2. Viral Polymerase Inhibitors

2.1. Remdesivir (Veklury, GS-5734)
Remdesivir is an adenosine monophosphate derivative and nucleotide-based antiviral prodrug (Figure 2). Remdesivir received, in May 2020, an emergency use authorization from the U.S. FDA for the treatment of laboratory-confirmed or suspected COVID-19 illness in children and adults hospitalized with severe disease [25]. The parenterally administered drug is being developed by Gilead Sciences, U.S., and has broad-spectrum antiviral activity [26]. It was first studied in 2016 as a potential treatment for Ebola virus [27]. In addition to its activity against SARS-CoV-2, remdesivir has a potential to treat a variety of infections caused by RNA viruses, including SARS-CoV and MERS-CoV [28].
The drug is metabolized to the pharmacologically active nucleoside triphosphate metabolite after being distributed into cells (Figure 2). The triphosphate metabolite acts as a competitive inhibitor of RdRp and thus eventually causes chain elongation termination, which decreases the viral RNA replication [29]. The termination is delayed and happens after the addition of more nucleotides (between 3 and 5). Therefore, remdesivir is described as a direct antiviral agent acting as a delayed chain terminator [30,31]. Importantly, remdesivir avoids proofreading by viral exoribonuclease [28,32]. Currently, remdesivir is being evaluated as a treatment for COVID-19 patients in about 15 studies across the globe. The drug is being tested alone or in combination with merimepodib (NCT04410354; n = 40), tocilizumab (NCT04409262; REMDACTA; n = 450), or baricitinib (NCT04401579; ACTT2; n = 1034). In particular, merimepodib is another antiviral agent that is inhibitor of inosine monophosphate dehydrogenase. The enzyme is required for the synthesis of guanine nucleotides. Merimepodib consequently inhibits the synthesis of DNA and RNA, leading to antiviral and immunosuppressive effects. Thus, remdesivir and merimepodib is a dual-acting antiviral combination with immunosuppressive activity.
Remdesivir itself demonstrated in vitro activity against Vero E6 cells infected with SARS-CoV-2 with an EC50 value of 0.77 µM (CC50 > 100 µM) [33]. Remdesivir also exhibited in vitro activity against SARS-CoV and MERS-CoV in multiple in vitro systems, including primary human airway epithelial cell cultures with sub-micromolar IC50 values [28]. Remdesivir was also effective against pre-pandemic bat-CoVs, bat-CoVs, and contemporary circulating human coronaviruses in primary human lung cells suggesting a broad-spectrum anti-coronavirus activity. In a mouse model of SARS-CoV, the prophylactic and early therapeutic use of remdesivir significantly decreased the lung viral load and improved the respiratory functions as well as the overall clinical signs of the disease [28]. Furthermore, remdesivir with interferon (INF)-b demonstrated better antiviral activity compared to lopinavir/ritonavir with INF-b in vitro. Compared to lopinavir/ritonavir/INF-b, the prophylactic and therapeutic use of remdesivir also more effectively diminished the pulmonary viral loads and improved the pulmonary function in mice model of MERS-CoV [34]. The efficacy of the prophylactic and therapeutic use of remdesivir was also demonstrated in the rhesus macaque model of MERS-CoV infection [35]. Very recently, remdesivir was also shown to inhibit SARS-CoV-2 replication in human lung cells and primary human airway epithelial cultures (EC50 = 0.01 μM). In mice infected with a chimeric SARS-CoV encoding RdRp, therapeutic administration of remdesivir diminished lung viral load and improved pulmonary function compared with vehicle-treated mice [36].
As far as clinical trials in humans, a randomized, placebo-controlled, double-blind trial in hospitalized adults (n = 236) with severe COVID-19 in China initially revealed that the median time to improvement was not substantially different in the remdesivir group (200 mg on the first day, and then 100 mg/day for 9 days) from that of the placebo group. The mortality rate was also similar in the two groups [37]. Yet, the trial was criticized for being insufficiently powered. Later, a phase 3 randomized, open-label trial in adults (n = 397) hospitalized with severe COVID-19 sponsored by Gilead revealed that the time to clinical improvement for 50% of patients was 10 days in the 5-day treatment group relative to 11 days in the 10-day treatment group. The dose regimen used was 200 mg on day 1, followed by 100 mg/day for total of 5 or 10 days. At day 14, about 64.5% of the patients in the 5-day group and 53.8% of the patients in the 10-day group achieved clinical recovery. Patients treated with remdesivir within 10 days of symptoms onset achieved better outcomes relative to those treated after more than 10 days of symptoms [38]. Similar results were obtained in hospitalized adults (n = 1600) with moderate COVID-19 (NCT04292730). In an uncontrolled study of hospitalized COVID-19 patients (n = 61), most patients needed less oxygen support after receiving remdesivir [39]. Importantly, a phase 3 adaptive, randomized, placebo-controlled study sponsored by the U.S. National Institute of Allergy and Infectious Diseases (NIAID) in hospitalized adults (n = 1063) indicated that: (a) the patients in the remdesivir group had shorter median time to recovery (11 days) than the patients in the placebo group (15 days) and (b) remdesivir may decrease the mortality rate from 11.6% in the placebo group to 8% in the treatment group [40]. As of now, the COVID-19 Treatment Guidelines Panel of the U.S. National Institute of Health recommends remdesivir for the treatment of COVID-19 in hospitalized patients with severe disease (requiring supplemental oxygen or on mechanical ventilation or extracorporeal membrane oxygenation). The Panel also indicates that there are no sufficient data to recommend either for or against the use of remdesivir in patients with mild or moderate COVID-19 [41]. Of note, the U.S. FDA warns against the concomitant use of remdesivir and chloroquine or hydroxychloroquine owing to in vitro evidence which suggests that chloroquine blocks the intracellular activation of remdesivir [42]. Moreover, data from the manufacturer’s compassionate use program suggested no safety concerns were identified for remdesivir in pediatric, pregnant, or postpartum patients [43].

2.2. Galidesivir (Immucillin-A, BCX4430)
Galidesivir is an adenosine nucleoside analog (Figure 3) that is an active site inhibitor of RdRp (EC50 < 50 µM). Similar to remdesivir, it is a prodrug that is metabolized by cellular kinases to the corresponding active form of nucleoside triphosphate. The triphosphate form binds to the active site of the viral enzyme and gets incorporated into the growing viral RNA chain resulting in premature chain termination. The drug is being developed by BioCryst, U.S., and being tested in a phase 1 clinical trial for COVID-19 or Yellow Fever in Brazil in collaboration with the U.S. NIAID (NCT03891420; n = 132) [44,45,46,47].
The drug is used parenterally and has demonstrated a broad-spectrum, showing in vitro antiviral activity against at least 20 RNA viruses across eight different virus families including coronaviruses. In animal studies, the drug was effective in protecting against dangerous viruses such as Zika, Yellow Fever, Marburg, and Ebola viruses [44,45,46,47].

2.3. Ribavirin (Virazole)
It is an open-ring analog of guanosine nucleoside (Figure 3) that was approved by the U.S. FDA in 1985 for the treatment of respiratory syncytial virus [48]. It is also used systemically for chronic hepatitis C virus (HCV) infection [49] and viral hemorrhagic fever [50]. The drug possesses broad-spectrum antiviral activity against both RNA and DNA viruses. To exert its antiviral activity, the drug is to be activated by phosphorylation to generate the triphosphate nucleotide that acts as an inhibitor of RNA synthesis and viral mRNA capping [51]. Other mechanisms have also been proposed to account for its broad spectrum of antiviral activity. Inhibition of host inosine monophosphate dehydrogenase by ribavirin-monophosphate and the resulting depletion of guanosine triphosphate (GTP) pool has been put forward to be another mechanism of action. Decreased intracellular GTP pool decreases viral protein synthesis and limits the replication of viral genome. Ribavirin is also a mutagen that leads to defective virions [52] and it has immunomodulatory actions [53]. Yet, the drug has a U.S. boxed warning pertaining to the risk of hemolytic anemia and potential complications during pregnancy [54].
Ribavirin is currently being evaluated in few trials for the treatment of COVID-19 patients. It is being tested alone (NCT04356677; n = 50) or in combination with nitazoxanide and ivermectin (NCT04392427; n = 100) or with lopinavir/ritonavir and INF β-1b (NCT04276688; n = 127). Recent computational work has shown that ribavirin binds with high affinity to RdRp of SARS-CoV-2 [55]. Furthermore, the MERS-CoV rhesus macaque model revealed promising results for ribavirin and IFN-α 2b [56]. Nevertheless, mixed results came out of treating MERS-CoV infections with a combination of ribavirin and IFNs (IFN-β1 or IFN-α 2a) [57]. Results from the in vitro testing of ribavirin in Vero E6 cells also indicated that the replication and/or the cellular spread of SARS-CoV was not inhibited at concentrations known to inhibit other sensitive viruses [58]. Interestingly, a recent open-label randomized controlled trial (NCT04276688; n = 127) indicated that early triple antiviral therapy of INF-β 1b, ribavirin, and lopinavir/ritonavir was safe and superior to lopinavir/ritonavir alone in alleviating the symptoms and shortening the duration of viral shedding and hospitalization in COVID-19 patients with mild to moderate symptoms [59].

2.4. Clevudine (Levovir and Revovir)
It a thymidine nucleoside analog (Figure 3) that was approved in Korea for the treatment of hepatitis B virus (HBV) infection [60]. Similar to previous agents, it is a prodrug that requires phosphorylation to form the corresponding active nucleotide, the triphosphate. Mechanistically, the triphosphate active form appears to noncompetitively inhibit the HBV reverse transcriptase protein priming and DNA synthesis [61]. Importantly, although clevudine showed a potent antiviral response, its long-term use for more than a year led to the development of viral resistance and myopathy [60]. The drug is being evaluated in a phase 2 as a treatment for COVID-19 in Korea (NCT04347915; n = 60).

2.5. Emtricitabine (Emtriva) in Combination with Tenofovir Disoproxil or Tenofovir Alafenamide
Emtricitabine is a cytosine nucleoside analog (Figure 3) that is a competitive inhibitor of human immunodeficiency virus-1 (HIV-1) reverse transcriptase. It is metabolized by cellular kinases-mediated phosphorylation to the triphosphate form. Emtricitabine triphosphate is the active form that blocks the HIV replication by terminating its genetic chain elongation, and thus, it prevents the generation of complementary DNA from the viral RNA and reduces the viral load. The drug was first approved by the U.S. FDA as an orally bioavailable, once-daily antiretroviral drug in 2003. It is now used in combination with other antiretroviral drugs for the treatment of HIV-1 infection [62,63]. Combinations with tenofovir disoproxil fumarate (Truvada), tenofovir alafenamide (Descovy; 2016), rilpivirine, and tenofovir alafenamide (Odefsey; 2016), or bictegravir and tenofovir alafenamide (Biktarvy; 2018) are available.
In particular, tenofovir disoproxil (Viread; 2001) is an adenine-based acyclic nucleotide analog (Figure S1) that, following activation, acts as a competitive inhibitor of reverse transcriptase, and subsequently, it leads to DNA chain elongation termination. Activation of the drug starts with the hydrolysis of the external esters followed by spontaneous release of carbon dioxide and formaldehyde to form the corresponding tenofovir, a nucleoside monophosphate, which subsequently undergoes two phosphorylation steps to form tenofovir diphosphate, the active drug (Figure S1) [64]. It was first approved in 2001 by the U.S. FDA and is prescribed for the oral treatment of HIV-1 and chronic HBV infections [65]. It is also available in many other combinations with emtricitabine, lamivudine (Cimduo; 2018), doravirine and lamivudine (Delstrigo; 2018), and efavirenz and lamivudine (Symfi; 2018). The efficacy of emtricitabine and tenofovir disoproxil as a prophylactic combination against SARS-CoV-2 infection is being evaluated in a large randomized, double-blind, controlled with placebo clinical trial for health care providers exposed to COVID-19 patients (NCT04334928). The two drugs have been reported by a recent computational work as potential inhibitors of RdRp of SARS-CoV-2 [55,66], yet this potential is to be experimentally confirmed.
Likewise, tenofovir alafenamide (Vemlidy; 2016) is an adenine-based acyclic nucleotide analog that, following activation, acts as a competitive inhibitor of reverse transcriptase and DNA chain elongation termination. The activation of the drug is, however, different and it usually takes place in infected cells by a series of bio-transformations similar to those of remdesivir (Figure S2) [67]. The main advantage of the prodrug, relative to the former prodrug, is that it increases the drug’s oral bioavailability, intestinal diffusion, selectivity of targeting the infected cells, and intracellular half-life. It also decreases the potential renal toxicity of the monophosphate intermediate. Tenofovir alafenamide was first approved in 2016 by the U.S. FDA and is prescribed for the oral treatment of HBV infection [68]. It is also available in many other combinations with emtricitabine (Descovy; 2016), bictegravir and emtricitabine (Biktarvy; 2018), emtricitabine and rilpivirine (Odefsey; 2016), and darunavir/cobicistat and emtricitabine (Symtuza; 2018). The efficacy of emtricitabine and tenofovir alafenamide as a prophylactic combination against SARS-CoV-2 infection is being evaluated in a large randomized, double-blind, controlled with placebo clinical trial for health care providers exposed to COVID-19 patients (NCT04405271; n = 1378).

2.6. Favipiravir (Avigan, T-705)
Favipiravir was originally developed by Fujifilm group, Japan. It is a pyrazine-carboxamide derivative (Figure 4) with a broad-spectrum antiviral activity. It selectively and potently inhibits the RdRp of RNA viruses [69]. Favipiravir is a prodrug that requires bioactivation in host-infected cells. Its active form is favipiravir-ribose-5′-triphosphate. The first step in the formation of the active species is potentially catalyzed by human hypoxanthine guanine phosphoribosyl-transferase [70], which converts favipiravir into ribose-5′-monophosphate intermediate. The latter intermediate undergoes two phosphorylation steps mediated by the action of host kinases leading to the formation of the ribose-5′-triphosphate active form.
Favipiravir is effective against several strains of influenza viruses, including those that are resistant to existing anti-influenza drugs. Favipiravir also showed an antiviral activity in experimental animals against other RNA viruses, including arenaviruses, alphaviruses, bunyaviruses, and flaviviruses [71]. Furthermore, preliminary results also indicated that favipiravir potentially possesses a moderate activity against Ebola [72]. Importantly, a recent nonrandomized, open-label study in patients (n = 80) with non-severe COVID-19 showed that favipiravir (1600 mg orally twice daily on the first day, then 600 mg orally twice daily for thirteen days) with INF-α had significantly better therapeutic effects on SARS-CoV-2 infection, in terms of disease progression and viral clearance, than lopinavir/ritonavir with INF-α [73]. Furthermore, an open-label, prospective, randomized, multicenter study in adults (n = 236) with COVID-19 pneumonia in China revealed that favipiravir (1600 mg orally twice daily on the first day, then 600 mg orally twice daily for 7–10 days) was associated with a higher 7-day clinical recovery rate compared to a control group treated with umifenovir, a potential inhibitor of the membrane fusion stage during the virus infection, (200 mg three times daily for 7–10 days). The 7-day clinical recovery rate in patients with moderate COVID-19 pneumonia was 71% in the favipiravir-treated patients, whereas the rate was 56% in the umifenovir-treated patients. Likewise, the 7-day clinical recovery rate in patients with severe to critical COVID-19 pneumonia was 6% versus 0%, respectively [74]. Currently, favipiravir is being studied alone or in combination with tocilizumab, hydroxychloroquine, or oseltamivir for the treatment of COVID-19 in more than 23 clinical trials across the world.
As of now, favipiravir is not available in the U.S. or European countries, perhaps because the animal experiments showed that the antiviral agent can be associated with teratogenic effects. Favipiravir is contraindicated in women with known or suspected pregnancy [75]. Favipiravir is also associated with QT prolongation [76]. It is currently approved to treat novel or re-emerging influenza outbreaks in China and Japan, and it is available as an oral solid dosage form [73,74,76].

2.7. AT-527
It is an investigational, orally active, purine nucleotide prodrug (Figure S3), which has exhibited antiviral activity against many single-stranded, enveloped RNA viruses, including human flaviviruses and coronaviruses [77]. It is a potent inhibitor of viral RdRp [78]. Following oral administration as hemi-sulfate salt, the drug gets converted to the monophosphate form via multiple metabolic activation steps. The first step is catalyzed by the action of human carboxylesterase 1 (CES1) and/or cathepsin A (CatA) to produce the L-alanyl intermediate. Spontaneous hydrolysis followed by histidine triad nucleotide-binding protein 1 (HINT1)-mediated hydrolysis results in the formation of the monophosphate metabolite. Then, the monophosphate is transformed to guanosine analog by adenosine deaminase like protein 1 (ADALP1) and further phosphorylated by guanylate kinase 1 (GUK1) and nucleoside diphosphate kinase (NDPK) to the pharmacologically active form of AT-527 diphosphate (also reported as AT-9010) (Figure S3) [78]. The safety, pharmacokinetics, and antiviral activity of AT-527 was earlier established in HCV-infected subjects with and without cirrhosis [79]. The drug is currently being evaluated in a phase 2 double-blind, randomized, placebo-controlled study to determine its efficacy and safety in patients with moderate COVID-19 symptoms (NCT04396106; n = 190).

2.8. EIDD-2801
It is the isopropyl-ester prodrug of β-D-N4-hydroxycytidine (Figure 5A). The prodrug has improved oral bioavailability as it avoids phosphorylation of the N4-hydroxyl group in the gastrointestinal tract. It is hydrolyzed in vivo to release the parent (EIDD-1931), which distributes into tissues, and upon tri-phosphorylation, it becomes the active triphosphate form. The tri-phosphorylated form has a broad-spectrum antiviral activity against various RNA viruses, including influenza, Ebola, Venezuelan equine encephalitis virus, MERS-CoV, SARS-CoV, SARS-CoV-2 and related zoonotic group 2b or 2c bat coronaviruses [80,81]. It also demonstrated increased potency against a coronavirus with resistance mutations to remdesivir [82]. By the action of RdRp, the active form is incorporated into the genome of RNA viruses, leading to the accumulation of mutations known as viral error catastrophe [80]. The active form exists in two forms (Figure 5B): the oxime form which mimics uridine and pairs with adenosine, while the other tautomer mimics cytidine and pairs with guanosine [81]. In mice infected with MERS-CoV or SARS-CoV, EIDD-2801 administration was found to diminish the virus titer and body weight loss and to improve pulmonary function [80]. Reduced MERS-CoV yields in vitro and in vivo was because of the increase in transition mutation frequency in only the viral RNA. The drug produced similar results in human airway epithelial cells. The drug showed similar results as a prophylactic and as a treatment [80].
The drug was developed at the Emory Institute for Drug Development and it was tested in a phase 1 randomized, double-blind, placebo-controlled, first-in-human study designed to evaluate its safety, tolerability, and pharmacokinetics following oral administration to healthy volunteers (NCT04392219; n = 130). It is now being tested in two phase 2 trials in COVID-19 patients (NCT04405570; n = 44 and NCT04405739; n = 60).