Effect of etelcalcetide on cardiac hypertrophy in hemodialysis patients: a randomized controlled trial (ETECAR-HD)

Background Fibroblast growth factor 23 (FGF23) is associated with left ventricular hypertrophy (LVH) in patients with chronic kidney disease, and calcimimetic therapy reduces plasma concentrations of FGF23. It remains unknown whether treatment with the calcimimetic etelcalcetide (ETL) reduces LVH in patients on hemodialysis. Methods/design This single-blinded randomized trial of 12 months duration will test the effects of ETL compared with alfacalcidol on LVH and cardiac fibrosis in maintenance hemodialysis patients with secondary hyperparathyroidism. Both treatment regimens will be titrated to equally suppress secondary hyperparathyroidism while alfacalcidol treatment causes an increase and ETL a decrease in FGF23, respectively. Patients treated thrice weekly with hemodialysis for ≥ 3 months and ≤ 3 years with parathyroid hormone levels ≥ 300 pg/ml and LVH will be enrolled in the study. The primary study endpoint is change from baseline to 12 months in left ventricular mass index (LVMI; g/m2) measured by cardiac magnetic resonance imaging. Sample size calculations showed that 62 randomized patients will be necessary to detect a difference in LVMI of at least 20 g/m2 between the two groups at 12 months. Due to the strong association of volume overload and LVH, randomization will be stratified by residual kidney function, and regular body composition monitoring will be performed to control the volume status of patients. Study medication will be administered intravenously by the dialysis nurses after every hemodialysis session, thus omitting adherence issues. Secondary study endpoints are cardiac parameters measured by echocardiography, biomarker concentrations of bone metabolism (FGF23, vitamin D, parathyroid hormone, calcium, phosphate, s-Klotho), cardiac markers (pro-brain natriuretic peptide, pre- and postdialysis troponin T) and metabolites of the renin–angiotensin–aldosterone cascade (angiotensin I (Ang I), Ang II, Ang-(1–7), Ang-(1–5), Ang-(1–9), and aldosterone). Discussion The causal inference and pathophysiology of LVH regression by FGF23 reduction using calcimimetic treatment has not yet been shown. This intervention study has the potential to discover a new strategy for the treatment of cardiac hypertrophy and fibrosis in patients on maintenance hemodialysis. It might be speculated that successful treatment of cardiac morphology will also reduce the risk of cardiac death in this population. Trial registration European Clinical Trials Database, EudraCT number 2017-000222-35; ClinicalTrials.gov, NCT03182699. Registered on

Funding for this study was provided by the Vienna General Hospital (Amtsforschung) and an unrestricted grant from Amgen (reference# 20167811).

Introduction
Background and rationale 6a FGF23 is associated with left ventricular hypertrophy (LVH) in patients with chronic kidney disease and calcimimetic therapy reduces plasma concentrations of FGF23. The majority of patients with terminal renal failure treated by dialysis exhibit LVH and have a dramatically increased risk of sudden cardiac death. The EVOLVE study was able to show that a reduction of FGF23 after 20 weeks of cinacalcet therapy showed a trend towards a decrease in cardiovascular mortality, sudden cardiac death and heart failure Wolf et al. showed a direct effect of FGF23 on mouse cardiomyocytes leading to hypertrophy. The HEMO study found that higher FGF23 levels in hemodialysis patients were a predictor of cardiac events, infections and all-cause mortality The aim of the present trial is the analysis of the causal inference and pathophysiology of LVH regression by FGF23 reduction by calcimimetic treatment. Study patients will receive medication approved for sHPT treatment. The diagnostic procedures are non invasive and harmless.

4-6 6b
Subjects will either receive Etelcalcetide or Alfacalcidol in order to achieve a similar reduction in PTH in both study groups while FGF23 is elevated in the Vitamin D arm and suppressed in the Etelcalcetide arm in order to analyze the causality of FGF23 reduction on LVH and fibrosis. The performance of laboratory analyses and BCM measurements is possible in both centres. Eligibility criteria for individuals performing the interventions: The MRIs will be performed and analysed by C.L., a radiology specialist who is blinded for patient's treatment allocation. The strain echocardiograms will be performed by R.R., a specialist for internal medicine who is blinded for patient's treatment allocation. 29-30, Table 1 Interventions 11a Cardiac MRI: at baseline and after 12 months of treatment; performed on a dialysis free day. Strain echocardiogram: at screening and after 12 months of treatment.
Lung ultrasound: will be performed at screening. Body composition monitoring: performed at the bedside for screening and in two-month intervals Laboratory analyses: blood will be taken from the hemodialysis machine. Analyses will be performed in regular intervals. Investigational products: Patients will receive either Etelcalcetide or Alfacalcidol intravenously 3 times per week at a dosage adapted for laboratory results. Both drugs are approved medication for the patient cohort and are well tolerated.

6-12
11b Drug dose will be changed depending on laboratory analysis (PTH, Calcium, Phosphate). Patients will discontinue the study in the case of a kidney transplant, pregnancy or severe side effects. 6-8 11c Study medication will be injected into the patient's blood stream via the dialysis machine by dialysis nursing staff.

11-12 11d
There are no restrictions on calcium supplements, the dialysate calcium concentration, or the type or dose of phosphate binders prescribed. Participants randomized to Etelcalcetide can receive additional vitamin D analogs as a rescue therapy only when the investigator thinks that it is necessary to protect participant safety.

12
Outcomes 12 Primary endpoint: The change of LVMI from baseline after a year-long treatment with either etelcalcetide or alfacalcidol. Secondary endpoints -cardiac structure: Difference in left atrial diameter measured by cMRI (continuous variable in mm) after a year long treatment with either drug. Change in LVMI and LAD progression in either treatment group (%). Difference in cardiac fibrosis and progression measured by cMRI (T1 mapping/relaxation) and cardiac strain (strain in different myocardial segments). Differences in cardiac function (ejection fraction -%) and wall motion abnormalities (% change) Secondary endpoints -others: Changes in metabolites of the RAAS (pg/ml) using mass spectrometry ("RAAS fingerprint") under either treatment. Change from baseline serum levels of FGF23 (RU/mL) and s-klotho (pg/mL) under either drug. Change from baseline in PTH (ng/l), 25-OH-Vit-D (nmol/L) and 1,25-(OH)2-Vit-D (pg/mL), serum phosphate (mmol/l), serum calcium (mmol/l) under either treatment. Changes from baseline in proBNP (pg/ml), pre-and postdialysis TnT (ng/ml) in either medication group.

8-10
Participant timeline 13 After achieving informed consent patients will be screened, followed by a washout phase of 4 weeks (if patient is taking calcimimetic or vitamin D therapy). Following this the baseline MRI will take place. Each patient will then receive study medication for a duration of one year followed by the second MRI.

6-8
Sample size 14 Under the assumption based on published data that the median LVM/BSA of hemodialysis patients determined by cMRI is 100g/m 2 with a SD of 25g/m 2 and an expected treatment effect of delta LVMI of 20g/m2 roughly 2x25 patients are needed to detect this difference with 80% power at an alpha level of 0.05. Assuming a 10% attrition (drop out/loss to follow up) rate within the one year of follow up an effective sample size in the ITT analysis is 2x31 patients.
14 Recruitment 15 Patients from both dialyses centers will be asked to participate in the trial. The hemodialysis center of the Medical University of Vienna has 160 prevalent patients and the Vienna Dialysis Center has 300 prevalent patients. 6

Methods: Assignment of interventions (for controlled trials)
Allocation: Sequence generation 16a Randomization is performed following first MRI by a computer algorithm (www.meduniwien.ac.at/randomizer/web) stratified by residual kidney function (RKF), defined as an amount of 500ml or more urine volume per day and the center where patients are recruited. To ensure that comparison groups are of approximately the same size and balanced in each centre a block randomization (block size of 4) of anuria vs. residual renal function groups is used. A descriptive analysis will be performed and visualized with in the form of boxplots and histograms. Variance homogeneity of parameters will be tested with histograms. T-tests will be used for normally distributed parameters and Wilcoxon sign rank tests as a nonparametric alternative. The primary endpoint (change in LVMI from baseline to final measurement) will be analyzed by Analysis of Covariance. The baseline values for each patient will be used as a covariate in the model. Furthermore, variables which will be used as stratification factors for the randomization procedure will enter the model in order to adjust for possible remaining imbalances between the groups. The secondary endpoints will also be analyzed analogously.
A detailed statistical analysis plan will be completed before statistical analysis. Patients will be randomized to etelcalcetide or alfacalcidol after the first cMRI. Major confounders in our analysis will be dialysis vintage, fluid status and vitamin D treatment. Our approach to this will be the following: only patients with a dialysis vintage of above 3 months and below 3 years will be included in the study. BCM will be performed before enrollment and only patients who achieve their optimal dry weight through dialysis will participate in the analysis. Randomization will be stratified by residual kidney function (≥ 500ml urine/day vs. ≤ 500ml urine/day) and center of recruitment. We are expecting a homogeneous group of patients in both treatment arms.

13-14 20c
An intention to treat analysis will be performed.

Methods: Monitoring
Data monitoring 21a A data safety monitor board (DSMB) is convened to assess the safety of treatment as well as the nonsuperiority of one treatment over the other based on the methods described by O'Brien & Fleming. The data safety monitor board consists of physicians who are not connected to the project or the study team and are be blinded for the patient's treatment allocation.

21b
Interim analysis was performed by the board after completion of 10 cases in each treatment group (= one third of the planned study population). The Lan and DeMets extension of the O'Brien-Fleming stopping rules were applied. Based on the statistical analysis and discussion of the reported adverse events it was recommended by the DSMB to continue the study as planned.

12
Harms 22 The investigators ensure that adequate medical care is provided in any clinical situation, including emergencies. All AEs observed by the investigator or reported by subjects are properly captured in the subjects' medical records. This collection period is from the time of first dose of investigational product to 30 days after the last dose.
It is left to the investigator's clinical judgment to determine whether an AE is related and of sufficient severity to require the subject's removal from treatment. There are no financial and other competing interests for principal investigators for the overall trial and each study site.

16,17
Access to data 29 Only members of the research team will have access to the database.

12,17
Ancillary and posttrial care 30 An insurance for each patient enrolled in this study is taken out. The company Zürich provides the insurance cover.

17
Dissemination policy 31a The findings of this study will be published by the investigators in a scientific journal and presented at scientific meetings. The manuscript will be circulated to all co-investigators before submission. Confidentiality of subjects in reports/publications will be guaranteed. Rights to an authorship are according to GCP and Good Scientific Practice (GSP) of the Medical University of Vienna. It is planned that Dr. Katharina Dörr, will be the primary author of the manuscript. 17

Appendices
Informed consent materials 32 Each patient will receive an informed consent form in German language previously approved by the Ethics committee.
*It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 Explanation & Elaboration for important clarification on the items. Amendments to the protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT Group under the Creative Commons "Attribution-NonCommercial-NoDerivs 3.0 Unported" license.