Coronavirus is the trigger, but the immune response is deadly.

As of mid-April, 2020, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the coronavirus responsible for COVID-19, has infected more than 4·4 million people worldwide, killing more than 300 000. A multifaceted methodology is necessary for this pandemic to reduce mortality (panel ). While awaiting safe and effective vaccines, and antivirals that substantially reduce death, we struggle to identify effective therapy for patients with COVID-19 who have been admitted to hospital. Early reports from China detailing clinical features (fever, confusion, respiratory distress, and coagulopathy) and laboratory features (lymphopenia, and elevated C-reactive protein [CRP], D dimers, lactate dehydrogenase, soluble interleukin-2 [IL-2] receptor-α, and ferritin) of severe COVID-19 are reminiscent of cytokine storm syndrome.

Safe and effective vaccines in the future

Public health measures (eg, social distancing)

Hospital or intensive standard of care (eg, proning)

Anticoagulation for patients at high risk of clots (eg, elevated D dimers)

Antiviral therapy—awaiting published effective treatment

Anticytokine storm therapy (eg, interleukin-1 blockade)

Cytokine storm syndrome is an umbrella term describing various frequently fatal hyperinflammatory conditions, such as macrophage activation syndrome, haemophagocytic lymphohistiocytois, and cytokine release syndrome. There are many associated diseases (eg, lupus and lymphoma) and triggers (eg, dengue and CAR T-cell therapy) that can result in cytokine storm syndrome. Viral infections, including pandemic influenza strains, are common triggers, and SARS-CoV-2 can be added to the list. There are several pathophysiological pathways that can result in cytokine storm syndrome, but the best studied pathway is defective lymphocyte killing via the perforin pathway. Homozygous defects in perforin pathway genes cause familial haemophagocytic lymphohistiocytosis, and heterozygous mutations in these same genes are associated with secondary haemophagocytic lymphohistiocytosis, including fatal H1N1 influenza. Whether similar, or novel, genetic defects contribute to the severity of COVID-19-associated cytokine storm syndrome is unknown, but should be answered by genomic sequencing of patients admitted to hospital with COVID-19-associated cytokine storm syndrome.

Regardless of mechanism, cytokine storm syndromes share features of inappropriately elevated proinflammatory cytokines (IL-1, IL-6, and interferon-γ] produced by a dysregulated host immune response, resulting in multiorgan failure. All cytokine storm syndromes are not identical, and COVID-19-associated cytokine storm syndrome has some unique features, including propensity for early acute respiratory distress syndrome and clotting, while having elevated but lower serum ferritins and lower IL-6 concentration than routinely noted in other cytokine storm syndromes.1, 2 Nonetheless, COVID-19 triggers a hyperinflammatory response in a substantial number of patients who require hospital admission. Different therapies have been used over the years for treatment of various cytokine storm syndromes. These therapies include broadly immunosuppressive approaches (eg, glucocorticoids and calcineurin inhibitors) and targeted immunomodulatory therapies (eg, anticytokines and Janus kinase inhibitors).1, 2 On the basis of previous experience with other deadly coronaviruses (severe acute respiratory distress syndrome and Middle East respiratory distress syndrome), WHO has recommended against glucocorticoid treatment of patients with COVID-19. However, targeting of cytokines in other cytokine storm syndromes has proved to be effective and is attractive because of the lower frequencies of associated side-effects. Thus, anticytokine approaches are being explored for treating COVID-19-associated cytokine storm syndrome.

As IL-6 is easily measured, and therapies targeting IL-6 are available in China, early approaches to treating COVID-19-associated cytokine storm syndrome involved IL-6 blockade. There are now dozens of clinical trials studying blockade of IL-6 or its receptor, and early publications of trials and non-trial cohorts are reporting mixed results. Nevertheless, there is evidence that blocking IL-6 signalling might benefit some patients with COVID-19. IL-1 is another pro-inflammatory cytokine targeted to treat various cytokine storm syndromes. IL-1 blockade with anakinra (a recombinant human IL-1 receptor antagonist) notably improved survival in a subset of patients with sepsis and features of cytokine storm syndrome, and was otherwise safe in this setting. There are currently about a dozen clinical trials exploring IL-1 blockade for COVID-19-associated cytokine storm syndrome.

While awaiting results of these clinical trials, case series of anakinra treating COVID-19-associated cytokine storm syndrome are being published. Three small cases series reported anakinra benefitting patients with COVID-19.6, 7, 8 Now, in The Lancet Rheumatology, Thomas Huet and colleagues describe a larger prospective, single-centre cohort study with a historical control cohort, demonstrating that anakinra significantly reduced requirement for invasive mechanical ventilation and death in patients who received anakinra (n=52) compared with historical controls who received usual care (n=44; hazard ratio 0·22 [95% CI 0·11–0·41], p<0·0001). Patients who were admitted to hospital were consecutively enrolled on the basis of having laboratory-confirmed SARS-CoV-2, bilateral lung infiltrates on imaging, and critical pulmonary function before admission to the intensive care unit. Admission to an ICU or death occurred in 13 (25%) of 52 patients in the anakinra group compared with 32 (73%) of 44 patients in the historical control group. The investigators also demonstrated a significant reduction in oxygen requirement from a median of 7 L/min (IQR 6–9) to 2 L/min (0–4) within a week of starting anakinra therapy. There was an associated decrease in CRP concentration in the anakinra group compared with the historical control group within 4 days of starting anakinra therapy (p<0·0001). The study has potential biases as do all non-randomised, non-blinded studies, but multivariate analysis was used to control for potentially confounding factors, such as body-mass index, without change in conclusions. Other than seven patients in the anakinra group and four patients in the control group with elevated liver enzymes, there were no substantial side-effects. This study is the most definitive evidence to date that anakinra can benefit patients with COVID-19-associated cytokine storm syndrome. The significant reduction in mortality associated with anakinra use for COVID-19 in this study is encouraging in these challenging times.

The choice of anakinra to treat deadly cytokine storms is wise because it has a remarkable benefit-to-side-effect ratio. It is a recombinant human protein that blocks both IL-1α and IL-1β with a half-life of about 4 h, has a large therapeutic window (1–48 mg/kg per day), can be given intravenously or subcutaneously, has a well established favourable safety profile, and works quickly and effectively for various cytokine storm syndromes. A top priority in this pandemic is to learn effective therapy from clinical trial enrolment. However, it is difficult for health-care professionals in regions overwhelmed by COVID-19 to witness so much death. Although it is important to treat the virus of infected patients as best we can, if a cytokine storm syndrome is present, then that also must be managed to reduce mortality from an excessive host immune response. Awaiting controlled trial results, anakinra provides hope for those severely affected by COVID-19.