key: cord-0703564-fsvacjgk
authors: Martinez-Vizcaino, V.; Mesas, A. E.; Cavero-Redondo, I.; Saz-Lara, A.; Sequi-Dominguez, I.; Pascual-Morena, C.; Alvarez-Bueno, C.
title: The race to find a SARS-CoV-2 drug can only be won by a few chosen drugs: a systematic review of registers of clinical trials of drugs aimed at preventing or treating COVID-19
date: 2020-05-09
journal: nan
DOI: 10.1101/2020.05.05.20091785
sha: e56fe4f8a6a5898a51e59b58e916c0cbb315d49a
doc_id: 703564
cord_uid: fsvacjgk

Considering the massive amount of clinical trial registers aimed to find effective drugs for the prevention and treatment of COVID-19, it is challenging to have a comprehensive view of which drugs are being studied more extensively and when is expected that we will have consistent results regarding their effectiveness. This systematic review included all clinical trials on pharmacological therapy related to COVID-19 and SARS-CoV-2 registered at the International Clinical Trials Registry Platform (WHO-ICTRP) up to April 22, 2020. Clinical trials characteristics (country, design, sample size, main outcomes, expected completion data, type of participants, length of the interventions, main outcomes). How many trials and he accumulated sample size by drug or combination of drugs, and by month in 2020 was depicted. We identified 412 clinical trials registers addressing the effect of pharmacological treatments on COVID-19, predominantly from Asia and Europe (42.2% and 31.1% of clinical trials registers, respectively). The most main outcomes studied were clinical recovery (54.4% of the clinical trials registers, respiratory recovery (28.2%) mortality (27.4%), viral load/negativity (20.4%). During 2020, a huge amount of clinical trials are expected to be completed: 41 trials (60,366 participants) using hydroxychloroquine, 20 trials (1,588 participants) using plasma, 18 trials (6,830 participants) using chloroquine, 12 trials (9,938 participants using lopinavir/ritonavir, 11 trials (1,250 participants) using favipiravir, 10 trials ( 2,175 participants) using tocilizumab and 6 trials (13,540 participants) using Remdesivir. The distribution of the number of registered clinical trials among the different therapeutic options leads to an excess of sample size for some and a lack for others. Our data allow us to conclude that by the end of June we will have results of almost 20 trials involving 40000 patients for hydroxychloroquine and 5 trials with 4500 patients for remdesivir; however, low statistical power is expected from the 9 clinical trials testing the efficacy of favipiravir or the 5 testing tocilizumab, since they will recruit less than 1000 patients each one.

The alarming figures of people that have suffered or died by COVID-19 have elicited an impressive pressure on public and private institutions to search drug treatments to prevent and treat the coronavirus. Even the agencies in charge of approving new drugs, or new indications for existing ones, are under extraordinary pressure that could even affect their well-deserved prestige as institutions guided solely and exclusively by scientific rigor in decision-making [1, 2] . Investigators and research agencies are determined not to repeat the main mistake made when facing, in 2014, the West African Ebola Epidemic, when, although they produced a plethora of experiments plenty of good intentions, the randomized clinical trials started too late and were not able to recruit enough patients [3] .

The spread of the virus appears to have been controlled in Asia and seems to be declining in most European and North American countries; moreover, it is anticipated that summer in the northern hemisphere will significantly reduce the infectivity and virulence of SARS-CoV-2 [4] . However, since no vaccine is expected to be available by then, everything seems to indicate that a resurgence of this pandemic will occur in autumn [5] .

As consequence of these hopeless predictions, a desperate race has begun aimed to demonstrate the efficacy of new drugs, and more urgently, to demonstrate that drugs already on the market for other clinical indications such as hydroxychloroquine, remdesivir, azithromycin, etc., are efficacious to manage the SARS-Cov-2 infection [6, 7] .

In this context, a synthesis effort is lacking to summarize the characteristics of the ongoing clinical trials, the expected dates of the completion of the recruitment, where they are being held, the drugs whose efficacy is being tested and the main outcomes of these trials. Although there are several databases and registries online, there is a concern about when we can expect to have results with a sufficient statistical power to clear up any doubts about the effectiveness of the different therapeutic approaches for the treatment or prevention of the COVID-19 disease.

The objective of this comprehensive systematic review is to gather and synthesize the information included in the clinical trial registers of candidate drugs to prevent and treat COVID-19 according to the pharmacological group and specific drugs name, study design, main outcomes, number and characteristics of participants recruited, and expected completion date. In addition, we graphically represent which drugs are most likely to achieve consistent results over the coming months of 2020.

Due to the novelty and importance of COVID-19, information regarding unpublished trials was systematically searched through the main clinical trial registries. First, we 

The criteria for the inclusion of clinical trials were as follows: (i) participants-patients with a COVID-19 diagnosis, a close contact with a COVID-19 patient or belonging to a high-risk exposure group such as health careers; (ii) design-randomized control trials (RCTs), nonrandomized experimental studies (non-RCTs) (including at least two-arm pre-post studies), and pilot studies; (iii) type of intervention-studies comparing the effect of pharmacological approaches as treatment or prevention of COVID-19; and (iv) outcomes-mortality, respiratory signs and symptoms' improvement, oxygen therapy,

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 9, 2020. 

The following data were extracted from each register: (1) sample size; (2) Additionally, the number of trials expected to be completed and the cumulative sample size expected per month for the year 2020 were shown graphically.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . https://doi.org/10.1101/2020.05.05.20091785 doi: medRxiv preprint Data extraction was independently performed by two reviewers (IC-R and AEM), and inconsistencies were solved by consensus of two more authors (VM-V, CA-B). The agreement rate between reviewers was reported by calculating kappa statistics.

The full dataset will be available in online Mendeley Data, a repository of research data that allows to assigns a permanent digital object identifier, un such a way that data of this study can be easily referenced (doi: http://dx.doi.org/10.17632/4rttjzsh4z.1).

We identified 412 clinical trials registers addressing the effect of pharmacological treatments on COVID-19 ( Figure 1 ), which were conducted in 40 countries, predominantly from Asia and Europe (42.2% and 31.1% of clinical trials registers, respectively), being China (26.5%), USA (13.6%), Iran (10.0%) and Spain (8.7%) the countries with the most registered clinical trials. Clinicals trials sample size ranged from 10 to 100,000 participants, and most clinical trials registers (91.0%) included confirmed COVID-19 patients (13.1% severe/critical clinical status) ( Table 1 ).

The estimated study completion date ranged from February 2020 to March 2025.

Regarding their experimental phase, the registers are distributed in percentages as follows: 2.4%, 1.5%, 26.5% and 29.4% for phases I, II, III and IV respectively. The remaining 14.3% corresponds with those trials including more than an experimental phase or that information is not clearly specified in the register. The allocation of participants to the study groups was done using randomized procedures in the 90.5% of the clinical registers, and 80.6% included two-arms (being placebo/control or usual care the most common comparison group). The intervention length ranged from 1 to 13 weeks.

On May 1, only 3.2% of the studies have completed the recruitment (Table 1) .

Regarding the COVID-19 outcomes studied, clinical recovery was included in the 54.4%, respiratory recovery in the 28.2%, mortality in the 27.4%, viral load/negativity in the 20.4% and other outcomes in the remaining 13.6%. Furthermore, the most frequent pharmacological treatment under analysis were: hydroxychloroquine (22.8%), chloroquine (6.6%), convalescents plasma (6.6%), lopinavir/ritonavir (5.6%), tocilizumab (4.6%), hydroxychloroquine + azithromycin (4.1%), favipiravir (3.4%), remdesivir (2.9%), methylprednisolone (2.7%), sarilumab (2.4%) and

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . hydroxychloroquine + lopinavir/ritonavir (1.9%) ( Table 2) . Moreover, during 2020, most of the clinical trials using hydroxychloroquine (41 trials) are expected to be completed, followed by those using plasma (20 trials), chloroquine (18 trials), lopinavir/ritonavir (12 trials), favipiravir (11 trials), and tocilizumab (10 trials) ( Figure 2 ).

Regarding the accumulated sample size by treatment, it is expected that during 2020, the estimated total number of participants to be included in trials with hydroxychloroquine is 60,366 participants, with lopinavir/ritonavir 9,938, with chloroquine 6,830, with remdesivir 5,245, with hydroxychloroquine + azithromycin 3,401, with tocilizumab 2,175, with convalescents plasma 1,588, with methylprednisolone 1,292, with favipiravir 1,250, with hydroxychloroquine + lopinavir/ritonavir 345, and only 60 with sarilumab ( Figure 2) . Additionally, the number of participants will significantly increase beyond 2021 for some treatments, such as hydroxychloroquine (93,630 participants), lopinavir/ritonavir (20,198 participants), remdesivir (13,540 participants), chloroquine (12,160 participants), sarilumab (2,140 participants) and hydroxychloroquine + azithromycin (1,376 participants). It should be noted that these figures may be underestimated because many registers did not indicate the estimated completion date, which would increase the sample size for hydroxychloroquine + lopinavir/ritonavir in 3,915 participants, for hydroxychloroquine+azithromycin in 2,252 participants, for sarilumab in 2,206, and for favipiravir in 1,107.

The world is currently facing a SARS-CoV-2 pandemic for which there are not effective therapeutic resources yet available. So much so that most hospitalized patients with COVID-19 diagnosis have had to receive off-label or compassionate use therapies, because there were not drugs approved for this disease [8, 9] . Because a resurgence of this pandemic seems likely to happen in the autumn [10] , it is urgent to test the effectiveness of new drugs, or repurposed ones, for preventing and treating the COVID-19 disease [11, 12] . In this context, this study provides an overview of the characteristics of the worldwide registered clinical trials until April 22, 2020.

The main characteristics of the registered trials are the following: more than half are aimed at recovery of clinical or respiratory parameters, are conducted in in China, Europe . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . https://doi.org/10.1101/2020.05.05.20091785 doi: medRxiv preprint and USA, participants are mostly hospitalized patients with confirmed COVID-19, have small/moderate sample size, up to 68.0% of trials have not specified the duration of the intervention in their registers, and up to 34.2% are going to end after the fall. Moreover, most trials are testing the efficacy of antimalarial drugs, particularly hydroxychloroquine (alone or in combination with azithromycin or lopinavir/ritonavir) and chloroquine, followed by the antiviral drugs lopinavir/ritonavir, favipiravir and remdesivir. Indeed, substantial interest has been placed in demonstrating the effectiveness of these specific drugs or drugs combinations [13] [14] [15] .

The challenge for which most of these clinical trials have been planned is to prevent or mitigate the devastating impact of the SARS-CoV-2 infection on the world's health while an effective vaccine emerges. And the obvious question is whether all this enormous deployment of research resources will not be a futile and disheartening effort. If we consider the cumulative sample size, the number of studies and the estimated date for completing the recruitment of participants for each of the treatment options (Figure 2 ), the current results seem to indicate that this risk is real. Thus, we can observe that by the end of June we will most likely to know whether hydroxychloroquine, chloroquine, remdesivir, or the combinations of hydroxychloroquine with azithromycin or lopinavir/ritonavir are effective to manage COVID-19 cases, because the ongoing trials will have accumulated, for each of these substances, results from more than 3000 patients, and in some cases as for hydroxychloroquine more than 15000 patients. Therefore, if the data of the registers are accurately completed, as scheduled many of the trials to assess the effectiveness of these therapeutic options planning to recruit patients after June could be cancelled or readapted.

In this context, it is worth highlighting two major recent research initiatives which, in parallel, aim to produce the same results. The Solidarity Trial [16] , sponsored by the WHO, which will involve more than hundred countries, and almost simultaneously, the European trial Discovery [17] , financed mainly through European institutions and governments. Both mega-trials are aimed to evaluate the efficacy of Remdesivir, Chloroquine or hydroxychloroquine, Lopinavir/Ritonavir and Lopinavir/Ritonavir plus Interferon beta-1 in patients hospitalized with COVID-19. Both are pragmatic trials and, because of their adaptive design, will allow incorporating new therapeutic options during their development. Although through meta-analysis techniques the results of these trials can be synthesized as if they were a unique trial, it seems thoughtful that it would have . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . https://doi.org/10.1101/2020.05.05.20091785 doi: medRxiv preprint been more efficient to coordinate these research efforts, gaining statistical power and, perhaps, to include some promising therapeutic alternatives not as extensively examined as favipiravir or tocilizumab.

Finally, the analysis of the data from the 412 clinical trials registered up to April 22, 2020 it can be summarized as follows: i) most of the trials are aimed at repurpose drugs or combinations of these drugs; ii) for all these therapeutic options the cumulative sample size seems excessive, albeit at expense of some small clinical trials with questionable quality in their registers; iii) most likely in June we will have consistent evidence about the effectiveness of the most promising therapeutic options (hydroxychloroquine, remdesivir, Lopinavir/Ritonavir), and with less certainty we will also know if favipiravir, tocilizumab, methylprednisolone, convalescent's plasma are useful to improve the clinical course of the COVID-19 disease; iv) it is unlikely that by the autumn resurgence of the illness we will have the results of the trials that are testing the efficacy of new drugs against the SARS-CoV-2 infection.

Among the limitations of this study it should be acknowledged that: i) the quality of the registers has not been evaluated due to the lack of a reliable tool for this purpose; ii) although the WHO-ICTRP is the main database of clinical trial registers and collects most national and international databases, some trials registered in non-WHO-ICTRP databases may not have been included; iii) the information collected from each clinical trial register could be heterogeneous, because each database encodes the information differently; iv) because of the lack of quality in the description of the sources of funding in many registers, we have not examined the potential relationship between therapeutic options and public or private sponsors of the trials; and v) the figures provided relating number of trials with sample size estimated to be available during 2020 may be underestimated, because many registers do not indicate the estimated end date of the trial.

In summary, regardless of whether it is financed by public or private resources, and especially in this pandemic crisis, research is a public good, and therefore has to be based on its scientific and social value, in order to produce the scientific evidence needed by the population, policy makers and, above all, clinicians [18] . For this reason, it is not justified that predictably by June more than 20 clinical trials enrolling tens of thousands of patients will finish their recruitment processes testing almost identical hypotheses, thus neglecting other promising therapeutic hypotheses. Probably, as has been recently suggested [19] , this pandemic context it is the right time to demand from academic . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . https://doi.org/10.1101/2020.05.05.20091785 doi: medRxiv preprint biomedical centres and supranational public health institutions led the coordination of the research initiatives, otherwise much of the research on the management of SARS-CoV-2 will be a waste of research resources [20] . Finally, in view of the low level of precision with which many of the database items have been recorded, a quality assessment tool for clinical trial records becomes necessary, since accuracy of these records would predict the risk of bias of the trial's results.

The authors declare that there is no conflict of interest regarding the present publication.

This study did not receive specific funding. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 9, 2020. . a The sum of the totals exceeds 100% because several trials are designed to test more than one drug and considered more than one outcome. b Improvement on one or more of the respiratory recovery indicators: signs and symptoms, oxygen therapy, mechanical ventilation or extracorporeal membrane oxygenation. c SpO2/FiO2 or blood gas/respiratory frequency/symptoms.

They Call Us and We Go

Drug Evaluation during the Covid-19 Pandemic

A randomized, controlled trial of ZMapp for Ebola virus infection

First-wave COVID-19 transmissibility and severity in China outside Hubei after control measures, and second-wave scenario planning: a modelling impact assessment

Beware of the second wave of COVID-19

Therapeutic options for the 2019 novel coronavirus (2019-nCoV)

SARS-CoV-2: Recent Reports on Antiviral Therapies Based on Lopinavir/Ritonavir, Darunavir/Umifenovir, Hydroxychloroquine, Remdesivir, Favipiravir and Other Drugs for the Treatment of the New Coronavirus

Review: Hydroxychloroquine and Chloroquine for Treatment of SARS-CoV-2 (COVID-19). Open Forum Infectious Diseases

In vitro testing of combined hydroxychloroquine and azithromycin on SARS-CoV-2 shows synergistic effect

A Trial of Lopinavir-Ritonavir in Adults Hospitalized with Severe Covid-19

Clinical Trial For COVID-19 Treatments

Trial Of Treatments For COVID-19 In Hospitalized Adults -Full Text View -Clinicaltrials.Gov

Improvement in one or more of the clinical indicators: unspecified or scored clinical improvement, laboratory results or imaging findings, time to clinical improvement, time to clinical recovery. e Body temperature, hematologic, immune, inflammation, coagulation, cardiovascular and chest imaging. f Admission to ICU, ICU discharge, length in ICU, hospital admission, discharge time, length in hospital. g Cure rate, incidence of adverse event, efficacy of treatment