Treatments Administered to the First 9152 Reported Cases of COVID-19: A Systematic Review

Abstract

The emergence of SARS-CoV-2/2019 novel coronavirus (COVID-19) has created a global pandemic with no approved treatments or vaccines. Many treatments have already been administered to COVID-19 patients but have not been systematically evaluated. We performed a systematic literature review to identify all treatments reported to be administered to COVID-19 patients and to assess time to clinically meaningful response for treatments with sufficient data. We searched PubMed, BioRxiv, MedRxiv, and ChinaXiv for articles reporting treatments for COVID-19 patients published between 1 December 2019 and 27 March 2020. Data were analyzed descriptively. Of the 2706 articles identified, 155 studies met the inclusion criteria, comprising 9152 patients. The cohort was 45.4% female and 98.3% hospitalized, and mean (SD) age was 44.4 years (SD 21.0). The most frequently administered drug classes were antivirals, antibiotics, and corticosteroids, and of the 115 reported drugs, the most frequently administered was combination lopinavir/ritonavir, which was associated with a time to clinically meaningful response (complete symptom resolution or hospital discharge) of 11.7 (1.09) days. There were insufficient data to compare across treatments. Many treatments have been administered to the first 9152 reported cases of COVID-19. These data serve as the basis for an open-source registry of all reported treatments given to COVID-19 patients at www.CDCN.org/CORONA . Further work is needed to prioritize drugs for investigation in well-controlled clinical trials and treatment protocols.

Keywords: COVID-19; Drug repurposing; SARS-CoV-2; Systematic literature review.

Fig. 1

PRISMA study selection

Fig. 2

Most frequently administered therapeutic classes of drugs. Most frequently administered therapeutic classes of drugs according to the Anatomical Therapeutic Classification system (a), and most frequently administered individual drugs reportedly used among the 9152 patients (b)

Fig. 3

Mean weighted time to clinically meaningful response associated with the most frequently used drugs. Clinically meaningful response was calculated as the duration from drug start until the earliest of author-reported resolution of symptoms or hospital discharge. Data on drugs with response times available from at least ten studies are included. A mean and standard deviation was imputed from summary-level reports with median (interquartile range), and a weighted mean was calculated from available data. Drugs may have been used alone or concurrently with others. TCMR observations were available for < 15 patients for oseltamivir, methylprednisolone, moxifloxacin, and umifenovir

Fig. 4

Potential mechanisms of treatments used to date against COVID-19. Treatments given to patients in this study may be acting through one or more of the following mechanisms or they may also have had no effect: (1) limiting entry into ciliated bronchial epithelial cells (N-acetylcysteine, heparin, meplazumab, umifenovir, hydroxychloroquine), (2) inhibiting viral replication (interferon-α/β, ritonavir/lopinavir, oseltamivir, ganciclovir, ribavirin, favipiravir, remdesivir, danoprevir), (3) preventing viral dissemination via antibody-mediated neutralization by increasing SARS-CoV-2-specific antibodies (convalescent plasma) or non-specific antibodies (IVIg, thymopentin), (4) strengthening a weakened immune response with immunostimulants (interferon-α/β, thymosin-α-1), (5) preventing a hyper-immune response with immunosuppressants (corticosteroids, hydroxychloroquine, IVIg), or (6) controlling a hyper-immune response (corticosteroids, tocilizumab). Antibiotics such as azithromycin, moxifloxacin, ceftriaxone, and cefoperazone were used more than 25 times but the potential mechanism of action is unknown. These drugs may be acting by preventing secondary infections, controlling inflammation, modulating the microbiome, or directly having an anti-viral effect; they may also have had no effect. A number of treatments also administered in this study support vital organ function, such as bacillis lichenformis (gastrointestinal), antacids (gastrointestinal), continuous renal replacement therapy (kidneys), vasopressors/vasocontrictive agents (cardiovascular), and expectorant and antitussive agents (respiratory). IVIg intravenous immunoglobulin. Dotted lines represent potential secondary mechanisms of action