Need for a Standardized Translational Drug Development Platform: Lessons Learned from the Repurposing of Drugs for COVID-19

Abstract

In the absence of drugs to treat or prevent COVID-19, drug repurposing can be a valuable strategy. Despite a substantial number of clinical trials, drug repurposing did not deliver on its promise. While success was observed with some repurposed drugs (e.g., remdesivir, dexamethasone, tocilizumab, baricitinib), others failed to show clinical efficacy. One reason is the lack of clear translational processes based on adequate preclinical profiling before clinical evaluation. Combined with limitations of existing in vitro and in vivo models, there is a need for a systematic approach to urgent antiviral drug development in the context of a global pandemic. We implemented a methodology to test repurposed and experimental drugs to generate robust preclinical evidence for further clinical development. This translational drug development platform comprises in vitro, ex vivo, and in vivo models of SARS-CoV-2, along with pharmacokinetic modeling and simulation approaches to evaluate exposure levels in plasma and target organs. Here, we provide examples of identified repurposed antiviral drugs tested within our multidisciplinary collaboration to highlight lessons learned in urgent antiviral drug development during the COVID-19 pandemic. Our data confirm the importance of assessing in vitro and in vivo potency in multiple assays to boost the translatability of pre-clinical data. The value of pharmacokinetic modeling and simulations for compound prioritization is also discussed. We advocate the need for a standardized translational drug development platform for mild-to-moderate COVID-19 to generate preclinical evidence in support of clinical trials. We propose clear prerequisites for progression of drug candidates for repurposing into clinical trials. Further research is needed to gain a deeper understanding of the scope and limitations of the presented translational drug development platform.

Keywords: COVID-19; clinical trials; drug repurposing; pandemics; translational medicine.

Figure 1

Ideal process for generating new preclinical data of repurposed drugs against SARS-CoV-2 to build a rationale for a clinical evaluation.

Figure 2

Simulated plasma concentration–time profiles for the molnupiravir metabolite EIDD-1931 in humans (70 kg) and hamsters (120 g), based on the final population pharmacokinetic models (for details, see Supplementary Materials, Section S2.6, Tables S8, S23 and S24). The black line (bounded by a grey shade showing the 90% confidence interval) represents the median simulated concentration-vs.-time profile in humans at a clinically relevant dose (800 mg molnupiravir, twice daily, for 10 days). The cyan line represents the median profile in hamsters, following administration of 50 mg/kg molnupiravir, twice daily for 10 days, corresponding to the approximate dose in hamsters matching C_max in humans. The dark blue line represents the median profile in hamsters following administration of 150 mg/kg molnupiravir, twice daily for 10 days, corresponding to the approximate dose in hamsters matching AUC_total in humans; 150 mg/kg twice daily was also the efficacious dose in the hamster infection model of SARS-CoV-2. The horizontal red line denotes the IC₅₀ (A549-ACE2TMPRSS2 cells), corrected for plasma protein binding in hamsters and humans (Supplementary Materials, Section S2.4, Table S7).

Figure 3

Simulated pharmacokinetic parameters of molnupiravir in humans and hamsters: (A) C_max; (B) AUC_total. Predicted pharmacokinetic parameters, i.e., C_max and AUC_total, in humans receiving 800 mg of molnupiravir twice daily for 10 days (grey box) were compared with pharmacokinetic parameters of hamsters receiving 20 to 200 mg/kg of molnupiravir twice daily (blue boxes) for same number of days. Boxes and whiskers represent the median with inter-quantile range and the 95% prediction intervals, respectively. The blue vertical line denotes the IC₅₀ (A549-ACE2TMPRSS2 cells) corrected for plasma protein binding in hamsters and humans.