Ivermectin in mild and moderate COVID-19 (RIVET- COV): a randomized, placebo-controlled trial


 Introduction: Till date, no drug has shown definite benefit in non-severe COVID-19. Ivermectin is an antiparasitic drug which has in-vitro efficacy in reducing coronavirus-2 (SARS-CoV-2) load in severe disease. Objectives: To determine if a single oral administration of Ivermectin to patients with mild and moderate COVID-19 is effective in converting SARS-CoV-2 RT-PCR to negative result and in reducing viral load.Methods: In this double-blind trial, patients were randomized to elixir formulation of Ivermectin in 24 mg, 12 mg or placebo in 1:1:1 ratio. The co-primary outcomes were conversion of RT-PCR to negative result and the decline of viral load at day 5 of enrolment and were assessed in patients with positive RT-PCR at enrolment (modified intention-to-treat population). Safety outcomes included total and serious adverse events and were assessed in all patients who received the trial drug (intention-to-treat population). Results: Among 157 patients randomized, 125 patients were included in mITT analysis. Forty patients each were assigned to Ivermectin 24 mg and 12 mg, and 45 patients to placebo. The RT-PCR negativity at day 5 was higher in the two Ivermectin arms but failed to attain statistical significance (Ivermectin 24 mg, 47.5%; 12 mg, 35.0%; and placebo, 31.1%; p= 0.30). The decline of viral load at day 5 was similar in the three arms. No serious adverse events were encountered.Conclusion: In patients with mild and moderate COVID-19, a single administration of Ivermectin elixir (either 24 mg or 12 mg) demonstrated a trend towards higher proportion of RT-PCR negativity at day 5 of enrolment. The protocol was registered in the Clinical Trial Registry – India (CTRI) vide ref No CTRI/2020/06/026001.


Introduction
The COVID-19 pandemic has become one of the biggest public health challenges of the 21 st century by already having affected around 50 million people globally and causing more than a million deaths (1). Although most patients have mild or moderate illness, the high contagiousness of the causative severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) contributes to rapid spread of infection.Unfortunately, despite aggressive efforts, no antiviral agent has yet been shown to be conclusively bene cial in non-severe COVID-19.
Several new and repurposed drugs are being trialled in mild and moderate COVID-19 to help suppress viral transmission and prevent disease progression. Ivermectin is one such drug which has an established record of safety with over 2.5 billion doses dispensed over the past three decades (2). Originally introduced as an anti-helminthic agent against tropical parasitic diseases, it has recently been found to possess additional antiviral, anti-in ammatory and anti-cancer actions (2). A broad-spectrum antiviral effect against single stranded RNA viruses such as HIV-1, dengue, yellow fever, West Nile virus and others has been observed in preclinical studies (3)(4)(5). This has been attributed to a host directed action against the importinα/β protein which is used by the viral nucleocapsid to enter the host nucleus (5).
In the urgency to search for effective drugs against COVID-19, ivermectin has also been evaluated. Recently, an in-vitro study by Caly et al demonstrated that micromolar concentrations (2-2.5 µg/mL) of Ivermectin can reduce viral load by 5000-fold at 48 hours in VERO/hSLAM cells (6). Although equivalent plasma concentrations are di cult to achieve with routine antiparasitic doses of Ivermectin (150-400µg/kg), there are inherent differences in the in-vivo and in-vitro responses to drugs. Ivermectin may act through its metabolites, get concentrated three-fold in lung tissue and have additional immunomodulatory actions at routine doses (7,8). Additionally, higher doses of ivermectin (1-2 g/kg), albeit unapproved, have been shown to be well tolerated (9,10). Till date, controlled trials evaluating Ivermectin in COVID-19 are lacking. Hence this exploratory study was designed to determine the e cacy and safety of this drug in COVID-19.

Methodology
We conducted a randomized, placebo-controlled, three-arm, parallel group study of a single oral administration of Ivermectin elixir at two dose strengths (12 mg and 24 mg) in patients with non-severe COVID-19. An independent data and safety monitoring board was constituted to oversee the conduct of the trial.

PATIENTS
Consecutive patients admitted at the trial site were screened and were considered eligible for inclusion if aged 18 years or above and diagnosed with non-severe COVID-19, i.e. room air saturation (SpO2) >90%, and with no hypotension or requirement of mechanical ventilation. Diagnosis of COVID-19 was based on a positive result on either SARS-CoV-2 reverse transcription-polymerase chain reaction (RT-PCR) or the rapid antigen test. Patients were excluded if they did not give informed consent. Other exclusion criteria included: pregnancy or lactation, known hypersensitivity to ivermectin, chronic kidney disease with creatinine clearance <30 mL/min, elevated transaminase levels (>5 x upper limit of normal), myocardial infarction or heart failure within 90 days prior to enrolment, prolonged corrected QT interval (>450 ms) on electrocardiogram, any other severe comorbidity as per investigator's assessment, or enrolment in a concomitant clinical trial.

TRIAL PROCEDURES
All subjects ful lling the trial eligibility criteria underwent a detailed clinical evaluation including history and physical examination. Assessment of comorbidities including diabetes mellitus, systemic hypertension, coronary artery disease, chronic obstructive pulmonary disease, tuberculosis, and obesity was done. Baseline laboratory investigations including complete blood count, renal function tests, liver function tests, in ammatory markers (including C-reactive protein and serum ferritin) and coagulation pro le (prothrombin time, D-dimer and serum brinogen) were performed. A baseline chest radiograph was obtained and graded using the Brixia score (11). Patients were managed using standard hospital protocol by the clinical team. The patients were followed up for a minimum of 14 days or till hospital discharge, whichever was later.

INTERVENTIONS AND RANDOMIZATION
Prior to study initiation, our group performed a pharmacokinetic simulation study of the dosing requirements for achieving an Ivermectin lung concentration of 2-2.5 µg/mL (unpublished work). As ivermectin is known to concentrate 2 to 10-fold in tissues (12), it was estimated that a plasma concentration of 150-500 ng/mL would enable su cient drug concentration in the lungs. Furthermore, plasma Ivermectin concentration may rise 2 to 2.5-fold when administered orally with a high-fat diet or in an alcohol-based formulation (10). Accordingly, we found that an alcohol-based elixir formulation of Ivermectin at a dose of 400 µg/kg administered after a meal may achieve a plasma Ivermectin concentration greater than 150 ng/mL. Accurately weighed Ivermectin was used for formulating Ivermectin elixir formulation. A 20 mL dose of nal formulation consisted of ivermectin (12 or 24mg) in ethanol (40% v/v) sweetened with syrup base which was suitability avoured and coloured.
Representative samples were subjected for the quality control to ensure the drug content and batch uniformity. It was compounded and dispensed from the in-house pharmacy by a quali ed pharmacist. Similar placebo was also prepared without ivermectin and formulations were coded before delivery to the trial site. After baseline evaluation, eligible patients were randomized in a 1:1:1 ratio to receive Ivermectin 12 mg (equivalent to 200 µg/kg) elixir, Ivermectin 24 mg (equivalent to 400 µg/kg) elixir, or identical placebo. A variable block randomization strati ed based on disease severity (mildor moderate illness) was done using a centralized telephone-based system and the patients, investigators, caregivers, and statisticians were blinded to the allocation. The intervention was given two hours after breakfast on the day of randomization as a single dose.

VIROLOGICAL ASSESSMENT
All patients who underwent randomization were evaluated using a baseline oropharyngeal and nasopharyngeal swab for COVID-19 RT-PCR. The sample was collected in a standardized viral transport medium using a nylon-tipped swab. Samples were transported at 2-8 degrees Celsius and were processed within 24 hours. RNA extraction was performed using an automated extraction system (Genolution, South Korea) which is an FDA-approved magnetic bead-based extraction system. For real time RT-PCR, Thermo sher's Quantstudio™ was used. All kits used for COVID-19 assay were pre-approved by the Indian Council of Medical Research (ICMR). To determine sample adequacy and ascertain adequate extraction of RNA, an endogenous control was used for each sample as part of the assay. A reference control was run in 8 serial dilutions to make a standard curve based on cycle threshold (CT) values at each dilution. Furthermore, with each set of samples one reference with high CT value and one with lowest CT value was run, hence a semiquantitative estimate of viral load (expressed as log 10 viral copies/mL) was provided.
In patients with positive baseline RT-PCR report, follow up RT-PCR was performed on days 3, 5 and 7 following drug intervention to estimate the change in viral load.

OUTCOMES
The primary outcomes were to evaluate the e cacy of the two different doses of oral ivermectin compared with placebo in reduction of viral load and conversion to negativity of nasopharyngeal/ oropharyngeal RT-PCR on day 5 after intervention. The viral load was estimated using the cycle threshold of the RT-PCR. The secondary outcomes included qualitative and quantitative results of RT-PCR on day 3 and 7 after intervention; time to clinical resolution; frequency of clinical worsening; clinical status of the subject on day 14; and hospital-free days at day 28. The clinical status was expressed using the 8-point World Health Organization (WHO) ordinal scale (13) as follows: 1-not hospitalized, no limitation of activities; 2-not hospitalized, limitation of activities; 3-hospitalized, not requiring supplemental oxygen; 4-hospitalized, requiring supplemental oxygen; 5-hospitalized, on non-invasive ventilation or high-ow oxygen devices; 6-hospitalized, on invasive mechanical ventilation; 7-hospitalized, on vasopressors, renal replacement therapy orextracorporeal membrane oxygenation; and 8-death. The frequency of total and serious adverse events in the study groups was documented.

STATISTICAL ANALYSIS
All consenting patients who were randomized and received a study medication were included in the intention-to-treat (ITT) analysis. Among these, patients with a positive result on nasopharyngeal/oropharyngeal RT-PCR on the day of enrolment were included in the modi ed intentionto-treat (mITT) analysis. All virological outcomes were assessed in the mITT population as viral load decline and conversion of RT-PCR to negative result was unmeasurable in patients with negative RT-PCR on the day of enrolment. Clinical outcomes were assessed in the mITT population, whereas the adverse effects were evaluated in the ITT population. Statistical analysis was performed using STATA (version 14). Categorical variables were expressed as number and percentage. Continuous variables were presented as mean and standard deviation, or median and interquartile range. Inter-group comparisons of categorical outcome variables were performed using Fisher's exact test. Inter-group comparisons of continuous outcome variables were performed using analysis of variance (ANOVA) or Kruskal-Wallis test.
The comparisons of decline of log 10 viral copies/mL between different pairings of study groups at various time points were performed using t-test and were expressed as mean difference with 95% con dence intervals (CI). In the presence of a negative RT-PCR test on follow-up sample, the viral load was imputed to 0 on the log scale. A p-value of less than 0.05 was considered statistically signi cant.
The funder had no role in study design, data collection, data analysis or writing of the report. The corresponding author had full access to the study data and had the nal responsibility for the decision to submit for publication.

Results
Between 28 July, 2020 and 29 September, 2020, a total of 278 patients with mild or moderate COVID-19 were assessed for eligibility( Figure 1). Of these, 157 patients were randomized, of whom 5 patients subsequently withdrew consent. The ITT population (n = 152) included 51 patients assigned to ivermectin 24 mg, 49 patients assigned to ivermectin 12 mg, and 52 patients assigned to placebo. Among these, 125 patients had a positive nasopharyngeal/oropharyngeal SARS-CoV-2 RT-PCR result on day of enrolment and were included in the mITT analysis. The mITT population included 40 patients in ivermectin 24 mg arm, 40 patients in ivermectin 12 mg arm, and 45 patients in the placebo arm. In the mITTgroup, 80 patients (64%) had mild illness, while 45 patients (36%) had moderate illness.
The mean (SD) age of participants was 35.3 (10.4) years and majority (88.8%) were males. The proportion of patients with moderate illness was 40% in ivermectin 24 mg arm, 32.5% in ivermectin 12 mg arm, and 35.6% in placebo arm (Table 1). In contrast, the proportion of asymptomatic patients at enrolment was 22.5% in ivermectin 24 mg arm, 27.5% in ivermectin 12 mg arm, and 17.7% in placebo. Baseline clinical severity by WHO ordinal scale was 3 in the majority (92%) of patients. The median duration of symptoms at the time of enrolment was 5 days (interquartile range, 3 to 7 days) and was similar in the three arms. There were no signi cant differences in the comorbidities or presenting symptoms in the three arms. Baseline laboratory parameters in the three arms were similar (Supplementary Table 1). A minority (10%) of patients received concurrent antiviral therapies including remdesivir, favipiravir or hydroxy-chloroquine as decided by site physicians without any difference in the three arms (Supplementary Table 2) Primary outcomes The proportion of subjects who became RT-PCR negative on day 5 of enrolment was numerically higher with ivermectin 24 mg arm (47.5%) compared with ivermectin 12 mg arm (35.0%) and placebo arm (31.1%)(Table2); however, this difference did not attain statistical signi cance (p-value =0.30) (Figure 2). Subgroup analysis based upon disease severity also demonstrated no signi cant difference in the negativity of RT-PCR at day 5. In subjects who received intervention early in the course of illness (within 4 days of symptom onset), Ivermectin 24 mg arm had numerically higher negativity of RT-PCR at day 5 compared with placebo (47.0% vs 28.6%, p-value = 0.38).The viral load at enrolment did not impact the e cacy of the therapies to convert to negative RT-PCR at day 5.
There was no signi cant difference in the viral load (expressed as log 10 viral copies/mL) in the three arms, either at baseline or at day 5 of enrolment (Table 3), or in the decline of viral load between the ivermectin and placebo arms at day 5 (Table 4& Figure 3). Furthermore, no difference was observed in the absolute viral load or the decline of viral load in either the mild or the moderate illness strata at day 5 (Supplementary tables 3, 4, 5 and 6).

Secondary outcomes
Among the secondary virological outcomes, there was no signi cant difference in the three arms in terms of conversion to negative RT-PCR (table 2), or in the decline of viral load at either day 3 or day 7 of enrolment (Table 4).
Secondary clinical outcomes were also similar in the three arms (Table 5). There was no difference in the mean (SD) duration of symptom resolution in the three groups or in the duration of hospital-free days at day 28. The proportion of patients with clinical worsening (de ned as an increase in the WHO ordinal score during treatment) was similar in the three groups (ivermectin 24 mg, 7.5%; ivermectin 12 mg, 5.0%; and placebo, 11.1%; p-value = 0.65).

Adverse events
There were no serious adverse events reported during the study ( Table 6). The frequency of all adverse events in the ITT population was similar in the three arms (ivermectin 24 mg, 11.8%; ivermectin 12 mg, 16.3%; and placebo, 11.5%; p-value = 0.76). The most frequent adverse event was epigastric burning sensation, which occurred in 17 (11.2%) patients.

Discussion
In this investigator-initiated, triple-blind, randomized, placebo-controlled trial, we examined the e cacy and safety of Ivermectin at two doses (24 mg and 12 mg) in the management of non-severe COVID-19. Patients in the Ivermectin 24 mg arm demonstrated a numerically higher rate of conversion to negative RT-PCR at day 5 compared to the placebo arm overall and also separately in the mild and moderate subgroups; however, this did not reach statistical signi cance ( Figure 2). Further, the decline in viral load at day 5 in all groups was similar.
The interest in Ivermectin in the treatment of COVID-19 was sparked by an in-vitro study by Caly et al, wherein they had demonstrated in Vero/hSLAM cells, that a single application of Ivermectin to achieve concentrations of 2-2.5 µg/mL enable a 5000-fold reduction in the viral load within 48 hours(6). Ivermectin has a plausible broad spectrum anti-viral action by inhibiting the importin α/β protein of the host(3). The inhibition of this protein blocks the entry of the viral nucleocapsid into host nucleus for subsequent replication. Previously, in a phase III clinical trial, Ivermectin increased rate of viral clearance of dengue virus compared with placebo without any demonstrable clinical bene t (14).
However, the micromolar doses described in the in-vitro study by Caly et al are di cult to achieve in vivo with the FDA-approved dose (200 µg/kg) of Ivermectin (15). Although Ivermectin is usually administered in tablet form, its bioavailability may increase upto 2.5-fold when given alongwith a fat-rich meal or in an alcohol-based formulation (10,16). Furthermore, Ivermectin may preferentially distribute into the tissues, including the lung (12). Hence, we included a higher dose (400 µg/kg) of Ivermectin in an alcohol-based elixir given after breakfast. Nonetheless, even higher doses may be required to achieve optimal therapeutic doses against SARS-CoV-2. Indeed, doses up to 1-2 g/kg have been found to be safe and may be explored further (9,10). Furthermore, Ivermectin may have immunomodulatory actions at nanomolar doses by inhibiting the nicotinic acetylcholine receptor (nAChR). The nAChR may act as a receptor for SARS-CoV-2 and drive dysregulated cytokine release (IL1, IL6, TNF and IL18) from macrophages (17,18).
In our study subjects, Ivermectin did not improve the time to symptom recovery, clinical status at day 14, or hospital-free days at day 28 after drug administration. Similar results were observed in the only other randomized-trial of Ivermectin (12 µg/kg) in predominantly mild COVID-19 patients (n=62) in Bangladesh, wherein Podder et al (19) found that Ivermectin failed to hasten the resolution of symptoms compared to usual care. The same investigators repeated RT-PCR only once on day 10 and found that most patients had attained a negative result (19). In contrast, we performed RT-PCR at days 3, 5 and 7 to serially evaluate decline in viral load with Ivermectin. Our rationale was that faster viral load decline may enable the non-severe COVID-19 patient to become non-infectious sooner, thereby limiting the contagion. Indeed, it has been shown that at a lower viral load (CT > 24), infectivity declines with lower viral culture positivity (20). Hence the trend towards increased viral negativity at day 5 with ivermectin 24 mg in our trial, particularly among mildly ill patients, encourages further exploration in this regard.
In a retrospective study of hospitalized patients in Florida (21), patients who received Ivermectin were found to have a signi cantly lower mortality that those who did not (15% versus 25%). The mortality bene t remained signi cant after propensity-matched analysis and adjusting for confounders. However, they included patients with greater illness severity than our study population, illustrated by lack of mortality in our trial. Furthermore, the greater use of concurrent therapies and retrospective design preclude drawing de nitive conclusions from their data. Nonetheless, we did nd a 56.2-61.5% RT-PCR negativity among moderately ill patients who received Ivermectin at day 5 of enrolment. The immunomodulatory rather than antiviral effect of Ivermectin may be hypothetically more important in moderate and severe COVID-19 (22).
There were no serious adverse events in our trial. Since we have used a novel elixir-based formulation with an aim to maximize plasma bioavailability of Ivermectin, this reassures us regarding its safety for further study. The frequency of mild adverse events was similar with ivermectin at either dose or placebo. Other studies of Ivermectin in COVID-19 have also found a low rate of adverse events (19,23). The predominant adverse event in our study was transient burning sensation in the epigastrium which could be attributed to the alcohol-based elixir preparation.
The major limitation of our study was that it was conducted at a single centre with a relatively small sample size. Most of our patient population was male and relatively young (mean age, 35.3 years) with few comorbidities which re ects the demographics of the catchment area of our centre. Such a patient population is likely to have an uncomplicated disease course (24,25). Furthermore, in the absence of previous clinical trials and considering the urgency of the research question, our sample size was exploratory. Hence, we cannot exclude the possibility that a similarly conducted study in a larger and more diverse population could have uncovered clinical e cacy of Ivermectin, if such bene t indeed exists. Furthermore, the favourable safety pro le is encouraging for the conduct of larger trials to further clarify the role of ivermectin in COVID-19.
Secondly, the elixir formulation of ivermectin used by us is not yet commercially available. Although our Ivermectin formulation and dosing strategy was determined by a simulation study to attain an adequate drug concentration in the lung, further pharmacokinetic studies are necessary to de ne the optimal therapeutic dosing of Ivermectin in COVID-19. Furthermore, Ivermectin has a plasma half-life of 18 hours and does not accumulate on repeat dosing (10). Whether multiple doses of Ivermectin in this disease may be superior to a single dose strategy is currently unknown. Hence, the translation of our ndings to the use of Ivermectin tablet at various dosing strengths and frequencies in clinical practice requires caution.
Finally, in our study we have recruited patients irrespective of the duration of illness prior to enrolment.
The median duration of symptoms at randomization was 5 days in the three arms. Hence, a signi cant number of patients had a negative RT-PCR result at baseline and were excluded from the modi ed intention-to-treat analysis. The recruitment of mild patients at a later stage of illness could also have contributed to high rates of RT-PCR negativity in the placebo arm at day 5 of enrolment (31.1%). In a previous study, over 90% of patients with mild COVID-19 have been found to have negative RT-PCR at day 10 of onset of illness(26). Furthermore, antiviral bene ts of Ivermectin are postulated to be maximal early in disease course, while hypothetical immunomodulatory bene ts may occur later in the illness. Hence, a better understanding of the cellular actions of ivermectin is necessary to de ne target populations precisely for future trials.

Conclusion
In conclusion, in this exploratory randomized placebo-controlled trial of a single oral administration of Ivermectin elixir at two different dosage strengths (12 mg and 24 mg) in patients with mild and moderate COVID-19, a trend towards higher negativity of RT-PCR at day 5 was observed with the use of Ivermectin 24 mg, while the decline in viral load was similar in all three arms. Reassuringly, there were no safety concerns with the use of Ivermectin at either dose. Larger studies employing different dosing regimens of Ivermectin are required to further elucidate its potential role in treatment of COVID-19.

Declarations
Authorship Statement: This is to state that all co-authors have contributed substantially to the conduct of this trial.

Disclosure Statement:
This is to state that none of the co-authors have any potential con icts of interest to declare pertaining to this manuscript or work.

Declaration of interests
We declare no competing interests.  RT-PCR -reverse transcriptase-polymerase chain reaction, CT -cycle threshold