Study design and setting
The RECUT trial is a prospective, randomized, double-blind, placebo-controlled, non-inferiority trial in a primary care setting. The coordinating study center in Liestal, Switzerland, consisting of the principal and co-principal investigators, study coordinators, and study physicians, organizes all global activities in connection with the trial, is responsible for data management, and supervises endpoint adjudication. It furthermore acts as a trial steering committee. The study is conducted in collaboration with general practitioners (GP) in Northwestern and Central Switzerland, as well as in the Innsbruck area, Austria, who identify potential participants and perform study related assessments. A list of participating GPs is available from the corresponding author on reasonable request.
Based on a sample size calculation, a total of 470 patients shall be enrolled, with a 1:1 allocation ratio to the experimental and conventional arm. Participating GPs assess patients with AECOPD for their eligibility criteria and perform diagnostic tests of their choice. Patients who fulfill the eligibility criteria and who are willing to participate receive 40 mg of oral prednisone per day for either five days (standard treatment group) or three days, followed by two days of placebo (experimental group). Antibiotics (amoxicillin/clavulanic acid, 625 mg 3/d, for seven days) are administered to all patients with a serum CRP (C-reactive protein) ≥ 50 mg/l at any of the study visits, or known diagnosis of bronchiectasis, as well as patients presenting with Anthonisen-type-I exacerbations (18). Additional initial treatment and further treatments during follow-up time are determined and documented by the treating GP. Participants are assessed with respect to primary and secondary endpoints after three and seven days by their treating GPs. The coordinating study center contacts patients by phone for further evaluation on days 30, 90, and 180. In cases where patients cannot give sufficient information with regard to the endpoints in the phone interview, their GP is interviewed in addition.
Patient characteristics
The first patient was enrolled in the study in August 2015 and recruitment is expected to conclude by September 2021. Enrolled patients must meet the following inclusion criteria (see also Table 1): age ≥ 40 years, history of ten or more pack-years of smoking (past or present smokers), airway obstruction (defined as FEV1/forced vital capacity (FVC) ratio ≤ 70%), and current AECOPD. The latter is defined by the presence of at least two of the following symptoms: change of baseline dyspnea, change of cough, change of sputum quantity or purulence. Key exclusion criteria include asthma/COPD overlap syndrome with predominant asthma component, initial necessity of hospitalization, known severe immunosuppression, and severe coexisting disease with a life expectancy of less than six months. Women who are pregnant or breastfeeding, premenopausal women with insufficient contraception, and patients with diagnosis of tuberculosis are also excluded from the study.
Table 1 Eligibility criteria
|
Inclusion criteria
|
|
Exclusion criteria
|
- Age ≥ 40 years
|
|
- Initial necessity of hospitalization
|
- Informed consent as documented by signature
|
|
- Previous enrollment into the current study
|
- History of ≥ 10 pack-years of smoking
(past or present smokers)
|
|
- Asthma/COPD overlap syndrome with a predominant asthma component
|
- Airway obstruction, defined as FEV1/FVC ≤ 70%
|
|
- Diagnosis of tuberculosis
|
- Current acute exacerbation of COPD by clinical criteria,
defined by the presence of at least two of the following:
× Change of baseline dyspnea
× Change of cough
× Change of sputum quantity or purulence
|
|
- Severe coexisting disease with life expectancy < 6 months
|
- Known severe immunosuppression or immunosuppression after solid organ or stem cell transplantation
|
|
|
- Inability to follow study procedures, e.g. due to language problems, psychological disorders, dementia, etc. of the participant
|
|
|
- Participation in another study involving an investigational drug
|
|
|
- Women who are pregnant or breast feeding
|
|
|
- Premenopausal women with insufficient contraception and anamnestic risk for pregnancy
|
FEV1: forced expiratory volume in one second. FVC: forced vital capacity. COPD: chronic obstructive pulmonary disease.
|
Blinding and Randomization
Identical looking blister packs with daily doses of 40 mg prednisone for either five days (standard treatment arm) or three days, followed by two days of placebo (interventional arm), are packed in a 1:1 ratio in the hospital pharmacy of the University Hospital Basel, Switzerland, in a GMP (Good Manufacturing Practice) regulated environment. Each blister pack is labelled with a computer-generated random alphanumeric code. A concealed envelope marked with this alphanumeric code on the outside contains group allocation and is kept safe at the study center until analysis of the final data. Depending on the expected number of eligible patients, each recruiting GP receives a certain number of the pre-randomized blister packs and hands them out to participating patients. Trial participants, general practitioners, outcome assessors and data analysts are blinded to group allocation. Unblinding is permissible if necessary for the urgent medical treatment of a participant.
Study intervention and assessments
The schedule of enrollment, interventions and assessments is presented in Fig. 1. On day one (inclusion visit), the treating GP informs patients presenting with AECOPD about the trial, checks eligibility criteria, gets their written informed consent, and performs a general clinical assessment including vital signs, dyspnea assessment (mMRC) and COPD assessment test (CAT). A blood sample for CRP, plasma glucose, and leucocyte cell count is taken and FEV1 and FEV1/FVC are assessed by spirometry. Participating patients receive 40 mg of oral prednisone per day for either five days (standard treatment group) or three days, followed by two days of placebo (experimental group). Additional newly started exacerbation medication besides the study medication will be documented. Furthermore, treating doctors can re-evaluate and change the treatment at any stage of the trial if necessary (e.g. in case of worsening of patient’s clinical condition). The implementation of either trial arm will not require any alteration of usual care. Thus, all concomitant treatment or medication considered necessary by treating doctors, including other newly started exacerbation medication, are permitted and their use will be recorded in the case report form (CRF).
The follow-up visits will take place on day three (+/- one day) and day seven (+/- one day) and consist each of a general clinical assessment, a blood sample, and an assessment of clinical course regarding treatment failure and need for hospitalization. Furthermore, any changes in medication (including COPD baseline medication and exacerbation medication), cumulative GC dose, other interventions such as COPD self-management and smoking cessation, as well as clinically manifested side effects of GCs will be documented. During the second follow-up visit on day seven, the detailed medical history with regard to COPD is recorded and a spirometry is performed additionally. Participants are further followed-up by phone on days 30, 90, and 180 after inclusion into the study (+/- seven days each). The phone interviews include dyspnea (mMRC) and COPD assessment (CAT) questionnaires, as well as questions regarding sputum, cough, any change in medication, and hospitalizations in the intervening time in order to assess re-exacerbation.
Participants have the possibility to withdraw from the study at any given time. Patients, who prematurely withdraw from the study, are nonetheless encouraged to attend follow-up appointments and data collected until the time of withdrawal will be analyzed in the intention to treat analysis.
Since this study tries to evaluate a real-life situation in an outpatient setting as closely as possible and needs to be simple with regard to feasibility, patients’ adherence is not assessed. Patients are encouraged to return empty blisters to their GPs though, which will be useful for double-checking if patients did take their medication according to plan. Unsatisfactory compliance concerning the study medication shall be reported to the local study center and the coordinating center.
Outcomes
The primary endpoint is time to next exacerbation during a six month follow-up period, which includes re-exacerbation during the index exacerbation (i.e. treatment failure). Exacerbation is defined as acute-onset worsening of the patient’s condition beyond day-to-day variations requiring interaction with a healthcare provider (19). We chose time to next exacerbation as our primary endpoint in order to evaluate effectiveness of the shorter steroid treatment. According to the studies by Leuppi et al. (6), Niewoehner et al. (9), and Aaron et al. (10), who investigated treatment failure rates, relapse rate, and time to relapse in AECOPD patients under GC, time to next exacerbation (which includes treatment failure) seems to be a valid measurement for effectiveness.
Secondary study outcomes are cumulative GC dose, GC side effects and complications, change in FEV1, hospitalization rate during index exacerbation and during follow-up, clinical outcome assessed by CAT and mMRC, as well as overall mortality. Cumulative GC dose and GC side effects are assessed in order to investigate safety of short-term and standard steroid treatment duration. Furthermore, change in FEV1, hospitalization rate during index exacerbation and during follow-up time, as well as clinical outcome and overall mortality are evaluated to compare effectiveness of different durations of systemic corticosteroid treatment.
Statistical Analysis
It is hypothesized that the experimental treatment (three days corticosteroid treatment) is non-inferior to the conventional treatment (five days corticosteroid treatment) with regard to the primary endpoint. For this, a Cox proportional hazards regression model will be fitted to the data. Non-inferiority will be concluded if the two-sided 95% confidence interval of the hazard ratio between the experimental and the control arm lies entirely below the critical hazard ratio defined as (see Formula 1 in the Supplementary Files)
where t is a fixed point of time, λe and λc are the hazard rates, πet and πct are the proportions of event-free patients at time t in the experimental and conventional arm respectively, and m is the non-inferiority margin, expressed as the additional proportion of patients having had an event in the experimental arm, assuming that the occurrence of events follows an exponential distribution (20). This approach is partly based on the methodology described in a previously performed study of our research group (6, 21). Following the recommendations of the CHMP (Committee for Medicinal Products for Human Use), a two-sided 95% confidence interval is used to assess non-inferiority (22). In case of missing data, GPs will be contacted by the study team with the aim to complete patients’ records, while data imputation is not planned. All statistical analyses will be performed on the per-protocol data set, complemented by a sensitivity analysis based on the intention-to-treat data set. Subgroup analyses or interim analyses are not planned.
Sample size calculation
For the estimation of necessary sample size, we assumed an exacerbation rate of 30 to 40 percent following an exponential distribution, and a 15 percent drop-out rate evenly distributed within the six months follow-up period, for both the interventional and the conventional arm. The non-inferiority margin was defined as a 15% increase of exacerbation rate within six months, the significance level was chosen to be 5%, and the desired power 80%. A simulation and a cox proportional hazards regression model were used to determine hazard ratios and 95% confidence intervals for the simulated data sets, which led to a sample size of N = 466 (95 % CI 461 to 471) for an exacerbation rate of 30%, and N = 464 (95 % CI 459 to 469) for an exacerbation rate of 40%, respectively. Therefore, we aim to include N = 470 patients into the study. Sample size will be re-estimated after approximately half of the initially estimated number of patients have reached the six month follow-up. If necessary, the sample size will be increased. We will re-estimate the exacerbation rates in a blinded manner, based on the overall observed exacerbation rate as described by Friede et al. (23). Since no hypothesis test is performed, no p-value adjustment to control the type-I-error rate is needed. If enrolment goals are anticipated not to be met, further geographical expansions of the study will be considered.
Safety, quality assurance and control
During the entire duration of the trial, all adverse events (AE) and serious adverse events (SAE) are collected, fully investigated, and documented in source documents and CRFs. GPs are obliged to report SAEs within 24 hours to the Sponsor-Investigator and the local project leader, who, in case of death, reports to the local ethics committee within seven days. AEs and SAEs are followed up until resolution or stabilization. Important protocol modifications are communicated to the relevant parties via E-Mail newsletters and personal phone calls.
All patient data are treated confidentially and are stored and analyzed in a coded way. Personal contact information, which is needed for follow-up phone calls, is stored separately and only accessible for the staff members executing these phone calls. Information on data monitoring and auditing can be found in the Appendix (file 1: SPIRIT checklist).