As of January, 2021, nearly 2-million deaths worldwide have been attributed to COVID-19, which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Much of the mortality has been associated with a cytokine storm syndrome in patients admitted to hos pital with COVID-19 pneumonia. 1 Defining the COVID-19 cyto kine storm syndrome has been challenging, but early reports proposed combinations of clinical (eg, fever) and laboratory (eg, hyperferritinaemia) features in determining patients most likely to benefit from cyto kine storm syndrome treatment. 2, 3 A vast array of anti-inflammatory therapies are being explored to dampen the cytokine storm syndrome to save lives. One of the first approaches to treat COVID-19 cytokine storm syndrome was targeting interleukin-6 (IL-6). Early during the pandemic, IL-6 concentrations were noted to be elevated, and IL-6-blocking therapies were available in China, whereas IL-1 inhibitors were not. Retrospective case series of monoclonal anti bodies binding IL-6 or its receptor presented mixed results in potential benefit in treating COVID-19 cytokine storm syndrome; however, most randomised controlled trials have not documented improved survival with agents targeting IL-6. 4 By contrast, targeting IL-1, another pro-inflammatory cytokine that has been targeted effectively in other cytokine storm syndromes, 5 has been reported to be largely successful in improving COVID-19 survival based on retrospective cohort studies. 6

In The Lancet Rheumatology, Giulio Cavalli and colleagues compared the effectiveness of IL-1 and IL-6 inhibition in the treatment of COVID-19 cytokine storm syn drome. 7 This single-centre, observational study of patients admitted to hospital with COVID-19, respiratory insufficiency, and hyperinflammation (elevated C-reactive protein≥ 100 mg/L or ferritin≥ 900 ng/mL) analysed mortality in those receiving an IL-1 receptor antagonist (anakinra; n= 62) or one of two monoclonal antibodies binding the IL-6 receptor (tocilizumab or sarilumab; n= 55) versus no interleukin inhibition (n= 275). The study suffers from potential biases that are frequent in non-randomised studies, but the authors controlled for baseline clinical differences among groups using multivariable Cox regression analysis, as well as immortal bias by excluding early (within 24 h from enrolment) deaths and intensive care admissions. Moreover, many