Can tocilizumab calm the cytokine storm of COVID-19?

The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has created a global pandemic, with ongoing regional COVID-19 outbreaks across the world. For many patients, COVID-19 manifests simply as a viral respiratory syndrome, but a subset of patients develops a life-threatening course with significant and prolonged systemic inflammation.¹ Increasing data show that many severe COVID-19 cases have features of cytokine storm syndrome, including characteristics of acute respiratory distress syndrome and macrophage activation syndrome, such as hyperferritinaemia, systemic hyperinflammation, and multi-organ system dysfunction.² Indeed, several early reports highlighted that patients with severe disease had clinical and laboratory features of hyperinflammation and cytokine storm, including elevated levels of interleukin-6 (IL-6).3, 4 Based on these observations and the encouraging results of an early case series using tocilizumab (a recombinant humanised monoclonal antibody against the IL-6 receptor),⁵ clinicians facing rising numbers of patients with COVID-19 have increasingly turned to cytokine-directed therapies for critically ill patients.

In The Lancet Rheumatology, Giovanni Guaraldi and colleagues⁶ report the results of the TESEO study, a large, multicentre, retrospective cohort study of tocilizumab for severe COVID-19 pneumonia. The study included 544 patients with confirmed COVID-19 and severe respiratory symptoms (defined as tachypnea, hypoxemia, poor oxygenation, and lung infiltrates of more than 50%) at three centres in the Emilia-Romagna region of Italy. 179 (33%) of these patients received tocilizumab (intravenous or subcutaneous) and standard of care therapy, and 365 (67%) patients with similar respiratory symptoms received only standard of care therapy (supplemental oxygen, hydroxychloroquine, azithromycin, combination antiretrovirals, and low molecular weight heparin). The primary outcome was a composite of death or progression to invasive mechanical ventilation. At day 14 from hospital admission, 22·6% (95% CI 16·2–29·0) of tocilizumab patients had the primary outcome, compared with 36·5% (30·7–42·2) of patients receiving standard of care. 13 (7%) of 179 patients treated with tocilizumab died, compared with 73 (20%) of 365 who received standard of care. When adjusted for age, gender, recruiting centre, duration of symptoms, and Subsequent Organ Failure Assessment (SOFA) score, tocilizumab treatment showed a hazard ratio for death alone of 0·38 (95% CI 0·17–0·83; p=0·015). Similar numbers of patients received intravenous and subcutaneous tocilizumab, on the basis of local availability, and no significant difference in the outcomes or side-effect profile were found according to the route of administration. Regarding safety, more secondary infections were diagnosed in the tocilizumab group (24 [13%]) than in the standard of care group (14 [4%]), which has also been seen in randomised trials of tocilizumab.⁷ The present study included one patient receiving tocilizumab who died from acute liver failure due to herpes simplex virus 1 reactivation.

This study is the largest of its kind reported thus far and represents a crucial addition to the literature regarding tocilizumab in COVID-19. It is also an impressively rigorous effort undertaken at the height of the pandemic in northern Italy, with patients enrolled in a systematic manner with informed consent, standardised data collection, and predefined study outcomes. Dosing was also standardised at either 8 mg/kg (up to 800 mg) administered twice intravenously, or 162 mg administered subcutaneously in two simultaneous doses (81 mg in each thigh). These doses were based on pharmacokinetic data and were intended to mimic peak plasma concentration. Finally, patients receiving tocilizumab were compared with a contemporaneous cohort of controls with the same inclusion and exclusion criteria. The primary limitation of this study, and all cohort studies, is that the patients and controls were not randomly chosen, thus introducing both measurable and unknown potential biases. Patients were offered tocilizumab treatment mainly on the basis of availability of the drug, but potentially also because of both demographic and disease-specific factors. Indeed, the patients treated with tocilizumab were younger, and in Modena (where more granular clinical data was available), they were more likely to have comorbidities. Treated patients also had worse baseline oxygenation and SOFA scores, and those in Modena had more severe markers of cytokine storm, including higher C-reactive protein, IL-6, and lactate dehydrogenase concentrations, and worse thrombocytopenia. However, the authors adjusted results for several of the key variables and found no differences based on these strata.

This study adds key new information to our understanding of tocilizumab in COVID-19. Previous studies of tocilizumab in COVID-19 have largely been encouraging, but either did not have a matched comparator group,5, 8 did not match to contemporaneous controls,⁹ or were small and probably underpowered for key safety and efficacy outcomes.10, 11 This study provides strong evidence that tocilizumab might prevent intubation and death in adults with severe COVID-19 pneumonia. These findings are also in agreement with emerging evidence that, in the setting of COVID-19-induced cytokine storm, immunosuppressive treatments might be most helpful earlier in the disease: after the onset of severe disease but before florid respiratory failure.⁹ The precise group of patients who might benefit from tocilizumab and the optimal biomarkers for identifying cytokine storm in the setting of COVID-19 remain unknown. The most crucial question concerns the relative utility of tocilizumab treatment versus other non-specific immunomodulatory agents (including corticosteroids) and other cytokine-directed therapies, which is the focus of multiple ongoing randomised trials.

© 2020 Shutterstock