The TOGETHER Trial Evaluates Fluvoxamine for COVID-19
“In those years, truth was elusive – as was my own faith that I could recognize and contain it.”
Previously, the very basic underpinnings of fluvoxamine as a pharmacological method to quell cytokine unruliness was outlined. In brief, fluvoxamine is a sigma-1-receptor [S1R] agonist – an effect that opposes inositol-requiring enzyme 1 alpha [IRE1]. Disruption of IRE1 activity has an anti-inflammatory-like effect because IRE1 is normally stimulated by pathogen fragments [e.g., lipopolysaccharide, viral motifs, etc.] and splices into activity x-box binding protein [XBP]. XBP, in turn, triggers the expression of tumour necrosis factor alpha [TNF-a], interleukin-6 [IL-6], and the like.
Prior to the COVID-19 pandemic, data supporting the aforementioned – including a murine model of gram negative endotoxemia – had accumulated and prompted the use of fluvoxamine against the SARS-CoV-2 virus. Accordingly, a unique, randomized, double-blinded controlled trial in outpatients with COVID-19 was published, followed by an observational study of a COVID-19 outbreak in congregate living. Both showed promising results in the outpatient setting prior to vaccine availability. Arguably, the clinical end-points enhanced by fluvoxamine were more robust than data supporting other, more expensive therapeutics championed for outpatient COVID-19.
In an effort to replicate the findings of Lenze et al. and Seftel and Boulware, the TOGETHER trial added fluvoxamine to its protocol. The TOGETHER trial is a well-regarded, randomised, adaptive platform trial investigating repurposed treatments for COVID-19 among high-risk, adult, outpatients. To date, TOGETHER has studied hydroxychloroquine, lopinavir–ritonavir, metformin, ivermectin, fluvoxamine, doxasozin, and pegylated interferon lambda versus matching placebos.
What They Did
The trial enrolled outpatients from 11 clinical sites in Brazil. Patients over 18 years of age, with no more than 7 days of symptoms, with an acute clinical condition consistent with COVID-19 and/or with a rapid test positive for SARS-CoV-2 were enrolled if they had at least one other criterion for high-risk disease [e.g., at least 50 years of age, diabetes, hypertension, chronic cardiopulmonary disease, stage IV chronic kidney disease, immunosuppression, cancer history within 6 months and/or unvaccinated].
Patients were excluded from study if any of the following criteria were met: a negative diagnostic examination for SARS-CoV-2, vaccinated status, dyspnea secondary to another cause, uncontrolled psychiatric disorders, unwillingness to follow research guidelines and procedures, or pregnancy.
All enrolled patients underwent a rapid antigen test for the SARS-CoV-2 virus, if negative, the patient was not included in the study. Patients were randomized via text message to a coordinating pharmacist, maintaining concealed allocation. The patient, trial team, and site staff were blinded to treatment allocation.
Patients randomly received either active treatment [i.e., fluvoxamine at 100 mg twice daily for 10 days] or identical placebo and were closely followed for 28 days.
The primary outcome was a composite end-point consisting of either observation for at least 6 hours for COVID-19 in an emergency setting or referral for hospitalization both within 28 days of randomization.
Secondary outcomes included: viral clearance, time to clinical improvement, number of days with respiratory symptoms, time to hospitalisation for any cause or due to COVID-19 progression, all-cause mortality and time to death from any causes, WHO clinical worsening scale score, days in hospital and on ventilator and adverse events and adverse reactions to the study medications.
The data was evaluated by intention-to-treat, modified intention-to-treat [i.e., the patient had an outcome after at least 24 hours of therapy] and per-protocol [i.e., the patient completed at least 80% of the randomized therapy] analyses.
What They Found
1497 recruited participants were randomly assigned to fluvoxamine [n=741] or placebo [n=756]. In the fluvoxamine group 79 [11%] had a primary outcome event compared with 119 [16%]; most events [87%] were hospitalizations. This was a statistically and clinically significant difference.
There were no major differences in the secondary outcomes, including viral clearance. By intention-to-treat, 28-day mortality was equivalent, however, based on per-protocol analysis [i.e., taking at least 80% of randomized therapy], mortality was reduced by roughly 2% in those who received fluvoxamine.
No doubt the TOGETHER evaluation of fluvoxamine is a much larger trial than the initial investigation by Lenze et al. In the earlier study, 152 patients were randomized and the primary end-point occurred in 0 of 80 patients in the fluvoxamine group and in 6 of 72 [8.3%] patients in the placebo group. Recall that the primary end-point was different, however, requiring both:
1.] presence of dyspnea or hospitalization for dyspnea or pneumonia and
2.] O2 saturation < 92% on room air or supplemental oxygen to maintain O2 saturation of > 91%.
We are not specifically told the burden of hypoxemia in the TOGETHER study, though the authors note that:
“Considering the transmissible characteristics of SARS-CoV-2 and the isolation recommendations of positive individuals, we collected few vital sign data.”
One wonders if the end-point included oxygen saturation criteria if the outcome would have changed? Within the secondary analyses, ‘hospitalization’ had a clinically, but not statistically-significant reduction in those who received fluvoxamine while treatment in an ‘emergency setting’ for at least 6 hours was statistically significantly reduced in those who were randomized to treatment. This might suggest that hypoxemia was not attenuated by fluvoxamine [i.e., assuming that supplemental oxygen would require hospital admission], but healthcare in Brazil at the time was significantly stressed. The authors note:
“Because many patients who would ordinarily have been hospitalised were prevented from admission due to hospital over-capacity during peak waves, the composite endpoint addresses both hospitalisation and a proxy for hospitalisation, retention in a COVID-19 emergency hospital setting. This region of Brazil implemented hospital-like services in the emergency settings with 50–80 bed units providing services including multiday stays, oxygenation, and mechanical ventilation.”
Thus, we presume that retention in one of these emergency care settings could have involved rather intensive treatment and that fluvoxamine diminished this risk significantly. Nevertheless, and somewhat buried in subgroup analysis were values that caught my eye. The authors pre-specified patients who received corticosteroid therapy as a subgroup of interest. Presumably, steroids are a marker of more severe disease, that is, requiring supplemental oxygen therapy. Overall, very few patients received corticosteroids, 8 total in the placebo group [i.e., ~ 1%] and 4 total in the fluvoxamine group [i.e., ~ 0.05%], numbers that are too small to make definitive conclusions but, nonetheless, hint at a study with relative euboxia in both arms.
Overall, the number needed to treat was 20 to prevent the composite end-point, which is clinically very significant and when the independent data safety monitoring committee met on Aug 5, 2021, it recommended that the TOGETHER trial stop randomly assigning patients to the fluvoxamine group, as this comparison had met the pre-specified superiority criterion for the primary endpoint.
Finally, and as mentioned by the accompanying editorial, questions remain, including the effects of fluvoxamine on individual outcomes such as mortality as well as whether being vaccinated has a clinical interaction with fluvoxamine. As well, did the difference in fluvoxamine dosing play any role [i.e., 100 mg BID for 10 days in TOGETHER versus a quick titration to 100 mg TID for 15 days in the original study]? Perhaps some of these questions will be enlightened by the forthcoming STOP-COVID-2 trial [NCT04668950]?