By Vibhu Sharma, MD, MS

Associate Professor of Medicine, University of Colorado, Denver

SYNOPSIS: These two trials had disparate findings with respect to interleukin-6 inhibition, with REMAP-CAP showing a benefit and COVACTA showing none.

SOURCES: The REMAP-CAP Investigators; Gordon AC, Mouncey PR, Al-Beidh F, et al. Interleukin-6 receptor antagonists in critically ill patients with Covid-19. N Engl J Med 2021;384:1491-1502.

Rosas IO, Brau N, Waters M, et al. Tocilizumab in hospitalized patients with severe Covid-19 pneumonia. N Engl J Med 2021;384:

The REMAP-CAP (Randomized Embedded Multifactorial Adaptive Platform Trial for Community-Acquired Pneumonia) trial randomized patients with severe COVID-19 pneumonia to tocilizumab, sarilumab, or placebo. This was the immune modulation therapy domain among multiple other domains in the study, including corticosteroid, antibiotic, and macrolide domains. Critically ill patients requiring organ support were enrolled. Organ support was defined as respiratory organ support in the form of heated high flow nasal cannula (HHFNC) oxygen > 30 L/minute, noninvasive ventilation (NIV), or invasive mechanical ventilation, or cardiac organ support in the form of vasopressors. Patients had to be enrolled within 24 hours of starting organ support in a critical care setting. Tocilizumab dosing could be repeated once within 12-24 hours at the discretion of the treating physician. Sarilumab was administered once at a dose of 400 mg. Overall, 895 patients underwent randomization at 113 sites across six countries. Approximately one-third of patients in each group were receiving HHFNC, with the rest on either NIV or invasive mechanical ventilation. Approximately one-quarter of patients in each group received vasopressors, except approximately 10% in those randomized to sarilumab.

REMAP-CAP used a Bayesian design with no sample size calculated; in this clinical trial type, interim analyses occur as randomization continues. A Markov chain Monte-Carlo simulation calculates odds of the primary outcome occurring — in this case, the odds of survival and the number of organ failure-free days up to day 21 as predicted by treatment group (tocilizumab, sarilumab, or placebo). A cumulative logistic model is fit to project possible outcomes using patient outcomes of those previously recruited (based on two-week epochs) and refining the model as recruitment continued. This technique has been used to predict stock prices using historical price data. Simplistically, an odds ratio of greater than one generated by the model implies improved survival or organ failure-free days. The trial was to conclude with a superiority conclusion if the posterior probability was greater than 99% that an intervention was more effective than the others or an inferiority conclusion when the posterior probability was less than 0.25% that a given intervention was better than the others. A calculated posterior probability of 99% that the odds ratio for the primary outcome was greater than one implied intervention efficacy compared to placebo. The intervention was deemed futile if the posterior probability that the odds ratio was greater than 1.2 was less than 5%, and equivalent if the posterior probability that the odds ratio was between 0.8 and 1.2 was greater than 90%.


The median number of organ support-free days was 10 (interquartile range [IQR], -1 to 16 days) in the tocilizumab group, 10 (IQR 0-16 days) in the sarilumab group, and 0 (IQR, -1 to 15 days) in the placebo group. The mean adjusted odds for organ failure-free survival were 1.65 (95% credible interval [CrI], 1.25 to 2.14) for tocilizumab and 1.76 (95% CrI, 1.17 to 2.91) for sarilumab compared to control, with posterior probabilities of > 99% for both, implying efficacy. Similarly, the mean adjusted odds for in-hospital survival were 1.64 (95% CrI, 1.14 to 2.35) for tocilizumab and 2.01 (95% CrI, 1.18 to 4.71) for sarilumab compared to placebo, implying efficacy for the mortality outcome.

The second trial (COVACTA) randomized 438 patients with COVID-19 acute respiratory distress syndrome (ARDS) (SaO2 < 93% or a PaO2/FiO2 < 300) to intravenous tocilizumab or placebo. The primary outcome was clinical status at day 28 on an ordinal scale: 1) discharged alive or ready to discharge; 2) hospitalization on a medical floor without oxygen; 3) medical floor with oxygen; 4) noninvasive ventilation or heated high flow oxygen; 5) intubation/mechanical ventilation; 6) extracorporeal membrane oxygenation (ECMO) or mechanical ventilation plus other life support; and 7) death. The baseline score on the ordinal scale was the score immediately prior to randomization. Secondary outcomes included clinical status at day 14 on the ordinal scale, day 28 mortality, number of ventilator-free days by day 28, the time to improvement from baseline by at least two categories on the ordinal scale, and the time to hospital discharge or readiness for discharge. This trial found no difference in clinical status between the tocilizumab and placebo groups at day 28; however, the trial did find a “possible benefit for tocilizumab in the time until hospital discharge (or readiness for discharge) and in the duration of ICU stay, both of which require additional study.”


The REMAP-CAP study found a mortality benefit to IL-6 antagonism with odds that are not impressive. Although the statistical technique used was robust, the odds ratios generated suggest a marginal benefit at best. Importantly, in this trial, patients received the first dose of the intervention drug (or placebo) within approximately a day of admission and within 13 hours of admission to the intensive care unit. PaO2/FiO2 ratios were approximately 100 at enrollment, implying moderate to severe ARDS. With respect to COVACTA, the mortality rate was identical in the tocilizumab and placebo groups; however, there was a signal toward a shorter length of hospital stay with tocilizumab. Approximately 40% of patients were mechanically ventilated and of these patients, the median time to initiation of the intervention drug was three days, with a range of 0-28 days, implying (possibly) that initiation was too delayed for any meaningful benefit in some patients. Patients were recruited a mean of 12 days from the onset of COVID symptoms, but the range was wide (up to 50 days for some), at which point, given the natural history of the disease, a benefit from immune modulation is unlikely. In the COVACTA trial, more patients in the placebo group received steroid therapy (29%) compared to the tocilizumab group (19%). It is clear that mortality is reduced by steroid therapy, and it is possible that the survival rates in the placebo group may have been inflated as a result of more patients receiving steroids. In addition, more Black and Native American patients received placebo purely as a result of the chance of randomization.

The RECOVERY trial was a much larger trial than COVACTA, enrolling more than 4,000 patients.1 This trial found a survival benefit for tocilizumab regardless of the amount of respiratory support among hospitalized patients with hypoxia (oxygen saturation < 92% on room air or requiring oxygen therapy) and evidence of systemic inflammation (C-reactive protein [CRP] ≥ 75 mg/L). Importantly, in this trial, positive outcomes were seen for the nine patients not receiving any oxygen and 1,859 patients receiving low-flow oxygen randomized to the tocilizumab group. The benefits were sustained for patients receiving corticosteroids and other subgroups as well (i.e., receiving noninvasive mechanical ventilation, invasive ventilation, or ECMO). A smaller trial assessed the use of tocilizumab in non-critically ill patients and found a smaller chance of reaching the primary endpoint (a composite of the percentage of patients who died, received invasive ventilation, or received noninvasive ventilation) among those with CRP levels > 150 mg/L.2 These patients with elevated CRP levels also were more likely to be alive at 90 days. Both REMAP-CAP and COVACTA excluded patients with active infection, tuberculosis, or those with imminent death. There were no excess infections in patients receiving IL-6 antagonism, suggesting that in the appropriately selected patient, IL-6 antagonism is safe. The current state of evidence suggests benefit to patients not yet critically ill with elevated inflammatory markers (CRP 75 mg/L to 150 mg/L). Although some patients may benefit later in the disease course, the key may be initiating IL-6 antagonism early in the course of disease in appropriately selected patients. 


  1. RECOVERY Collaborative Group. Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): A randomized, controlled, open-label, platform trial. Lancet 2021;397:1637-1645.
  2. Mariette X, Hermine O, Tharaux PL, et al. Effectiveness of tocilizumab in patients hospitalized with COVID-19: A follow-up of the CORIMUNO-TOCI-1 randomized clinical trial. JAMA Intern Med 2021; May 24:e212209. doi: 10.1001/jamainternmed.2021.2209. [Online ahead of print].