EUVIMEDEUVIMED entdecken
EditorialKommentar

Repurposing Revisited: Exploring the Role of Metformin for Treatment of COVID-19

Mark J Siedner iD, Paul E Sax

Clinical Infectious Diseases · 16.08.2024 · 79 (2) · 292-294

DOI: 10.1093/cid/ciae154 ↗

Kurz gefasst

Editorial in der Zeitschrift Clinical Infectious Diseases, online erschienen am 01.05.2024. Verfasst von Mark J Siedner und 1 weiteren Autor:innen. Beteiligte Einrichtung: Medical Practice Evaluation Center and Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts, USA. Hauptthemen laut MEDLINE: Metformin, Drug Repositioning, COVID-19 Drug Treatment, SARS-CoV-2. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 25 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

(See the Major Article by Bramante et al. on pages 354–63.) Approximately 4 years into the coronavirus disease 2019 (COVID-19) pandemic, available treatment options for outpatients have a mixed track record. The most notable success is nirmatrelvir-ritonavir, which significantly reduces the risk of hospitalization and death in high-risk, unvaccinated outpatients [1]. Yet, nirmatrelvir-ritonavir is not a panacea. Whereas observational studies suggest a similar benefit in high-risk, vaccinated individuals [2, 3], a randomized study in a standard-risk population did not show improvement in either time to resolution of symptoms or the incidence of hospitalization or death [4]. Multiple drug interactions and virologic rebound [5, 6] further complicate its use. While remdesivir and molnupiravir are reasonable alternatives in select populations, they also have weaknesses. Remdesivir must be given intravenously, which is a substantial logistical challenge, and molnupiravir has both questionable efficacy in vaccinated or otherwise immune populations as well as safety concerns. This latter issue is related to the mechanism of action of molnupiravir, which leads to possible mutagenesis, limiting its use in women and men with “pregnancy potential” [7]. Meanwhile, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has developed escape mutants to monoclonal antibodies faster than these intravenous agents can be developed; although one such agent, pemivibart, has recently been made available for prevention of COVID-19 in high-risk groups, none have been available for treatment since the US Food and Drug Administration halted use of bebtelovimab in November 2022 [8]. Several other experimental agents are in development, yet the timeline for their availability is uncertain. What about already approved drugs that have been “repurposed” for treatment of COVID-19? After all, these agents have established safety profiles and regulatory approval; furthermore, they are often inexpensive and widely available. Early enthusiasm generated from in vitro and preliminary clinical data for candidate drugs has been justifiably replaced by skepticism, as several well-designed, randomized, controlled trials failed to demonstrate clinical benefits [9–12]. With more than 60 000 deaths attributed to COVID-19 in the United States in 2023 and year-round circulation of SARS-CoV-2 disrupting work, schooling, and family life, development of additional therapies that address the weaknesses of current treatments remains a critical need. The Outpatient treatment of COVID-19 and Incidence of Post-COVID-19 condition over 10 months (COVID-OUT) clinical trial was one of the many studies that dampened enthusiasm for repurposed drugs for COVID-19 [13]. In the elegant, albeit complex, parent study, overweight and obese ambulatory adults aged between 30 and 85 years with a diagnosis of COVID-19 in the past 3 days and symptoms for less than 7 days were randomized to 14 days of placebo versus metformin, fluvoxamine, or ivermectin in a 2 × 3 factorial design. Compared with placebo, none of the therapies led to improvements in the primary outcome, which was a reduction in a composite of hypoxemia, emergency room visitation, hospitalization, or death at day 14. At first glance, and in combination with the aforementioned other trials that failed to show benefits, the results of this strategy appeared to put these 3 drugs on the scrap heap of failed repurposed drugs for COVID-19—an ignominious pile already littered with hydroxychloroquine, azithromycin, and lopinavir/ritonavir, among others. Yet, a prespecified secondary analysis suggested potential clinical benefits for 1 of the treatments, metformin. When the authors focused on the more severe components of the primary end point (emergency room visits, hospitalizations, or deaths), they reported an adjusted odds ratio for metformin versus placebo of 0.58 (95% confidence interval, .35 to .94) [13]. These are arguably the most important components of the composite end point; in addition, home pulse oximetry is now known to be notoriously unreliable, particularly at lower oxygen saturations [14]. A separate secondary analysis that identified a significant reduction (6% vs 10%, P = .012) in the incidence of reported Long COVID-19 for those taking metformin versus placebo further supported the clinical benefit of metformin in COVID-OUT [15]. Of critical importance for the potentially subjective end-point of long COVID-19 diagnosis is that the data were accrued while study blinding was maintained. Bramante and colleagues now present compelling virologic evidence for a potential mechanism behind the benefits of metformin in this issue of Clinical Infectious Diseases [16]. They analyzed SARS-CoV-2 polymerase chain reaction data from nasal swabs self-collected by participants on days 1, 5, and 10 during the trial. Compared with placebo, individuals in the metformin group had significantly lower mean viral loads at days 5 and 10 and were less likely to have a detectable viral load at both time points. Results were consistent across demographic subgroups, variant, symptom duration, and baseline viral load and remained significant after adjustment for potential confounders including baseline viral load and vaccination status. By contrast, no virologic effect was seen for either ivermectin or fluoxetine. The study was not without limitations. Subanalyses of factorially randomized studies are often limited by power and can be complex to analyze if more than 1 of the intervention arms is effective or if there is interaction between them. Fortunately for the COVID-OUT trialists, the other drugs evaluated had no impact on SARS-CoV-2 decay, which helped reduce this concern. The nasal swab substudy was optional, and only approximately 75% of people in the parent study participated. That said, participants and nonparticipants were similar in both the metformin and placebo arms, mitigating the potential for selection bias or confounding due to post-randomization dropout. The absolute differences in the proportion with detectable viral load between the metformin and placebo arms at day 5 (5% difference) and day 10 (8% difference) were modest, as were the differences in mean viral reduction. Lacking from the report was data on the proportion in each arm with a viral load above a threshold associated with live viral shedding (ie, >104 copies/mL [17]), which would have augmented the potential public health implications of metformin use. Nonetheless, the benefit in viral load reduction observed with metformin over placebo in COVID-OUT, that is, approximately 0.5 log10 reduction at day 5 and 0.6 log10 reduction at day 10, was very similar to that reported for nirmatrelvir-ritonavir in the EPIC-HR study (0.8 and 0.4 log10 reductions, respectively). Finally, although they attempted to evaluate differences in viral rebound incidence, the use of only 3 time points in the study limits interpretation of those results [5]. Notwithstanding these limitations, this study makes a strong case for a potential effect of metformin on COVID-19 virologic decay and prompts reevaluation of existing data in support of its use. In vitro studies have identified both antiviral and anti-inflammatory activities of the drug, and the investigators initially identified metformin as a promising agent through use of sophisticated in silico modeling [18, 19]. Furthermore, observational studies have suggested that individuals treated with metformin for diabetes have improved COVID-19 outcomes compared with those not on this drug [20, 21]. While cross-study comparisons have limitations, it is notable that the absolute risk reduction for hospitalization or death for metformin versus placebo in COVID-OUT was nearly identical to that from a recent study that compared nirmatrelvir-ritonavir to lack of treatment in a propensity-scored matched analysis in the vaccination era (ie, 1% vs 3%) [3]. The fact that more than 50% of participants in COVID-OUT were vaccinated further amplifies the relevance of the study results to the current immunologic profile of today's population, where nearly everyone has been vaccinated, or had COVID-19, or both [22]. How can the results of the COVID-OUT study be applied to clinical practice, and should treatment guidelines recommend metformin for routine use? Points in favor of this approach are the beneficial clinical and virologic outcomes identified in this study, as well as the drug's well-known safety profile, low cost, widespread availability, and lack of drug interactions. Countering these benefits is that COVID-OUT is a single trial and the primary study end point was negative, limitations in pulse oximetry notwithstanding. Furthermore, another randomized, placebo-controlled trial of metformin initially showed no benefit [23]. However, that trial used a different dose of metformin and is currently being reanalyzed by the study investigators due to a concern about treatment allocation [24]. For clinicians faced with the dilemma of prescribing or recommending metformin, there may be a tiebreaker on the horizon. The ACTIV6 trial is currently randomizing individuals with acute COVID-19 to placebo versus metformin [25], with a planned enrollment of 3000 participants in the metformin arm, and allowing for concomitant use of other antiviral therapy. This will provide additional evidence about the potential clinical benefits of the drug and inform us whether at least 1 repurposed therapy deserves a place in COVID-19 management after all.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MetforminDrug RepositioningCOVID-19 Drug TreatmentSARS-CoV-2HumansCOVID-19Hypoglycemic AgentsAntiviral Agents

Substanzen

MetforminHypoglycemic AgentsAntiviral Agents

Autor:innen und Institutionen

  • Medical Practice Evaluation Center and Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts, USA.
  • Department of Medicine, Harvard Medical School, Boston, Massachusetts, USA.
  • Clinical Research Department, Africa Health Research Institute, KwaZulu-Natal, South Africa.
  • Division of Infectious Diseases and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts, USA.

ORCID-Profile: Mark J Siedner

Zitieren

APA

Siedner, M. J., & Sax, P. E. (2024). Repurposing Revisited: Exploring the Role of Metformin for Treatment of COVID-19. Clinical Infectious Diseases, 79(2), 292-294. https://doi.org/10.1093/cid/ciae154

Vancouver

Siedner MJ, Sax PE. Repurposing Revisited: Exploring the Role of Metformin for Treatment of COVID-19. Clinical Infectious Diseases. 2024;79(2):292-294. doi:10.1093/cid/ciae154

BibTeX
@article{Siedner2024,
  author = {Siedner, Mark J and Sax, Paul E},
  title = {Repurposing Revisited: Exploring the Role of Metformin for Treatment of COVID-19},
  journal = {Clinical Infectious Diseases},
  year = {2024},
  volume = {79},
  number = {2},
  pages = {292--294},
  doi = {10.1093/cid/ciae154}
}
Kostenlos starten →