., G.J.) blinded to the patient's identity and the treatment received. Measures were obtained at the level of the annulus, the sinuses of Valsalva (aortic root), the sino-tubular junction, the ascending aorta, the aortic arch immediately prior to the left subclavian artery, descending thoracic aorta (maximal diameter), and abdominal aorta (at the level of the celiac artery). As recommended by the ASE and EAE, measures were made using 2D imaging at end-diastole, in a strictly perpendicular plane to that of the long axis of the aorta using the leading edge to leading edge convention for aortic root, sino-tubular junction, and the ascending aorta; inner edge to inner edge convention for the annulus, aortic arch, descending aorta, and abdominal aorta.
A two-step sequencing strategy was performed to identify the molecular defect in patients in the course of diagnosis. In a first step, the FBN1 gene was investigated by combining Sanger sequencing and multiplex ligation-dependent probe amplification analysis. If no mutation was identified, other genes associated with the disease (TGFBR2, TGFBR1, SMAD3, TGFB2, ACTA2, and FBN2) were sequenced. To assess the deleterious effect of identified sequence variants, a series of widely accepted prediction algorithms was used (for details on molecular methods and algorithms, see Ref. 11).
Outcomes
The primary endpoint was the rate of change in aortic root diameter (sinuses of Valsalva), normalized to its theoretical value and expressed as mean change in z-score per year.
The secondary endpoints included: mean rate of change in raw unnormalized aortic root diameter (expressed in millimeter per year) and aortic complications (aortic root surgery, aortic dissection, cardiac death, and death).
Sample Size
Sample size calculation was based on the assumption that Losartan efficacy in patients already receiving prophylactic therapy would be half of the β-blocker efficacy reported in the study by Shores, i.e. from 0.02 to 0.01 with a SD of 0.03. Choosing a 0.80 power and a two-sided type I error of 0.05, the number of patients required by group (of similar size) would be 142. One hundred and fifty patients were required in each group.
Randomization. Randomization was made using an internet-based computerized randomization system stratified on study centre, age (< or ≥18 years old), and treatment with baseline preventive therapy. To ensure allocation concealment, Losartan and matching placebo were supplied to study sites in masked identical packages. Drug packages were given to the patients during follow-up visits.
Blinding. All study personnel involved in the operations of the study or with any potential site contact, such as medical monitors, remained blinded to treatment assignments from the time of randomization until after completion of statistical analysis.
Statistical Methods. Statistical analyses were performed on an intention-to-treat basis. For the primary endpoint, standard linear regressions were first used to estimate individual slopes for the change of aorta diameters over time, using all available measurements of each subject. A linear model (ANOVA) with adjustment on the stratification factors (centre, protective treatment at baseline, age <18 years old at baseline), was used to compare the mean slope of aortic dilatation between groups. All calculations were performed using R software version 3.1.1.
