Predicting the dose for repurposing nitazoxanide in COVID-19

By Dr. Liji Thomas, MDMay 7 2020

A new study published on the preprint server medRxiv* in May 2020 reports a model that could help evolve appropriate dosage protocols for future clinical trials of the repurposed drug nitazoxanide in COVID-19 prevention and treatment. The researchers say, “It was possible to achieve plasma and lung tizoxanide concentrations, using proven safe doses of nitazoxanide, that exceed the EC90 for SARS-CoV-2.”

The need to achieve effective drug concentration

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes COVID-19 disease, a respiratory illness with significantly high mortality, which has spread rapidly and extensively to become a pandemic. While treatment and prevention measures are being sought with the utmost urgency, none have yet been identified.

Novel Coronavirus SARS-CoV-2 Colorized scanning electron micrograph of a VERO E6 cell (blue) heavily infected with SARS-COV-2 virus particles (orange), isolated from a patient sample. Image captured and color-enhanced at the NIAID Integrated Research Facility (IRF) in Fort Detrick, Maryland. Credit: NIAID

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources

A promising route has been to repurpose existing drugs developed and used for other conditions since they have already been approved and have passed safety and tolerability tests. However, one bottleneck has been the determination of the drug concentrations in the blood and the lung tissue at a given dose, though this is obviously an important step.

Even though SARS-CoV-2 targets the ACE2 receptor present in high concentration on some lung cells, which therefore serve as the primary site of viral replication, the virus infects the upper respiratory tract initially. Still, the presence of these receptors on the kidneys, nervous tissue, and liver cells causes these organs to be affected as well later in the course of the disease.

For this reason, any effective antiviral must achieve effective or therapeutic levels not only in the upper airway, to treat and to prevent infection but also in the blood to eliminate the virus in these other tissues and organs as well. However, a recently developed analytic method that benchmarked the in vitro activity of a drug against the concentrations in target tissue achieved with the clinical doses showed that most of the repurposed drugs could not reach useful concentrations at approved doses.

Nitazoxanide

One such drug is nitazoxanide, an antiparasitic drug used to treat cryptosporidiosis and giardiasis, both parasites which cause diarrhea, as well as some anerobic bacteria, viruses, and other protozoa. It is rapidly metabolized to the metabolite tizoxanide, which has not been tested for its activity against SARS-CoV-2 as yet.

However, it does show in vitro activity, like nitazoxanide, against the influenza virus and other coronaviruses, as well as rotaviruses, and the hepatitis B and C viruses. It may reduce the symptoms and shedding of SARS-CoV-2 as well as enhancing the innate immune response, increasing interferon type 1 production. It also blocks TMEM16A ion channels, and so causes bronchodilation.

It is known to be relatively safe, with minimal adverse gastrointestinal effects when given as a single oral dose at up to 4 g. it is recommended to be taken after meals as this increases the plasma levels.

An ongoing clinical trial is expected to provide data on the activity of nitazoxanide at a dose of 500 mg BD, either alone or along with hydroxychloroquine, against SARS-CoV-2. However, there is no publicly available supporting evidence for this dose against the virus.

How was the current PBPK modeling done?

The current study aims to identify a nitazoxanide dose and schedule that will achieve plasma and lung trough concentrations above the minimum effective concentration reported for the SARS-CoV-2 in vitro, in over 90% of patients. The tool used is physiologically-based pharmacokinetic (PBPK) modeling.

PBPK modeling is a computational tool that allows the selection of a suitable dose and dosing schedule of a drug based on the physiology of the human organism and the manner in which the drug is typically metabolized in the body.

As a first step, a PBPK model for tizoxanide concentrations following nitazoxanide was validated. This was then used to evaluate the plasma and lung concentrations of tizoxanide following the administration of 600 m BD nitazoxanide in a previous trial of the drug for influenza. This allowed the exposure to be measured with respect to the previously demonstrated in vitro activity in a situation where the drug is already known to have clinical benefit.

Then the researchers tested different doses and schedules of nitazoxanide to find those that would keep the plasma and lung levels above the reported EC90 for most patients, at 4.64 199 μM (1.43 mg/L).

A virtual patient set of 100 healthy adults, equally composed of both sexes, aged 20-60 years, was created using patient demographics from CDC charts. Existing literature was used to arrive at appropriate parameters for physiological measures like blood flow rate and organ weights and volumes.

Using a seven-compartment model to describe the drug movement through the stomach and small intestine, the absorption, and distribution of the drug was described mathematically. The ratio of plasma drug levels to tissue drug levels was assessed the same way.

The model was then validated using clinical data acquired from healthy people for a range of oral doses from 500 mg to 400 mg, and for multiple doses at 500 mg BD to 1000 mg BD with food. Assuming apparent permeability data to calculate the absorption of the prodrug, and that it would be converted to the active form tizoxanide as soon as it reaches the blood (it is known to be wholly converted in 6 minutes in plasma), the disposition of the drug was finally defined.

The researchers assumed the model would have been validated if the error in plasma concentrations in the observed and simulated plasma concentrations was less than two-fold, and if the maximum concentration, an area under the curve for plasma concentration vs time, and trough concentration at the end of the interval between doses, were all less than twice the mean observed values.

Were the PBPK results realistic?

The results showed the PK values for all these parameters were between 1 and 1.55 for the average error in the fasted and fed states. Other parameters ranged from 0.8 to 2.2 times the observed measurements in the fasted and fed states.

Simulation of the dosing schedule used in the previous clinical trial in influenza patients showed that all patients had plasma and lung drug levels below the EC90. Still, the maximum levels of the drug in the plasma and the lung reached the EC90 or above in 71% and 14%, respectively.

Finally, the optimal doses to keep the trough concentrations above the calculated 1.43 mg/L for at least 50% of the simulated data set were predicted to be 1.200 mg QID, 1600 mg 243 TID, 2900 mg BID in the fasted state. In the fed state, these were 700 mg QID, 900 mg TID, and 1400 mg BID. Steady-state concentrations were achieved within 48 hours in both states.

The previous influenza trial showed that nitazoxanide brought down the median duration of symptoms from 117 hours to 96 hours at 600 mg BD, and lowered the viral load. However, the PBPK model showed that this dosage schedule actually underdosed most of the patients, with the trough drug values falling below the reported EC90 of 8.4 mg/L. This might account for the moderate effects.

How is the study useful in SARS-CoV-2?

Simulating the effect of the drug at various doses and dosing intervals against SARS-CoV-2, the study indicates the potential for use with or without food. Smaller doses at three or four times daily achieve good plasma exposure when taken with food, and this may be preferred in sick patients to reduce the chances of gastrointestinal side effects.

However, single doses are not found to reduce the plasma trough levels to below the EC90, and this may be essential for long-term preventive administration, provided gastrointestinal side effects are tolerable.

The drug may inhibit the glycosylation of the spike protein, similar to its action on the hemagglutinin antigen of the influenza virus.

Limitations of the study include the unavailability of data on certain PK parameters. The amplifying indirect drug effects achieved by its stimulation of the innate immune system would not be reflected in the in vitro study on which this simulation was based. The model included only healthy participants, and PK data on drug disposition in underlying kidney and liver disease is not available for this drug. There is no clinical data for multiple doses over 1000 mg, which limits the accuracy of prediction for higher multiple doses. The limited data on many fronts, including lack of PK data on tizoxanide administered with food rather than in the fasted state, and corrected in vitro activity to account for the over 99% protein binding of tizoxanide in plasma,  is the chief hindrance to the acceptance of the accuracy of the predictions of this simulation.

The implications

The PBPK model presents data showing the feasibility of dosing with nitazoxanide to achieve effective plasma and lung concentrations. This allows for better design for future clinical trials. It also shows, for multiple dosing regimens, an optimal sparse sampling strategy of 0.25, 1, 3, and 12h post-dose for use in treatment trials.

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources