Trial design, locations and participants
A double-blind, randomized controlled clinical trial of ivermectin versus placebo was conducted simultaneously at four healthcare institutions in Colombo district; the National Institute of Infectious Diseases (NIID)-Angoda, Base Hospital-Homagama, District Hospital-Wetara, and intermediate care centre (ICC)-Kandawala which was affiliated to Colombo South Teaching hospital, Kalubowila, enrolling participants from 29/07/2021 to 17/03/2022. During this study period, delta (sub-lineages AY.28 and AY.104) was the dominant variant of SARS-CoV-2 from July to October 2021 [21] and then Omicron up to the early part of 2022 [22].
Inclusion and exclusion criteria
Patients over 18 years of age, hospitalized within 4 days of onset (as per patient’s history recorded by the doctor at the admission unit) of mild to moderate symptoms of Covid-19 and confirmed by SARS-CoV-2 reverse transcriptase- polymerase chain reaction (RT-PCR), were enrolled. The severity of patients was categorised according to the World Health Organization (WHO) severity criteria (the latest version available at the time of conducting the trial) [23]. Patients with severe disease were excluded. Pregnant females, breast-feeding mothers, those with HIV co-infection, known persons with an allergy to ivermectin or anthelmintics, patients with critical stage such as those who require mechanical ventilation or anticipated impending need for mechanical ventilation, patients recruited to any other trials simultaneously, and patients who have already been started on ivermectin were excluded.
Interventions
Ivermectin 6 mg tablets (manufactured according to Good Manufacturing Practice (GMP) by Popular Pharmaceuticals Limited, West Panthapath, Dhaka 1207, Bangladesh) from one single batch was supplied by the local agent (ABC Pharma Services (Pvt) Ltd, Colombo, Sri Lanka) for the study. The placebo tablet containing all inactive excipients to match the ivermectin tablets was manufactured by the State Pharmaceutical Manufacturing Corporation (SPMC, Sri Lanka), to be identical in appearance, shape, weight, taste, colour, smell, and texture. Recruited patients were randomly allocated into two groups (ivermectin arm and placebo arm). The allocation ratio was one to one (1:1). Each participant received oral ivermectin 24 mg (400 µg per kilogram of body weight for an average 60 kg patient) or placebo daily for 05 days. For all the interventions (medication, data collection, and investigations) the date of intervention /enrolment was considered as day zero. Both ivermectin and placebo arms received standard care according to a circular issued by the Ministry of Health which included symptomatic treatment (as and when necessary) with age-appropriate doses of paracetamol, omeprazole, domperidone, and salbutamol nebulisation. Ivermectin arm had 45/127 (35%) and the placebo arm 49/122 (40%) patients who continued with their long-term medication with antihypertensives, anti-diabetic and lipid-lowering drugs.
Outcomes
The main primary outcome measure was the difference in the viral load (calculated using the natural log Cycle threshold (Ct) value of RT- PCR test for SARS-CoV-2) between D10 and day zero in the two groups. The viral load was calculated using the Ct values as done in a previously reported study [24]. The viral load was calculated using the formula given in the protocol of the paper by Chen at al (2021) with a revision of the cut-off Ct value for our study [24]. The maximum Ct value in our study was 40; the maximum Ct value in the reference paper was 45. Therefore, the formula to calculate log base 10 viral load was revised as - (40 - Ct)/ log (base2)10.
Clinical progression of the patient was also measured using WHO Clinical Progression Scale measured on days three, six, ten, 14, 21, and 28 [25].
Secondary outcomes included RT-PCR on day five, improvement of the symptoms by day six and by day ten of intervention using the WHO clinical progression scale, and adverse effects noted in the two arms. Each symptom was scored as absent (score 0), mild (1), moderate (2), or severe (3) [24].
The outcome of viral load could be studied only in those who were discharged at Day 10. This was a change to trial outcomes after commencement.
Sample size
For sample size calculation, retrospective data from a sample of 30 hospitalized patients were analyzed using their Cycle threshold (Ct) values for SARS-CoV-2 RNA on day one (D1) and day ten (D10). The difference in the natural log Ct values of D1 and D10 had a mean of 0.23 and standard deviation of 0.27. The sample size was calculated to observe at least a 50% difference in the viral load between treatment groups (from 0.23 to 0.34), assuming a standard deviation of 0.3. A sample size of 106 per arm was determined to have 80% power to detect this difference at a significance of 0.05. The final sample size was calculated to be 236 (118 per arm) with a dropout rate of 10%.
Data Safety Monitoring Board (DSMB) evaluated all adverse events after 50% of patients were recruited and granted approval to continue the trial. Termination of study was to be decided by DSMB, trial related regulatory authorities, who was to be notified of any major adverse event.
Randomization and double blinding
For randomization computer-generated random numbers were created by the statistician in variable blocks of four, six, or eight. Ivermectin or placebo tablets were packed into identically labelled packages numbered from 1 to 400 according to the computer-generated random numbers. The list was kept sealed and locked until the completion of the trial. The label of each pack indicated only the randomization number and directions for taking the tablets. Patients were screened for eligibility and recruited by medically qualified pre-intern research assistants guided by the clinical investigators under whose care the patients were admitted. The clinical and other data and the randomisation number were entered into each patient’s case record form (CRF).The study was blinded to participants, investigators, data analysts, healthcare providers and outcome assessors.
Approvals
Ethics approval was obtained from Ethics Review Committee of Faculty of Medicine, University of Colombo, Sri Lanka (EC-21-EM02). The trial was registered at the Sri Lanka Clinical Trials Registry [26] (SLCTR/2021/020) before patient recruitment. Good clinical practice guidelines for the conduct of clinical trials in Sri Lanka of the National Medicine Regulatory Authority (NMRA) were adhered [27]. The protocol was approved by the Clinical Trials Evaluation Committee (CTEC) of NMRA (CT/P38/15/2021). Ministry of Health, Sri Lanka also approved the trial (ETR/AC/M3/33/2021). Relevant heads of each institution where the study was conducted granted administrative approval. Informed written consent was obtained from all participants.