key: cord-286525-0354438s
authors: Lee, Todd C.; Butler-Laporte, Guillaume; Chagla, Zain; McDonald, Emily G.
title: Tocilizumab versus the covid19 tempest: all’s well that ends well or much ado about nothing?
date: 2020-09-29
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2020.09.033
sha: 
doc_id: 286525
cord_uid: 0354438s

nan

Nearly nine months have passed since the first mention of SARS-CoV-2, initially described as a mysterious respiratory illness [1] that has gone on to catch the world offguard. Since then we have learned a great deal about COVID-19 and made important strides in patient care including pivoting away from pre-emptive intubation, which may have inflated early mortality [2] . However, many questions remain unanswered. Since the outset, a severe phenotype of the disease associated with elevated markers of inflammation and culminating in substantial lung injury and death has been recognized to develop in a subset of patients [3] . Studies of potential treatments aimed at prevention and treatment of this phenotype have focused on medications with antiviral, antithrombotic, and/or anti-inflammatory properties. One of these treatments, dexamethasone, has already demonstrated a mortality benefit in a large randomized controlled trial [4] . Analogies with cytokine storm, which may or may not be an apt comparison [5] , have led to an interest in IL-6 inhibition as a therapeutic modality for severe COVID-19. Tocilizumab is a biological IL-6 inhibitor therapy that is used for the treatment of rheumatoid arthritis and giant cell arteritis. Indeed, several observational cohorts have explored the repurposing of tocilizumab for COVID-19 [6] [Also Cite

Martinez-Sanz et al., CMI when in press] and there are, at present, dozens of registered clinical trials involving tocilizumab or sarilumab.

Rodríguez-Baño et. al [7] tackle the management of patients with an inflammatory presentation by providing observational data in support of tocilizumab for select patients with COVID-19. The authors conducted a retrospective cohort study in patients admitted to sixty hospitals in Spain between February 2 and March 31, 2020. Patients selected for inclusion in this analysis were clinically ill with both fever and oxygen requirements at study enrollment but without immediate need for mechanical ventilation. Most importantly, they selected patients with signs of a "hyperinflammatory" response, which the authors defined by presence of an elevation in ferritin, D-dimer, and/or IL-6 levels.

After adjustment for several potential confounders, using a variety of techniques, the authors found that the receipt of tocilizumab without corticosteroids was associated with J o u r n a l P r e -p r o o f reduced hazard for intubation or death. By contrast, neither steroid therapy alone nor in combination with tocilizumab was convincingly protective for either outcome. Of note, a second observational study from Spain [Cite Martinez-Sanz et al. when it is in press] was subsequently published in Clinical Microbiology and Infection and reported a decreased risk of death when tocilizumab was given to patients with high levels of Creactive protein (>150mg/L).

Rodriguez-Baño et al appropriately adjusted for several comorbidities and used propensity score methods. Prior to propensity score matching, the treatment groups differed, with a higher prevalence of cardiac disease and severe renal insufficiency in the untreated group. Comparatively, the tocilizumab group also had a longer median duration of illness and hospitalization prior to enrolment as well as lower adverse negative prognostic markers (ferritin and D-dimer). Even after propensity score matching, some of these imbalances persisted. The major challenge due to the study design is that we will never know with certainty why in some cases a given therapy was chosen over no therapy at all. This confounding by indication [8] is extremely difficult to eliminate. Likewise, it is hard to ascertain through observational data alone whether patients who received tocilizumab monotherapy were systematically treated differently than those who did not.

Additionally, the authors attempted to address immortal time bias by excluding early primary outcomes and performing a sensitivity analysis with a time dependent covariate.

Yet, there is likely residual survivor bias as the fundamental definition of day 0 did not correspond to the day of admission, rather to the time of enrollment in the cohort. As such, the tocilizumab group was further along in the illness trajectory with a median of 

One of the unexpected findings from this study was an observed lack of benefit from steroids in contrast to the RECOVERY randomized controlled trial results [4] or a recent meta-analysis of steroid trials in critically ill patients. [9] There are several ways this can be reconciled. Perhaps like in the tale of "Goldilocks", the "porridge was too hot": many patients received 10x the dose of steroids used in the RECOVERY trial and less than 10% of the cohort was treated with dexamethasone. While it is tantalizing to assume that the steroid effect is a class effect, dexamethasone may work differently with its lack of mineralocorticoid activity and longer half-life. Indeed, a recent randomized controlled trial of methylprednisolone at higher equivalent doses to RECOVERY's dexamethasone failed to demonstrate a mortality benefit [10] and methylprednisolone appeared to have the least benefit in the meta-analysis [9] . Finally, the patients in the SAM-COVID-19 cohort differed from those in RECOVERY with a lower prevalence of heart and kidney disease and fewer mechanically ventilated patients (1-3% vs. 15%) who seem to benefit most from steroids. Finally, we must always consider the possibility of residual confounding by indication and/or other differences in the care received by the steroid group in the present trial.

It is also important to put this observational data in context with several industrysponsored trials that have examined IL-6 inhibiting monoclonal antibodies with disappointing conclusions. The phase III COVACTA trial (NCT04317092) found tocilizumab did not reduce mortality in hospitalized patients with severe COVID-19 pneumonia [11] . Similarly a large trial of sarilumab (NCT04315298) in severe and critical COVID-19 was stopped by the data safety monitoring board due to lack of benefit and a potential signal for harm in non-ventilated patients [12] and a second international trial (NCT04327388) also failed to meet its primary or key secondary outcomes [13] . Whether or not a specific subgroup of patients with a hyperinflammatory response might benefit as proposed by the authors remains to be seen in future (ideally randomized) trials.

J o u r n a l P r e -p r o o f Implications Early rapid mobilization of research efforts allowed the investigators to collect valuable data through an observational cohort study and important attempts were made to control for confounding by indication and survivor bias in the analyses. The authors appropriately conclude with a proposal to investigate the role of IL-6 inhibition as dexamethasone emerges as standard of care, and they caution against widespread use based on observational data alone. Reconciling their results with those from randomized control trials raises important questions about the causal effect of the hyperinflammatory response and its role in the development of severe COVID-19. Given the promising findings associated with the use of corticosteroids, we speculate whether a broader anti-inflammatory approach is the best option for most patients, whereas only a subset of patients might benefit from targeted anti-inflammatory management. We hope that trials like RECOVERY (www.recoverytrial.net) and REMAP-CAP (www.remapcap.org) will soon bring substantial clarity to the role, if any, that drugs like tocilizumab might play in combating the worldwide COVID-19 tempest.

Pneumonia of unknown aetiology in Wuhan, China: potential for international spread via commercial air travel

Ventilation Techniques and Risk for Transmission of Coronavirus Disease, Including COVID-19: A Living Systematic Review of Multiple Streams of Evidence

The role of IL-6 and other mediators in the cytokine storm associated with SARS-CoV-2 infection

Dexamethasone in Hospitalized Patients with Covid-19 -Preliminary Report

Is a "Cytokine Storm" Relevant to COVID-19?

Efficacy of Tocilizumab in COVID-19: A Systematic review and Meta-Analysis

Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases

Confounding by Indication in Clinical Research

Association Between Administration of Systemic Corticosteroids and Mortality Among Critically Ill Patients With COVID-19: A Meta-analysis

Methylprednisolone as Adjunctive Therapy for Patients Hospitalized With COVID-19 (Metcovid): A Randomised, Double-Blind, Phase IIb, Placebo-Controlled Trial

Roche provides an update on the phase III COVACTA trial of Actemra/RoActemra in hospitalised patients with severe COVID-19 associated pneumonia

Sanofi and Regeneron provide update on Kevzara® (sarilumab) Phase 3 U.S. trial in COVID-19 patients

Sanofi provides update on Kevzara® (sarilumab) Phase 3 trial in severe and critically ill COVID-19 patients outside the

Drs. Lee and McDonald receive research salary support from the Fonds de Recherche Québec -Santé. Dr Butler-Laporte is supported by a scholarship from the Fonds de