A systematic review and meta-analysis, investigating dose and time of fluvoxamine treatment efficacy for COVID-19 clinical deterioration, death, and Long-COVID complications

Abstract

There have been 774,075,242 cases of COVID-19 and 7,012,986 deaths worldwide as of January 2024. In the early stages of the pandemic, there was an urgent need to reduce the severity of the disease and prevent the need for hospitalization to avoid stress on healthcare systems worldwide. The repurposing of drugs to prevent clinical deterioration of COVID-19 patients was trialed in many studies using many different drugs. Fluvoxamine (an SSRI and sigma-1 receptor agonist) was initially identified to potentially provide beneficial effects in COVID-19-infected patients, preventing clinical deterioration and the need for hospitalization. Fourteen clinical studies have been carried out to date, with seven of those being randomized placebo-controlled studies. This systematic review and meta-analysis covers the literature from the outbreak of SARS-CoV-2 in late 2019 until January 2024. Search terms related to fluvoxamine, such as its trade names and chemical names, along with words related to COVID-19, such as SARS-CoV-2 and coronavirus, were used in literature databases including PubMed, Google Scholar, Scopus, and the ClinicalTrials.gov database from NIH, to identify the trials used in the subsequent analysis. Clinical deterioration and death data were extracted from these studies where available and used in the meta-analysis. A total of 7153 patients were studied across 14 studies (both open-label and double-blind placebo-controlled). 681 out of 3553 (19.17%) in the standard care group and 255 out of 3600 (7.08%) in the fluvoxamine-treated group experienced clinical deterioration. The estimated average log odds ratio was 1.087 (95% CI 0.200 to 1.973), which differed significantly from zero (z = 2.402, p = 0.016). The seven placebo-controlled studies resulted in a log odds ratio of 0.359 (95% CI 0.1111 to 0.5294), which differed significantly from zero (z = 3.103, p = 0.002). The results of this study identified fluvoxamine as effective in preventing clinical deterioration, and subgrouping analysis suggests that earlier treatment with a dose of 200 mg or above provides the best outcomes. We hope the outcomes of this study can help design future studies into respiratory viral infections and potentially improve clinical outcomes.

Keywords: (E)‐5‐methoxy‐1‐[4‐(trifluoromethyl)phenyl]pentan‐1‐one O‐2‐aminoethyl oxime); Antidepressant; Coronavirus; Drug repurposing; Pandemic; SARS-CoV-2; Sigma-1 receptor (σ1R).

Figure 1

Study Selection. Studies were selected from various literature databases, identifying clinical trials with fluvoxamine for COVID-19. Studies were split into all studies and randomized placebo-controlled studies. *Oskotsky et al., retrospective trial included all SSRI drugs and combined fluvoxamine with fluoxetine and, therefore, was excluded from the meta-analysis. § Studies with no deaths in either group (fluvoxamine or control) were not entered into the COVID-19 mortality meta-analysis. †Only one placebo-controlled study of fluvoxamine in COVID-19 had deaths in either group (fluvoxamine or control), so the meta-analysis could not be carried out.

Figure 2

Meta-analysis for every study regarding fluvoxamine treatment of COVID-19 (A) Forrest plot. The Log OR was 1.087 (95% CI 0.200 to 1.973) z = 2.402, p = 0.016. (B) Funnel Plot. Rank correlation (p = 0.1925) and regression test (p = 0.9398) do not indicate any funnel plot asymmetry.

Figure 3

Meta-analysis for the placebo-controlled studies on fluvoxamine treatment of COVID-19 and clinical deterioration. (A) Forrest plot. LogOR 0.359 (95% CI 0.1111 to 0.5294), which differs significantly from zero (z = 3.103, p = 0.002) (B) Funnel Plot. Rank correlation (p = 0.5619) and the regression test p = 0.4018 do not indicate any asymmetry.

Figure 4

Meta-analysis for all the studies carried out regarding fluvoxamine treatment and COVID-19 mortality. (A) Forrest plot. LogOR 1.502 (95% CI 0.621 to 2.391) z = 3.301, p < 0.001. (B) Funnel Plot. The rank correlation p = 0.0863 and the regression test indicated p = 0.0629 did not reveal any plot asymmetry.

Figure 5

Meta-analysis for all the studies on fluvoxamine treatment of COVID-19 and clinical deterioration sub-grouped by dose. (A) Forrest plot. High dose LogOR 0.6957 (95% CI 0.3777 to 1.0137) z = 4.2882, p < 0.0001. Low dose LogOR 1.198 (95% CI -0.930 to 3.325) z = 1.082, p = 0.279. (B) Funnel Plot. No Asymmetry was observed at either high or low dose grouping.

Figure 6

Meta-analysis of sub-grouped Fluvoxamine/COVID-19 studies based on dose, up to 100, up to 200, and up to 300 mg/day. (A) Forrest plot. 100 mg/day LogOR 1.43 (95% CI -1.16 to 4.01) z = 1.082, p = 0.279. 200 mg/day LogOR 0.550 (95% CI 0.236 to 0.865) z = 3.433 p < 0.001. 300 mg/day LogOR 1.153 (95% CI 0.593 to 1.713) z = 4.035 p < 0.001 (B) Funnel plots. No Asymmetry was observed at dose grouping.

Figure 7

Meta-analysis for all the studies on fluvoxamine treatment of COVID-19 and risk of mortality sub-grouped by dose (High and Low). (A) Forrest plot. High dose LogOR 11.593 (95% CI 0.530 to 2.656) z = 2.938, p = 0.003. Low dose LogOR 1.420 (95% CI -1.802 to 4.462) z = 0.864, p = 0.338. (B) Funnel Plots. No Asymmetry was observed in either dose grouping.

Figure 8

Meta-analysis for all the studies on fluvoxamine treatment of COVID-19 clinical deterioration sub-grouped by time to treatment (early, medium, or late). (A) Forrest plots. Early treatment LogOR 2.441 (95% CI 0.078 to 4.805) z = 2.025, p = 0.043. Medium treatment LogOR 0.453 (95% CI 0.130 to 0.776) z = 2.748, p = 0.006. Late treatment LogOR 0.869 (95% CI 0.09 to 1.718) z = 2.004, p = 0.045 (B) funnel Plots. No asymmetry was observed at any time point in grouping.

Figure 9

Meta-analysis for fluvoxamine sub-grouped by time to treatment (early, medium, or late) in studies reporting COVID-19 mortality as an endpoint. (A) Forrest Plot B Funnel plots. (B) Funnel Plot. No Asymmetry was observed at time point grouping.