Trial Design and Oversight
In this single-center, open-label, randomized, controlled, parallel-group and clinical trial, eligible patients with confirmed SARS-CoV-2 infections were randomly assigned in a 1:1 ratio to receive one of the two following therapeutic regimens: 1) IFN-β1a (Recigen) (Subcutaneous injections of 88 µg (24,000 IU) on days 1, 3, 6) + lopinavir/ritonavir (Kaletra) [intervention group], 2) IFN-β1a (Recigen) (Subcutaneous injections of 44 µg (12,000 IU) on days 1, 3, 6) + lopinavir/ritonavir (Kaletra) (400 mg/100 mg twice a day for 10 days, orally, in all two arms) [control group]. Intervention and control groups received standards of care including necessary oxygen support and non-invasive or invasive mechanical ventilation. The study was commenced at August 20th and terminated at September 4th, 2020 at Loghman Hakim hospital.
We tried to collect our data on a potential treatment regimen by performing a pragmatic randomized controlled trial for moderate to severe COVID-19 patients without sacrificing any critical investigational component in a reasonable time frame.
Due to emergency situation as well as brisk rate of infected patients, blinding of all caregivers was not possible. All patients and outcome assessor were blinded to the arms of study.
Deputy of the vice chancellor of research and technology in Shahid Beheshti University of Medical Sciences provided all supports.
All patients were randomly assigned to each arm of the study via permuted block randomization (each block-sized for three or six patients) in order to minimize allocation bias in each study group. The sequence of the randomization was generated via “randomizeR” package using R project for statistics computing version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). All randomization codes for individuals enrolled to the study, were sealed in unrecognizable opaque envelopes by the responsible statistician for randomization. The investigator (IAD, MMR, and FH) enrolled the patients and only then open envelopes to assign patients to the different treatment groups. This method of allocation concealment leads to minimizing selection and confounding biases.
The Ethics in Medical Research Committee of the Shahid Beheshti University of Medical Sciences had been confirmed the study. Signed informed consents were obtained from all participants or their legally authorized representatives. The trial was carried out under the declaration of Helsinki and the International Conference on Harmonization of Good Clinical Practice (ICH-GCP) guidelines for the conduct of clinical trials on human participants. The trial has been registered with ClinicalTrials.gov, NCT04521400 and the full protocol is freely available on the BMC Trials(12).
Patients
In this randomized controlled trial, patients were assigned to the intervention group or the control group. The inclusion criteria were as follow: age ≥ 18 years, oxygen saturation (SPO2) ≤ 93% or respiratory rate ≥ 24, presence of at least one of following manifestations on admission: Cough, shortness of breath, nasal congestion/ discharge, myalgia/arthralgia, radiation contactless body temperature ≥ 37.8, diarrhea/vomiting and headache or fatigue. The patients’ symptoms must be in acute phase (≤ 14 days).
Exclusion criteria were refusal to participate, receiving drugs with interactions with lopinavir/ritonavir or Interferon-β 1a, a fivefold rise in serum AST/ALT relative to upper limit of normal laboratory results, pregnant or lactating women, history of alcohol or drug addiction in the past 5 years and intubated less than one hour after admission to the hospital.
Clinical And Laboratory Monitoring
For assuring safety, daily monitoring for adverse effects (AEs) and treatment-related AEs, vital signs (pulse rate, respiratory frequency, body temperature, and blood pressure), spo2, Glasgow Coma Scale (GCS) and laboratory tests was performed. Time, severity, symptoms of adverse effects and their relation with aforementioned drugs on a daily basis were recorded.
Before enrollment, nasopharyngeal swab samples were obtained from each patient. The samples were tested using RT-PCR kits including Liferiver (W-RR-0479-02, China) for E, N, and Rdrp genes. Patients’ data were recorded on paper checklists and Hospital Information System (HIS) by FH and MMR. The recorded data were entered into a pre-designed EXCEL sheet and later confirmed by a third investigator (IAD).
Outcome Measures
Primary outcome measure was TTCI defined as the time from enrollment to discharge or decline of two steps on the seven-step ordinal scale. Beigel and colleagues in a posthoc analysis of an influenza study introduced a six-step ordinal scale. WHO R&D Blueprint Team (Accessed May 15, 2020, at https://www.who.int/teams/blueprint/covid-19) have recommended a nine-step ordinal scale for COVID-19. In the current study the utilized seven-step ordinal scale consists of the subsequent categories: (I) Not hospitalized, and has no activity limitations; (II) Not hospitalized, but has activity limitations; (III) Hospitalized, but does not need any supplemental oxygen; (IV) Hospitalized, and needs supplemental oxygen; (V) Hospitalized, and needs either High-Flow Nasal Cannula (HFNC) or non-invasive ventilation; (VI) Hospitalized, and needs invasive ventilation; and (VII) Dead.
Secondary outcomes include mortality from the date of randomization until day 21, by which all of the patients will have at least one of the following outcomes: 1) A decline of two steps on the seven-step ordinal scale, 2) Hospital discharge or 3) Death. SpO2 improvement defined as the difference between the last and the first recorded measurement during the hospitalization, using pulse oximetry; length of stay in the hospital until the date of discharge from hospital or death from any cause, whichever came first; the incidence of new mechanical ventilation uses from the date of randomization until day 21. Follow-ups of discharged patients were done by utilizing telemedicine visits, online, or over the telephone.
Statistical Analysis:
Total sample size was calculated according to the Latouche and colleagues approach for estimating sample size in survival analyses with 90% power, alpha = 0.05, Hazard Ratio (HR) of 2.0 (as the ratio of the hazard rates of TTCI) and assuming that 60% of patients would reach the primary outcome. The calculations were carried out using Package ‘powerSurvEpi’ in R and accounted for a dropout rate of 10%. According to above-mentioned assumptions, 168 patients should have been recruited for this trial. Patients who failed to reach the primary endpoint (TTCI) or died prior to day 21 were regarded as right-censored in analysis.
Kaplan–Meier (compared with a log-rank test) was used to analyze the TTCI. Cox proportional-hazards model was also applied to calculate the HRs with 95% Confidence Intervals (CIs). All the participants who had undergone randomization were included in Intention-To-Treat (ITT) analysis (Fig. 1).
For categorical variables frequencies and percentage were employed. For distributed continuous variables Mean (SD) and for none-normally distributed variables, median (interquartile range) were used, respectively. Categorical variables were analyzed using chi-squared or the Fisher’s exact test (when the expected frequency was less than 5 in one or more cells). Continuous variables were evaluated using T-test (for normally distributed) and Mann-Whitney U test (for non-normally distributed). A p-value of < 0.05 was considered to be statistically significant. All of the carried-out tests were two-tailed. R software version 3.6.1 was used to perform the statistical analyses.