Significant effort by scientists to identify drugs with antiviral activity against the SARS-CoV-2 virus has led to investigations involving new drugs or FDA-approved drugs for other indications, states the U.S. Food and Drug Administration (FDA). Identification of these potentially effective candidates is occurring at an unprecedented speed, and a necessary step in this process is translating in vitro antiviral activity to appropriate clinical dosing regimens. This step is complex and multifactorial, but essential to finding the right products, administered at the right doses, to the right patients, at the right time for successful treatment of COVID-19.
Many compounds have been evaluated in vitro for their potential antiviral activity against SARS-CoV-2, the virus that causes COVID-19. The most commonly reported parameter to quantify the in vitro antiviral potency of these compounds is the EC50, the drug concentration in cell culture media that provides 50% maximal antiviral activity. Many experimental factors can influence the estimation of antiviral potency (i.e., the EC50), such as the SARS-CoV-2 viral strain and quantity added in culture, cell line, drug treatment duration, percent of serum present in the culture medium, and methods used to quantify virus.
One common practice is to estimate the EC50 based on initial drug concentrations added to the cell culture media (extracellular concentration), even though antiviral activity might be related to intracellular concentrations of the drug or drug metabolites. EC50 value alone is not sufficient to judge a drug’s in vivo antiviral activity because the maximum in vitro antiviral activity could provide far less than 100% inhibition, despite a very low EC50 value, and drug concentrations necessary to achieve meaningful antiviral activity in vivo may not be safely tolerated.
For drugs previously approved by the FDA, the in vivo drug concentrations (i.e., exposure) associated with the approved dosing regimen are generally known, as are other important pharmacokinetic data that can be used to help evaluate the potential utility of a drug to treat a new indication such as COVID-19.
For example, drug exposure determinations observed under the previously approved regimen, or PK model-based predictions of drug exposure under a new dosing regimen, can be used as a reasonable estimate of in vivo drug exposure in healthy subjects (for prophylaxis) or COVID-19 patients (for treatment).
Importantly, drug exposure can be measured in different matrices (e.g., tissue, whole blood, serum, or plasma) and, depending on the physicochemical properties of the drug, may vary substantially from one matrix to the next. For example, chloroquine phosphate (CQ) and hydroxychloroquine sulfate (HCQ) are known to reach significantly higher concentrations in certain tissues (including blood) than in plasma, with a serum-to-plasma concentration ratio estimated to be 2 for both CQ and HCQ, and specific blood processing procedures should be followed in order to accurately measure their plasma concentrations. PK models used to predict CQ/HCQ plasma concentration should similarly be based on properly processed plasma samples to yield reliable conclusions on the adequacy of dosing regimens.
n vivo antiviral efficacy requires sufficiently high in vivo drug exposure. Typically, the most relevant in vivo drug exposure should be that at the site of action. However, since in vitro antiviral potency experiments only measure the drug concentration in the cell culture media (i.e., extracellular drug concentration), the most relevant in vivo exposure is the free/unbound drug concentration in the tissue interstitial fluid. Since the free drug concentration in the tissue interstitial fluid is generally similar to the free plasma concentration under equilibrium, we can link existing data regarding a drug’s in vitro concentration to its in vivo tissue interstitial fluid concentration.
Consequently, even if we are examining drugs that reach significantly higher concentrations in tissues than in plasma, in the absence of actual intracellular concentration data, the free drug concentrations in plasma – which are likely to be comparable with the free extracellular tissue concentrations – are nevertheless the most relevant exposure values that can be compared with the reported EC50 values derived from drug concentrations in cell culture media.
As noted above, this in-vitro-to-in-vivo extrapolation assumes similar in vivo cellular drug accumulation as those seen in in vitro experiments. In situations involving an active metabolite, then, the intracellular conversion efficiency of the parent drug to the active metabolite in vivo is assumed to be similar to the efficiency in vitro.
Much is unknown about the pathogenesis of SARS-CoV-2-induced COVID-19, as well as the relevant mechanism of action for any drugs that may prove to be safe and effective for COVID-19 prophylaxis and treatment. Adequate and well-controlled clinical trials will ultimately determine appropriate treatment modalities for COVID-19 and will identify the stages of infection and disease at which these treatments are effective, with which administration routes, and at what dosing regimens. Sound PK modeling and accurate in-vitro-to-in-vivo translation of data will be crucial for the appropriate design of dosing regimens administered in these trials.