Study design. ThIIo is a multicenter, randomized, single blinded clinical phase II trial assessing the efficacy of inhaled iloprost in the development and progression of ARDS in critically ill patients. Based on the risk of pulmonary hemorrhage which is very rare especially in patients with ARDS, study medication was unblended. For safety reasons, after treatment of 100 patients (day 28 after last dose IMP Patient 100) within the study an interim analysis for an increased risk for pulmonary hemorrhage ≥°III according to CTCAE Version 5.0 in the treatment (Iloprost) arm will be performed and the results discussed with the DSMB. It was approved by the local ethics committee (University Tuebingen, Germany) on June 4th 2019 and by BfArM (Bundesinstitut für Arzneimittel und Medizinprodukte, Bonn, Germany (EU/1/03/255/001)) on the 14th of March 2019. The trial is registered at EUDRA-CT (2016-003168-37) and at clinicaltrails.gov (NCT03111212).
Population. The target population for this clinical trial is adult critically ill patients suffering from ARDS. Patients will be included in the trial if they present with ARDS as defined by the Berlin Definition (Table 1 and (1)) and meet the inclusion criteria. The trial population will consist of both sexes. One hundred fifty intensive care patients suffering from ARDS will be included in the study at the Department of Anesthesiology, Eberhard Karls University Tübingen, Germany; the Department of Intensive Care and Intermediate Care, University Hospital RWTH Aachen, Germany; and the Department of Anesthesiology, University Hospital Münster (UKM), Münster, Germany.
Inclusion Criteria. Patients meeting the following criteria will be included: age ≥ 18 years, PaO2/FiO2 ≤ 300, bilateral infiltrates consistent with pulmonary edema on frontal chest radiograph, need for positive pressure ventilation via an endotracheal tube or noninvasive ventilation and no clinical signs of left atrial hypertension detected via echocardiography, or if measured, a Pulmonary Arterial Wedge Pressure (PAOP) less than or equal to 18 mmHg. The term “acute onset” is defined as follows: the durations of the hypoxemia criterion and the chest radiograph criterion must be ≤ 48 hours at the time of randomization. Patients must be enrolled within 48 hours of ARDS onset and no later than 7 days from the initiation of mechanical ventilation.
Exclusion criteria. The exclusion criteria are defined as follows: subject age < 18 years; time interval more than 7 days since the initiation of mechanical ventilation; more than 48 hours since the onset of ARDS; patient, surrogate or physician not committed to full intensive care support; positive pregnancy test at the time of screening; and contraindications against iloprost. These are defined as conditions in which the effects of iloprost on platelets might increase the risk of hemorrhage (e.g., active peptic ulcers, trauma, intracranial hemorrhage), severe coronary heart disease, myocardial infarction (within the last 6 months), decompensated heart failure, severe arrhythmias, unstable angina pectoris, pulmonary arterial hypertension caused by the occlusion of pulmonary veins, cerebrovascular events (e.g., transient ischemic attack, stroke) within the last 3 months, and congenital or acquired valvular defects with clinically relevant myocardial function disorders not related to pulmonary hypertension. Patients who received iloprost treatment for any indication within 48 hours prior to enrolment in the clinical trial or patients who were on thrombin inhibitors or NO within the previous 24 h before study randomization were also excluded. Additionally, patients dependent on the sponsor, investigator and their employees were not included in the study.
Study drug. The investigational medicinal product is iloprost (Ventavis ®; Drug Code: SUB14185MIG; ATC code: B01AC11), manufactured by Berlimed S.A., Spain (for Bayer Pharma AG, Germany); it will be used as a concentrate for use in nebulizers and will be administered by inhalation three times a day (20 µg per administration). The administration of the drug will occur at the same time each day ± one hour. In cases of severe adverse effects, the dosage will be reduced to 20 µg once a day (morning). Other dose modifications or temporary cessation of the study drug will not be allowed.
Iloprost is usually dissolved in 0.9% NaCl, which is used to keep the ventilator circuit moist as standard of care. Therefore, in the control group, NaCl 0.9% will be used to keep the airway circuit moist, which is the standard of care for the treatment of patients with pulmonary insufficiency (7). Looking at the pharmacokinetic and dynamic profile of iloprost, we have suggested an approach of an application of three times per day, with a dose of 20 μg, which seems to be an average dose in the trials reported up to date. The rationale behind this was that iloprost also exerts an anti-inflammatory effect that may last up to 6 h (29-33). Therefore, an administration of iloprost 3 times a day would allow a significant time frame per day to be covered by anti-inflammation due to this drug. The duration of 5 days was included in the trial because the pathophysiology of ARDS develops within the first few days and is progressive during that period.
Randomization. Randomization lists will be generated at the biostatistical center. Based on these lists, numbered envelopes will be provided and used for randomization.
Concomitant Medication and Treatments
Relevant additional medications and treatments such as vasopressors, inotropes, anti-infective agents, inhalative therapy or sedation, steroids and immunosuppressive therapy administered to the subjects on entry to the trial or at any time during the trial are regarded as concomitant medications and treatments and must be documented on the appropriate pages of the CRF, these data will be grouped according to class of medication. Depending on the substance the documentation varies in details (e.g. dosing).
Intervention plan. This study will consist of the following consecutive phases: study entry, treatment and follow-up. The time-points and trial procedures are listed in Table 2. All patients included in this trial will receive standard care for ARDS according to the ARDS network, with special consideration of lung protective ventilation strategies.
In this trial, ARDS patients present an emergency situation, such as the diagnosis of ARDS
requiring ICU admission and ventilation therapy, not allowing for any delay of diagnostic
work-up or therapy. Additionally, due to severe symptoms, the vast majority of patients who
meet the eligibility criteria for the trial are assumed to be unable to give consent in the acute
admission phase and legally authorized representatives (LAR) might not be available in most
cases. This is also in line with local regulations: e.g. §41 of the German Drug Law allows the
start of a treatment in an emergency situation without prior consent in case the immediate
treatment is necessary to save the patient’s life, recover the patient’s health or ease the
patient’s suffering. In this situation the consent of an independent physician not directly
involved in the study conduct will be sought before the beginning of any study-related activity. The consent has to be obtained as soon as the patient is able to give consent or a LAR is available. Independently, the personal consent will be obtained from each patient after recovering consciousness and competence for decision making or by a legal representative in cases recovering is not achieved during study duration (i.e., day 27). When possible however the patient or his legal representative is to be informed both in writing and verbally by the investigator before any study-specific procedure is performed. Each patient or his legal representative will be informed about the modalities of the clinical study in accordance with the provided patient information. Informed consent from the patient will be obtained using a form approved by the Ethics committee (EC) of the Universitätsklinikum Tübingen or the local EC if the patient is treated in a collaborating institution.
The treatment group will receive 20 μg of nebulized iloprost three times per day for 5 days in addition to standard care. Iloprost will be measured in blood samples to determine the serum levels within this setting.
The control group will receive nebulized 0.9% sodium chloride (NaCl) with an equal volume three times per day for 5 days. After 5 days, the trial treatment will be complete (Figure 1). Blood samples will be drawn at defined time points for a variety of biomarkers to better assess the associations among coagulation, inflammation and iloprost treatment. Key cointerventions (infection control, aspiration precautions, fluid and transfusion) will be standardized across all patients. Mechanical ventilation will be standardized (see supplemental file).
Hospital survivors will undergo a brief follow-up phone survey to assess functional status (Barthel Index), a health-related questionnaire, and frailty (VES) six months after enrollment. The patients will be visited daily until day 28 or until discharge from the ICU, which could be beyond day 28. If discharged, the next visit will be on day 90, if patients are still in the ICU, there will still be daily visits until this time point. Data will be collected according to the study procedure until then. Each visit will consist of a clinical examination, a blood sample, assessment of the functional capacity through the Barthel Index and assessment of the severity of illness through the SOFA score. All data will be recorded on an eCRF form; this will be used as a visit diary. Blood samples will be drawn at defined visits for a variety of biomarkers to better assess the associations among coagulation, inflammation and iloprost treatment (Table 3).
The primary objective and endpoint is to assess the effect of iloprost on the improvement of oxygenation (PaO2/FiO2 ratio) in patients suffering from ARDS.
For secondary objectives the absolute incidence of the following parameters will be determined:
- Overall survival in the 90-day follow-up period (90-day all-cause mortality).
- Duration of mechanical ventilation support
- ICU length of stay
- Ventilator-associated pneumonia
- Pulmonary hemorrhage
- Gastrointestinal hemorrhage
- Pulmonary embolism
- ICU-acquired weakness
- Discharge location (home, skilled nursing facility, rehabilitation)
The exploratory objectives are 6-month survival, QOL (SF12), functional status (Barthel Index), and frailty (VES) assessed by phone follow-up interview.
The following parameters will be used to determine the efficacy:
- Improvement of oxygenation (PaO2/FiO2) on a daily basis in relationship to baseline
- Overall survival in the 90-day follow-up period
- Decrease in duration and severity of ARDS
- SOFA scores: will be calculated based on data in hospital records
- Duration of mechanical ventilation support: documentation in hospital records
- ICU length of stay: documentation in hospital records
- Ventilator-associated pneumonia: documentation of microbiological findings in hospital records
- Incidence of barotrauma: documentation of ventilator parameters in hospital records
- Reduced morbidity assessed through SOFA score, also according to the incidence of complications and increased functionality assessed through the Barthel Index
- Delirium: documentation (e.g., CAM-ICU) in hospital records
- ICU-acquired weakness: documentation in hospital records
- Discharge location: documentation in hospital records, phone call
The demographic parameters at enrolment include age, sex, race, ICU admission diagnosis and comorbidities (such as diabetes, existing malignancy, any kind of pre-existing pulmonary disease, and hypertension).
The main clinical data during the ICU daily assessment are as follows:
- Laboratory data: Blood count, procalcitonin, IL-6, creatinine, urea, PTT, D-dimers, INR, AST, ALT, albumin, CHE, BNP
- Ventilation support
- Invasive or noninvasive ventilation
- Prone positioning Yes/No
- Max Pmax on daily basis
- max Pmean on daily basis
- PEEP on daily basis
- PEEP on daily basis
- Driving pressure at max Pmax
- compliance on daily basis
- FiO2 on daily basis
- High frequent ventilation Yes/No
- Tracheotomy Yes/No
- Any relaxation therapy performed Yes/no
- ECMO therapy: Cannulation, blood flow, FiO2, Sweep gas
- Volume resuscitation: Volume crystalloid/day, volume colloid/day, volume albumin/day, daily balance sheet
- Transfusion: Units red blood cells/day; units thrombocytes/day
- Substitution of clotting factors if necessary: Fibrinogen, PPSB, factor XIII, cyclocaprone, vWF, factor VIII, factor VII
- Anticoagulation: Fractionated heparin (cum. dose/day), unfractionated heparin (cum. dose/day), epoprostenolum (cum. dose/day), argatroban (cum. dose/day), others (cum. dose/day)
- Hemodynamic parameters: Heart rate, arrhythmia, lowest and highest MAP per day, lowest and highest ZVD per day, cardiac index, lowest and highest SpO2 per day, lowest and highest lactate per day (mmol/l) SvO2,
- Renal replacement therapy: AKIN criteria. CRRT, CVVHD, CVVHDF, CVVHF, intermittent dialysis, SLED/dialysis, anticoagulation, duration, dose
- Delirium: CAM-ICU score
- Immunosuppressive therapy: Yes/no
Weekly assessments of the ICU will include the following:
- Differential blood count
- Creatinine clearance
- ECMO post oxygenator PaO2
- SOFA score
- Assessment at discharge
- Chronic renal failure at discharge
- Hepatic failure at discharge
- Length of stay in the ICU
- Length of stay in the hospital
- Discharge from hospital to a nursing home
- Discharge from hospital to home
- Discharge from hospital to a rehabilitation unit
- Residence in nursing home at 6 months
The final assessment will consist of the following:
- Days of ECMO support
- Ventilator days
- Need for mechanical ventilation at home
- Incidence of pulmonary hemorrhage defined by an indication for blood transfusion, radiological finding, or a decrease in oxygenation
- Incidence of barotrauma
- Incidence of pleural drainage
- Incidence of pulmonary embolism defined by the following parameters:
- New hypotension
- Sign of right ventricular failure on echocardiography
- CT-scan (optional)
- Incidence of gastrointestinal bleeding defined by the following parameters: upper gastrointestinal bleeding, blood vomiting, lower gastrointestinal bleeding, melena, indication for blood transfusion, endoscopic diagnosis/intervention
- Incidence of cerebral hemorrhage defined by following parameters: Impairment as measured by the Glasgow coma scale, CT scan
- Infections: Incidence of positive blood culture, incidence of pneumonia, incidence of wound infection, incidence of peritonitis, incidence of surgical intervention due to infection, incidence of bacterial infection, incidence of fungal infection, incidence of viral infection, incidence of MRGN infection
- Anti-infective therapy: Generic, duration, incidence of changing antiinfective therapy due to inadequate treatment
- Incidence of surgical intervention
Data collection and management
Case Report Form
The trial case report form (CRF) is the primary data collection instrument for the trial. For this project, electronic Case Report Forms (eCRFs) will be used. Entered data will be subjected to plausibility checks directly implemented in the CRF, monitoring and medical review. The trial master file, the CRFs, and other material supplied for the conduct of the study will be retained by the sponsor/CRO according to applicable regulations and laws. The investigator(s) will archive all trial data (source data and Investigator Site File (ISF), including the subject identification list and relevant correspondence) according to the ICH Consolidated Guideline on GCP and local laws or regulations.
Study Population Definition
The study population will consist of the following: to be assessed for eligibility: (n = 300); to be assigned to the trial: (n = 150); to be analyzed: (n=150 in the intention to treat analysis, other endpoints n=120). The sample size and power consideration refers to 120 evaluable patients, and it is assumed that the power will not be decreased in the analysis of the ITT population using multiple imputation. Furthermore, baseline adjustment will not be taken into account, which leads to a conservative sample size estimation.
In a previous study on iloprost with 20 patients, an increase from 177±60 to 213±67 was observed for the PaO2/FiO2, which was significant at the 0.01 level (27). Recalculation shows that the intraindividual standard deviation must have been considerably smaller as a p-value of 0.01 corresponds to an effect size of 0.93 (intraindividually) and thus to an intraindividual standard deviation of approximately 40 in this study.
In our study, we can show effect sizes of 0.525 assuming 116 error degrees of freedom, taking into account 1 day for baseline adjustment and 3 days for the study center (inquiry, power 80%, level of significance 0.05 two-sided, t-test). If we assume the recalculated standard deviation from the previous study in our study (which is still conservative due to the linear baseline adjustment used in our study), an (interindividual) effect size of 0.525 corresponds to a difference of approximately 21 in the PaO2/FiO2 ratio in the treatment arm compared to the control arm. This seems to be a reasonable and relevant effect.
Analysis of Primary Variables
The primary endpoint of PaO2/FiO2 at day 6 after the baseline will be analyzed daily using a baseline adjusted analysis of covariance model with the last measurement of the PaO2/FiO2 ratio before treatment as the baseline, with the study arm as a second level factor. The study center will be included in the analysis as a nuisance factor. Additionally, an interaction term between baseline and treatment will be included in the model if this term is significant. In the case of interaction, the main effect will be retrieved for the arithmetic mean of the baseline values using the centered variable for PaO2/FiO2. Multiple imputation will be applied in the intention to treat population of patients receiving at least one dose of treatment or the control.
Analysis of Secondary Variables
Statistical analysis of the prespecified secondary endpoints will be performed with descriptive and exploratory statistical methods according to the scale and observed distribution (absolute and percentage frequencies, chi-square tests, logistic regression models for categorical variables; means and standard deviations, medians and quartiles, or ranges with t-tests or Mann-Whitney tests and linear regression models for continuous variables; Kaplan-Meier curves, log-rank tests and Cox proportional hazard models for censored data). P-values will be reported but should not be considered part of the confirmatory analysis.
Planned subgroup analyses will be performed according to the following:
- Sex and race (only for subgroups larger than 40 subjects)
- Patients with increased pulmonary arterial pressure
- Direct vs. indirect lung injury
- Age stratified by decades
For safety reasons, after the enrolment of 20 patients (day 28 after last dose IMP Patient 20), an interim analysis of the following will be performed:
- An increased risk of pulmonary hemorrhage ≥°III according to CTCAE Version 5.0 in the treatment (iloprost) arm and
- Levels of IMP in the serum.
The results will be discussed with the DSMB. The DSMB has to assess whether the result allows continuation of the study as planned.
Moreover, after treatment of a total of 100 patients (day 28 after the last dose IMP Patient 100), an interim analysis of an increased risk of pulmonary hemorrhage ≥°III according to CTCAE Version 5.0 in the treatment (iloprost) arm will be performed, and the results will be discussed with the DSMB. The DSMB has to assess whether the result allows continuation of the study as planned.
Moreover, in the case of the following situations, a premature termination of the trial must be considered:
- Substantial changes in risk-benefit considerations
- New insights from other trials and
- Insufficient recruitment rate.
The biometric report will be delivered according to the SOP BI07 of the statistical center (IKEaB). In summary, the report will contain sections on the statistical methodology, preprocessing of data, and the descriptive, exploratory, and confirmatory analyses. It will be reviewed by the PI before presenting the final version.