The World Health Organisation (WHO) declared the coronavirus disease (COVID-19) outbreak as a global pandemic on 12 March 2020. As the novel coronavirus SARS-CoV-2 continues to spread worldwide, it has already resulted in over 110 million cases and 2.5 million deaths. Multiple mutant strains have also complicated the landscape. While numerous studies are under way worldwide to uncover its prevention and cure, there are currently no known therapeutics for COVID-19. SARS-CoV-2, however, is not the first coronavirus to cause serious outbreaks. COVID-19 can be compared with previous human coronavirus diseases, such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS), to better understand the development of treatments. This rapid review with elements of evidence briefing aims to provide a comprehensive overview of current and emerging COVID-19 treatment options based on clinical trials. Finally, we will highlight new emerging therapies and the future prospects of this pandemic.
Since its discovery in December 2019, the coronavirus SARS-CoV-2 has been responsible for the worldwide COVID-19 pandemic. The coronavirus infects humans through the binding of its spike glycoprotein (S) with the host cell receptor, angiotensin-converting-enzyme 2 (ACE2). Peak viral replication occurs in the initial 7-10 days of infection and the primary immune response occurs in days 10-14. This progresses to either viral clearance or an uncontrolled immune response (cytokine storm) . The time of therapy administration is crucial in determining COVID-19 outcomes. However, with no known cure, as of 8 March 2021, more than 100 million people have been infected and at least 2.5 million have died . In addition, mutant strains have contributed to further uncertainty. The major impact on human health is only one aspect of the pandemic. Public health interventions to “flatten the curve”, including nationwide lockdowns, social distancing and travel restrictions, will have lasting effects on the economy, human behaviour and education . Due to the absence of effective treatments and vaccines, current measures are largely reliant on public compliance which have uncertain benefits  and puts the world at risk of multiple waves of infections . It is therefore of great interest and urgency that treatments for COVID-19 should be found.
There are currently seven coronaviruses known to infect humans: the endemic HCoV-HKU1, HCoV-OC43, HCoV-NL63 and HCoV-229E which cause 15% of the cases of the common cold; and the epidemic SARS-CoV, MERS-CoV and SARS-CoV-2 (Table 1). SARS-CoV which caused SARS and MERS-CoV which causes MERS are much more deadly than previous coronaviruses. SARS rapidly spread to 29 countries globally, causing an epidemic with 8096 cases, 774 mortalities and a CFR of 11% . Likewise, MERS spread to more than 27 countries and as of November 2019 has resulted in 2494 cases and 858 deaths. Its 35% CFR makes it the deadliest coronavirus to date . As patients acquire immunopathological damage, they can present with dangerous complications such as acute respiratory distress syndrome (ARDS) which developed in around 16% of SARS  and over 50% of MERS patients .
Table 1. Characteristics of coronaviruses that infect humans
ARDS is a set of pulmonary conditions caused by uncontrolled inflammation, leading to acute respiratory failure. In response to SARS-CoV, MERS-CoV and SARS-CoV-2, the body produces a massive immune response, producing pro-inflammatory neutrophils and cytokines like IL-6, TNFα and IL-1β . This “cytokine storm” attacks lung tissue and pneumocytes to cause vascular leakage and pulmonary fibrosis. Damage to type 1 cells cause decreased perfusion by shunting while damage to type 2 cells puts patients at risk of lung collapse due to decreased surfactant production. Thrombi formation due to activation of the coagulation cascade in response to the fluid exudates further restricts oxygen exchange .
SARS-CoV-2 is the first coronavirus to cause a pandemic of its scale in modern times, which could be attributed to a high basic reproduction number (R0). The R0 of COVID-19 is estimated to be 3.77, which is higher than both SARS and MERS . After the incubation period, COVID-19 patients develop moderate symptoms lasting around 5-8 days . Around 5%-15% of COVID-19 patients develop ARDS, acute organ failure and shock . The case fatality rate (CFR) varies by country – from less than 0.1% in Singapore to over 19% in Yemen . Chest x-rays appear consistent with that of pneumonia, and laboratory findings include lymphopenia, thrombocytopenia, and leukopenia . Similar to SARS and MERS, children are less likely to become infected and present with milder cases  whereas older men with co-morbidities have higher mortality .
In face of the current crisis, numerous treatments are being tested based on past experience of SARS and MERS. There are 3 common approaches for the discovery of COVID-19 drugs. The first is screening chemical libraries with databases containing large numbers of existing compounds that may have antiviral properties (eg, remdesivir). Second is the testing of existing broad-spectrum antiviral drugs (eg, ribavirin and interferons). The final method involves the de novo development of novel, specific agents such as monoclonal antibodies based on the genomic and biophysical understanding of the SARS-CoV-2 . Development of novel agents takes years and given the urgency of the current pandemic, COVID-19 treatments are mainly repurposed from SARS and MERS.
Promising treatments have been approved for worldwide clinical trials. Remdesivir; Lopinavir/Ritonavir; Interferon β; hydroxychloroquine and dexamethasone are some of the drugs included in the WHO Solidarity Trial and the Recovery Trial [38,39]. Novel vaccines, emerging “cytokine storm” targeting treatments and convalescent plasma therapies are also being explored. Aiming to provide a comprehensive overview of the area of COVID-19 therapeutics, the main objective of this rapid review is to describe existing SARS-CoV-2 therapeutic approaches and present the results of their current COVID-19 clinical trials.
The databases Medline, Embase and WHO COVID-19 were searched for all publications from 2019 to 9 February 2021 using a combination of the terms “COVID 19”, “SARSCoV2”, “coronavirus”, “therapy”, “lopinavir”, “ritonavir”, “ribavirin”, “remdesivir”, “hydroxychloroquine”, “corticosteroids”, “anakinra”, “tocilizumab”, “convalescent plasma”, “interferon” and “monoclonal antibodies”. (Table S1 in the Online Supplementary Document). From the articles retrieved, additional references were identified by a manual search among the cited references and through searching journals online. The types of studies included were clinical trials, case reports, observational studies and systematic reviews. To prevent bias, studies with less than 10 patients were not included. Articles which did not have full text available or were not in English were also excluded.
Data extracted from eligible studies included: Drug/treatment, author, year, study aim, study type, study design, status, main findings and limitations.
The literature search and screening for eligible articles was conducted by 2 reviewers (AS, YD). Data extraction was performed by 2 reviewers (AS, YD).
After removal of duplicates, the literature search identified 1409 unique articles. A further 7 studies were obtained through manual and citation searches. From the 1416 total articles identified, 1114 were excluded after title review, 234 were excluded after abstract review. Sixty-eight articles were reviewed by full text and a total of 40 articles were selected to be included (Figure 1). This rapid review includes 4 studies on remdesivir, 6 on convalescent plasma, 7 on hydroxychloroquine (Table 2), 2 on lopinavir/ ritonavir, 4 on interferon, 7 on corticosteroids, 6 on cytokine storm inhibitors (Table 3) and 6 on monoclonal antibodies (Table 4). Table 2, Table 3 and Table 4 summarise the studies we reviewed regarding current COVID-19 therapy and trials.
Figure 1. Flowchart summarising study identification and selection.
Table 2. Summary of COVID-19 therapy studies: Compounds with anti-viral properties
Table 3. Summary of COVID-19 therapy studies: Existing broad-spectrum antiviral drugs
Table 4. Summary of COVID-19 therapy studies: Novel specific treatment agents
Compounds with anti-viral properties
Remdesivir (GS-5734) is a broad-spectrum nucleoside analogue which is metabolised by the host cell and converted into nucleoside triphosphate (NTP). NTP targets RNA-dependent RNA polymerase (RdRp) which inhibits the transcription of viral RNA  (Figure 2). Remdesivir was developed in 2009 by Gilead Sciences . It is most commonly known as a drug to treat Ebola but was proved to be an inferior treatment to other contenders .
Figure 2. SARS-CoV-2 gene targets. Proteins are translated from the SARS-CoV-2 genome. Lopinavir targets 3CL protease produced by the ORF1a gene. Remdesivir and ribavirin targets RNA-dependent RNA polymerase (RdRp) translated from ORF1b. Chloroquine/ hydroxychloroquine, convalescent plasma and monoclonal antibodies have effects against the spike protein.
Remdesivir is currently FDA approved for emergency administration to COVID-19 patients. A study in severe COVID-19 patients showed very positive results. 68% of patients improved and only 13% of patients died . These findings are highly significant, considering the mortality of patients with severe COVID-19 is over 40% . However, viral load was not collected to confirm the antiviral effects and the patient cohort was small and therefore there is a higher chance of false-positive results.
Remdesivir was studied further in a randomised, double-blinded, placebo-controlled trial of 237 patients in which Wang et al reported a statistically insignificant median recovery time of 21 days compared to 23 days in the control . There was no decrease in mortality rate and more patients receiving the treatment dropped out of the trial due to adverse effects which included gastrointestinal symptoms and worsened cardiopulmonary status. This trial did not complete full enrolment. A randomised, placebo-controlled, double-blinded clinical trial of 1062 patients where patients received the same drug dose as the previous study showed that remdesivir had a superior recovery time of 10 days compared to 15 days in the placebo group . Contrary to Wang, this trial found that compared to the placebo, patients who received remdesivir experienced less adverse effects and the mortality rate decreased from 11.9% to 6.7%. Although this is significant, the high mortality rates reported make it clear that remdesivir alone is unlikely to be sufficient in COVID-19 treatment. This conclusion is supported by the results of a randomised clinical trial which found that despite patients receiving a 5-day course of remdesivir having a significantly better status, it was not clinically important . In accordance with in vivo studies, the drug is most likely only beneficial to patients in the early stages of infection (Figure 3).
Figure 3. Progression to cytokine storm. Peak viral replication occurs in the initial 7-10 days and primary immune response usually occurs by day 10-14 which is followed by virus clearance . Therefore, therapies appear to be most beneficial when given before 14 days of infection. Days 0-5: incubation period. SARS-CoV-2 enter cell via ACE2 receptors on human type 1/2 pneumocytes and start replicating. Hydroxychloroquine (HCQ) targets glycosylation of viral surface. Lopinavir, remdesivir targets the viral replication. Until day 10: moderate symptoms. The initial immune response produces IL-18, IP-10, MCP-1 and MIP-1α which starts to recruit specific immune cells. Interferon (IFN), HCQ and convalescent plasma (CP) is most effective. Day 10: severe infection leading to risk of mortality. Macrophages via BTK, T-cells via JAK, neutrophils and B-cells produce cytokines, resulting in a cytokine storm by day 14. Ruxolitinib, baricitinib, tocilizumab, anakinra, acalabrutinib and corticosteroids can be used .
In convalescent plasma (CP) therapy, blood plasma is taken from a patient previously infected by the virus and transfused into a symptomatic patient. The theory is that the antibodies transferred to the symptomatic patient will suppress viremia . Antibodies, or immunoglobins, bind to invading viruses which neutralises the pathogen (Figure 3). It can also cause virus destruction by processes like activating the complement pathway or opsonisation via viral Fc receptors. CP is generally regarded as safe and has been recorded to be used for the Spanish Flu pandemic as early as 1918  and has been used to treat Machupo virus, Junin, Lassa fever and Ebola virus . It also reduces mortality rates in patients with severe influenza A .
Convalescent plasma therapy is currently FDA approved to be used in COVID-19 patients for investigational purposes. Observational studies have demonstrated it to be safe and patients treated have a significantly improved oxygen saturation, CRP, 30-day survival rate and reduced need for mechanical ventilation [45,46]. Results also suggest CP could alleviate inflammation and overreaction of the immune system. This potential is further demonstrated by a study which found a 76% improvement rate in patients, of which more than half had ARDS .
A randomised, open label clinical trial was performed on 103 patients with life threatening COVID-19 . Results show that CP therapy was clearly highly effective in viral clearance. Treatment was associated with 87% negative conversion rate of viral PCR compared to 37.5% of the control group. Although those patients receiving the treatment had a shorter time to clinical improvement and a lower 28-day mortality rate, these results were found to be statistically insignificant. The study was terminated early due to a shortage of patients. Another randomised controlled trial involving 1210 patients found no clinical or mortality improvement between patients receiving CP and patients receiving standard care. There was also no evidence of anti-inflammatory properties . Zeng et al found that treatment inhibits viral shedding but doesn’t reduce mortality, and late treatment intervention could even be attributed to poorer clinical outcomes . In concordance with previous findings in SARS and MERS, the authors recommend transfusion to be given before day 14 of infection.
Hydroxychloroquine and chloroquine are 4-aminoquinolines. Chloroquine is thought to be a drug that blocks viral entry by inhibiting glycosylation of cell surface receptors, proteolytic processing and endosomal acidification . (Figure 2). It also has immunomodulatory effects and inhibits cytokine production, so could potentially help patients with cytokine storm (Figure 3). Hydroxychloroquine (HCQ), an analogue of chloroquine, has the same antiviral mechanisms but is associated with less drug-drug interactions and adverse effects . Both drugs are used to treat malaria, lupus and rheumatoid arthritis, and are cheap and easy to attain .
Hydroxychloroquine is one of the most controversial treatments and is not recommended by the FDA. The WHO discontinued its trial in July 2020 which followed their temporary suspension of the trial in May after a French hospital found the adverse effects of prolonged QTc heart rate intervals caused death due to sudden cardiac arrest . There was a rapid increase in demand of the drug after USA president Donald Trump publicly claimed to use the drug . Initial trials showed both hydroxychloroquine and chloroquine displayed antiviral effects against SARS-CoV-2 in vitro . Strikingly, an open label, non-randomised hydroxychloroquine clinical trial in France found that patients had clear nasopharyngeal viral results in 3 – 6 days compared to the mean clearance time of 20 days .
Other studies show contrasting findings. A randomised, controlled trial of 4716 patients by the Recovery Group found that not only did hydroxychloroquine not provide any benefit, but it also resulted in worse outcomes . The 28-day mortality rate of patients receiving hydroxychloroquine was 27% compared to 25% in those receiving standard care. The group receiving the drug was also associated with longer hospitalization and higher progression to invasive mechanical ventilation. A double-blinded, randomised trial was conducted in patients with severe COVID-19 where half were given low dosages of hydroxychloroquine and half were given high dosages of hydroxychloroquine . Worryingly, patients receiving high dosages had a mortality rate of 39% compared to 15% receiving low dosages. There was also no evidence of viral clearance. Another randomised, double-blind placebo trial showed that the drug did not prevent postexposure prophylaxis for the illness either . In addition to cardiac side effects, hydroxychloroquine has been reported to cause vomiting and abdominal pain in half of the patients treated . A randomised trial of patients with moderate COVID-19 symptoms conducted in China, found that hydroxychloroquine not only failed to provide benefit, but caused a much higher rate of mainly gastrointestinal adverse effects . Hydroxychloroquine is unlikely to be beneficial in the treatment of COVID-19 but it has been suggested that if given at a low dose with an anti-inflammatory drug it might help mitigate the cytokine storm in critically ill patients .
Existing broad-spectrum antiviral drugs
Lopinavir and ritonavir
Based on clinical evidence in SARS-CoV and MERS-CoV, the most commonly used antivirals for COVID-19 are lopinavir, ritonavir and ribavirin. Lopinavir/ritonavir (LPVr) combined formulates Kaletra, an FDA approved drug used to treat HIV. Lopinavir inhibits 3-chymotrypsin-like protease (3CLpro) which disrupts viral replication and RNA release from host cells  (Figure 2). Ritonavir is a human CYP3A4 enzyme inhibitor. In addition, lopinavir efficacy was found to be inferior to remdesivir and interferon-β . Randomised, controlled clinical trials of COVID-19 patients found that LPVr was not associated with clinical improvement [57,58]. Furthermore, up to 50% of patients could experience adverse effects including gastrointestinal problems and hepatoxicity .
Interferons (IFNs) are cytokines discovered to have antiviral properties in 1957 and have since been approved for clinical use by the FDA . They are widely available to treat hepatitis, leukaemia and carcinoma . Human cells naturally produce type I interferons, IFN-α and IFN-β, in response to viral infections. Of these, IFN-β is the most potent . They are crucial in inhibiting viral replication and can also activate the transcription of hundreds of genes which produce virus controlling proteins  (Figure 3).
IFN-β 1a combined with Lopinavir/Ritonavir was previously approved for international trialling by WHO. A randomised clinical trial of 81 patients found that IFN-β 1a decreased the 28-day mortality rate from 43.6% to 19%, although it did not change time taken to reach clinical response . Further support of this treatment was found in a recent randomised, double-blind, placebo-controlled trial which found patients receiving nebulised IFN-β 1a were twice as likely to recover than the standard group (44% vs 22%) . Similarly, IFN-β 1b is also thought to be beneficial. The results of a randomised clinical trial in 66 patients showed that it caused a significantly shorter time for clinical improvement and it reduced 28-day mortality from 18.8% to 6% . However, an open-label, randomised, phase 2 clinical trial carried out in 127 patients suggested that although patients with moderate symptoms had a significantly better clinical and virologic response, it is unlikely to benefit patients with severe symptoms .
Corticosteroids are hormones naturally produced by the adrenal cortex, principally in the form of glucocorticoids and mineralocorticoids. They have widespread effects on the human body including metabolic regulation and reducing vasodilation . Glucocorticoids bind to glucocorticoid receptors (GR) on the cell cytoplasm and regulates gene expression . This can produce anti-inflammatory and immunosuppressant factors which is useful pharmacologically. Synthetic glucocorticoids include betamethasone which treats asthma, and prednisolone which treats COPD and rheumatoid arthritis.
Due to its controversy, corticosteroids were not previously recommended by WHO to treat COVID-19. However, on 2 September 2020 it was approved for the treatment of patients in critical condition . Theoretically, corticosteroids should inhibit the symptoms of a “cytokine storm” which only presents in patients classified as severe [63,64,95] (Figure 3). Clinical studies reported that around 20%-50% of observed COVID-19 patients received this treatment. In an observational study, Zhang et al suggested that corticosteroids could possibly produce early clinical benefits in COVID-19 . The effectiveness of corticosteroid therapy was contested by Zhou et al, who thought that high dose corticosteroids could have contributed to poor clinical outcomes in a cohort study . Recent findings have reinstated corticosteroids as an important drug in this pandemic. In contrast to most outcomes of corticosteroid trials , the steroid dexamethasone seems very promising. Furthermore, a recent review suggested that hydrocortisones has a similar effectiveness . A trial carried out in hospitalised patients by the Recovery Collaborative Group found dexamethasone reduced the mortality rate for patients on ventilators from 41.4% to 29.3%, and from 26.2% to 23.3% for patients needing oxygen . However, there was no benefit in its use for patients with moderate symptoms. For patients who were not receiving any respiratory support, patients who received dexamethasone had a mortality rate of 17.8%, compared to a 14% mortality in the control group. Prolonged corticosteroid treatment may be effective in saving lives of severe patients who develop ARDS and multiple organ failure, but short therapy is suggested to increase death, so cannot be thought as an effective cure .
Drugs targeting the cytokine storm
The most critically ill COVID-19 patients develop a massive immune response, resulting in a cytokine storm. This causes ARDS and multi-organ failure. Moderation of immune mediators such as Janus Kinase (JAK), bruton tyrosine kinase (BTK) and interleukins (IL) are crucial in saving patient lives (Figure 3). JAK inhibitors include baricitinib and ruxolitnib. A clinical trial found patients treated with baricitinib achieved significantly greater clinical improvements. After 2 weeks, 58% of the treated patients were discharged compared to 8% of the control group . It also significantly reduced IL-1β, IL-6, and TNF-α plasma concentrations . Patients given ruxolitinib in a randomised, controlled trial displayed significantly decreased levels of cytokines and numerically faster clinical improvement . Acalabrutinib, a BTK inhibitor, was found to improve symptoms of severe patients . In a cohort study the IL-1 receptor antagonist Anakinra was found to greatly reduce the need for invasive mechanical ventilation and decrease mortality in a cohort study from 73% in the historical group to 25% in the treatment group . This result has been supported by other similar studies . The therapeutic effects of these cytokine storm inhibiting drugs seem favourable, but more high quality, randomised controlled trials are needed to fully assess its affects.
Development of novel specific treatment agents
Monoclonal and polyclonal antibodies
Monoclonal antibodies are produced from immunized animals, antibody human phage libraries and memory B cells of recovered patients . Compared to convalescent plasma treatment, monoclonal antibodies are an alternative strategy which has higher efficacy, is able to be expressed in larger quantities and reduces the risk of antibody dependent enhancement .
Tocilizumab is a recombinant, anti-human IL-6R monoclonal antibody that inhibits hyperinflammation in patients with COVID-19. It is currently most well known in treating rheumatoid arthritis, another inflammatory immune disease. A randomised, double-blind, placebo-controlled trial in 389 moderate to severe patients found that tocilizumab improved the combined progression to either mechanical ventilation or death (12% vs 19.3%). However, it did not improve survival alone . Another randomised trial also found that there was no improvement in 28-day mortality in moderate and severe cases . In fact, in more critical cases, tocilizumab may lead to worse outcomes. A randomised, open label trial of severe or critical COVID-19 patients found death at 15 days increased from 3% in the standard care group to 17% in the patients receiving treatment and was linked to a higher occurrence of adverse events . The trial was not completed due to the large number of deaths.
REGN-COV2 is a monoclonal antibody cocktail that target the RBD of the S1 or the S2 segment of spike proteins, thus interfering with entry into the host cell. Recent results of a double- blind, randomised controlled trial of REGN-COV2 show a reduced viral load especially in patient whose immune response had not yet been initiated . Some other potential monoclonal antibodies effective against SARS-CoV-2 include B38, H4 and CR3022 [98,99].
In this rapid review, we summarise the literature on COVID-19 treatments. There is not yet a completely effective cure for COVID-19. This review found no certain evidence of any single therapy to significantly improve clinical outcomes. Of the 40 studies included, 18 reported clinical improvement while 17 reported no significant result and 5 reported a worsening of clinical progression. Most of the studies reporting clinical improvement are observational and have a limited sample size. Many of these studies report that although the therapy studied was effective in improving patient recovery time and symptoms of mild disease, it is not effective in preventing mortality rates of severe patients. In contrast, many randomised controlled trials report no significant improvements or even worsened outcomes.
Timing of administration is crucial in drug efficacy. In accordance with the SARS-CoV-2 peak replication time and the patient’s primary immune response, therapies appear to be most beneficial when given before 14 days of infection [1,49,100,101]. Broad-spectrum antiviral treatments could benefit patients with mild symptoms in the early stages of infection but are ineffective in treating severe disease patients. Remdesivir, convalescent plasma and interferon were found to reduce viral load and improve patient recovery time but failed to prevent mortality. Conversely, dexamethasone and hydrocortisones have been proven to save lives in severe patients but are not effective in mild cases. Other drugs like hydroxychloroquine do not seem to be effective in either mild or severe patients.
There is evidence that remdesivir can improve recovery time in COVID-19 patients and reduce the mortality rate from 11.9% to 6.7% . Nonetheless, other randomised controlled trials suggest that any benefit is not clinically significant [41,43]. These results are consistent with in vitro and animal studies. Remdesivir is highly effective in reducing replication of SARS-CoV-2 in human cell lines . However, in rhesus macaques, although it reduced clinical disease and lung damage, there was no reduction of virus replication in the upper respiratory tract . Current evidence from previous coronaviruses do not suggest added benefit of the drug. In initial studies, remdesivir was found to have prophylactic and therapeutic effects against SARS- and MERS-CoV in human airway epithelial cells and in animals, as well as having antiviral properties against HCoV-299E and HCoV-OC43 . However when used therapeutically in MERS patients, it was unable to prevent mortality or loss of pulmonary function .
Convalescent plasma therapy was found to be effective in reducing viral clearance in COVID-19 patients [47,49]. However, there is limited effect on the clinical outcomes of patients [47,48]. Efficacy of CP was initially supported in trials involving SARS and was reported to be effective in decreasing mortality rate in several studies . In deteriorating SARS patients, switching to CP therapy was associated with better outcomes than continuing high dose methylprednisolone . Studies on MERS better reflect the outcomes for COVID-19 patients. Initial experiments on immune camel serum was found to be able to diminish weight loss and lung histological changes, and improve virus clearance in mice . However, it was found that antibody titres collected from recovered patients were too low in the plasma to covey any therapeutic effect . Thus in 2014, the WHO’s stance on CP changed from being the most promising near-term MERS therapy to be only regarded as investigational . In concordance with findings in SARS and MERS, it is recommend transfusion to be given early, before day 14 of infection, for greatest therapeutic effect in COVID-19 patients .
A treatment suggesting very limited potential is hydroxychloroquine. The drug was associated with a higher mortality of 27% compared to 25% in the control, as well as longer hospitalisation . There are also several worrying side effects including cardiac symptoms. In previous coronaviruses, chloroquine was shown to inhibit SARS-CoV, MERS-CoV and HCoV-229E in vitro , . Hydroxychloroquine was similarly effective in SARS, suggesting it to be a potential pharmacological agent for coronaviruses . However, the effects of chloroquine/ hydroxychloroquine were never fully explored in vivo  and the discrepancy with COVID-19 highlights the need for vigorous drug testing upon translation from in vitro studies to patients.
Broad spectrum antivirals LPVr are common treatments for COVID-19 patients. However, it is unlikely to provide any clinical improvement [57,58]. Lopinavir inhibits SARS-CoV-2 in vitro but it was recently found that ritonavir did not . Studies in SARS and MERS provide further evidence of their limitations. Although LPVr inhibited the replication of SARS-CoV, it was not effective in MERS-CoV . Other antivirals such as Ribavirin, a nucleoside analogue that inhibits viral RdRp is similarly unlikely to provide clinical value. In vitro it was found that it could only inhibit SARS-CoV at very high levels which is difficult to achieve clinically  and has no benefit in SARS and MERS patients . Although unlikely to be effective alone, broad spectrum antivirals may play a role in conjunction with other therapies.
Several randomised controlled trials demonstrate the benefits of IFN-β. The drug was shown to improve time for clinical improvement and recovery, and also decrease 28-day mortality from 43.6% to 19% . Evidence from previous years show IFN is able to inhibit SARS- and MERS-CoV replication both in vitro and in animals  and had maximum effect used in combination with the broad spectrum antivirals, lopinavir/ritonavir and ribavirin . A clinical trial of IFN- α 2a and ribavirin in MERS patients showed improvement in 14-days survival but not 28-day survival. Significantly, IFN-β 1a resulted in a mortality rate of 64% compared to 85% in IFN- α 2a .
It is suggested that corticosteroids such as dexamethasone are vital in preventing mortality in patients with severe COVID-19 . Current evidence reported from SARS and MERS suggest varied benefit. A non-randomised and uncontrolled trial showed that 89% of 107 patients who received high dose methylprednisolone recovered from the disease . In contrast, a randomized, double-blinded, placebo-controlled trial showed that corticosteroids caused an increase in patient viral load and decrease in viral clearance . Similarly in MERS, a study concluded that corticosteroids not only had no effect in preventing ARDS and pulmonary fibrosis, it was also associated with osteonecrosis, delirium and aspergillosis . Further contradiction arises in a 2020 randomised trial of ARDS patients which found that methylprednisolone was associated with a significant increase in death .
There is promise in new cytokine storm targeting drugs like JAK and BTK inhibitors, but further investigation is required to discern its value for treating critically ill. Previously, monoclonal antibodies were shown to have neutralizing effects in vitro, in mice and in rhesus macaques for SARS and MERS [90,116]. However, there is no evidence that tocilizumab is effective in treating COVID-19 patients and REGN-COV2 requires further investigation to fully understand its effects.
Future prospects of the pandemic could depend mainly on the development of novel therapies like monoclonal antibodies and prevention through effective and widespread vaccination which will be effective against mutated strains. Meanwhile, to halt the spread of the virus, social distancing and good hygiene practises such as handwashing and face coverings or masks are essential .
As of 8 March 2021, there are 308 candidate COVID-19 vaccines of which 16 have entered phase 3 trial, and 4 have entered phase 4 . There are several vaccines currently in use. The WHO has issued emergency use listings (EULs) for Pfizer-BioNtech COVID-19 vaccine and AstraZeneca/Oxford COVID-19 vaccine. The FDA have further approved of the Pfizer-BioNtech, Moderna and Janssen vaccines. ChAdOx1 nCoV-19, developed by the University of Oxford and AstraZeneca is in use in the UK. It is a chimpanzee adenovirus-vectored vaccine. Results from phases 1/2 and 2/3 trials show neutralising antibody responses in >99% of participants by 14 days, and potent cellular and humoral immunogenicity in all participants after a second dose [119,120]]. Interim results also show up to 71.2% efficacy from the phase 3 Brazil trial . BNT162b2 developed by Pfizer and Biotech, is a mRNA vaccine. It is also being rolled out in the UK, as well as the US and following phase 3 trial is reported to be 95% effective . Prospective studies of ChAdOx1 nCoV-19 and BNT162b2 from real world data in the UK confirms their protective effect against hospitalisation [123,124]. Another mRNA vaccine, mRNA-1273, developed by Moderna is currently rolled out in the US, and is reported to be 94.5% effective on preventing COVID-19 .
Other vaccines include the CoronaVac (formerly PiCoVacc) which was found in phase 1 and 2 trials to be safe in inducing neutralizing antibodies with a 97.4% seroconversion rate after 28 days . Other promising vaccines are the adenovirus vectored vaccines. Phase 1 and 2 trials on Ad5 by CanSino Biologics show its safety in causing a 97% antibody seroconversion rate in participants by 28 days . The Janssen vaccine (Ad26) by Johnson& Johnson is similarly favourable in causing neutralising antibodies in over 90% of participants .
This review presents a current overview of COVID-19 therapies and links their development to their efficacy in previous coronaviruses for a more comprehensive picture of their likely trajectory. Future directions of the pandemic are considered by discussing initial studies of novel therapies and vaccines. There are also a number of limitations. First, due to the massive amount and fast pace of publications regarding COVID-19, new findings are constantly changing the treatment landscape. We included some preliminary results of ongoing studies in which not all data was available or presented. Second, due to more limited evidence, we were unable to include all identified COVID-19 therapies in this review such as oseltamivir and colchicine. Third, in observational and retrospective studies included, the duration and dose of therapy were not standardised and may be subject to bias. Fourth, the literature was limited to studies published in English and was searched for in only two databases. The search and screening process was not screened twice.
While the pandemic continues to spread worldwide, there is urgent need to understand the benefits and risks of each treatment. Living drug treatment systematic reviews which updates when new evidence becomes available can be followed . Currently there is no single effective treatment against COVID-19. A combination of therapies administered at different stages of infection seems to provide the best outcomes in patients. As most clinical trials are carried out in severe patients, there is a need for more high quality randomised controlled trials, especially of patients in earlier stages of infection. An effective and safe vaccine distributed worldwide will be a turning point in resolving the COVID-19 pandemic.
UNCOVER (Usher Network for COVID-19 Evidence Reviews) authors that contributed to this review are: Ms Marshall Dozier, Dr Ruth McQuillan, Prof Harish Nair, Ms Emilie McSwiggan, Prof Gerry Fowkes.