No perfect therapy for the imperfect COVID-19 cytokine storm.

More than 2 years into the pandemic, almost 6 million people have died from COVID-19 worldwide. Many people who succumbed to the virus had cytokine storm syndrome, a dysregulated immune response to the pathogen. Progress toward treating COVID-19 has been substantial on several fronts, including rapidly developed safe and effective vaccines, and various antiviral therapies (eg, monoclonal antibody therapies, protease inhibitors, and nucleoside analogues). Antiviral approaches are particularly effective early during infection, but cytokine targeted therapies have shown benefit during later stages of illness, when hyperinflammation is present.

The most promising treatment for COVID-19 hyperinflammation is glucocorticoids when given to patients admitted to hospital with COVID-19 who require oxygen. Nonetheless, this broadly immunosuppressive approach has not been that effective. Targeting individual pro-inflammatory cytokines (eg, interleukin [IL]-1 and IL-6) has shown some survival benefit but, again, the effect has been rather underwhelming.3, 4 Somewhere in between these two approaches, Janus kinase (JAK) inhibitors disrupt signalling downstream from receptors that bind multiple cytokines. Different small molecule JAK inhibitors target different kinases associated with various cytokine receptors, and ruxolitinib preferentially targets JAK1 and JAK2, which signal downstream of numerous pro-inflammatory cytokines, including IL-6 and interferon-γ. In The Lancet Rheumatology, MeiLan Han and colleagues report results from a randomised, double-blind, placebo-controlled trial of ruxolitinib to treat patients with COVID-19 (RUXCOVID).

In the RUXCOVID trial, 432 patients with COVID-19 were randomly assigned (2:1) to ruxolitinib (5 mg twice daily for 2 weeks) plus standard of care or placebo plus standard of care. The primary endpoint was a composite of death and requirement for invasive ventilation or intensive care by day 29. 34 (12%) of 284 ruxolitinib-treated patients and 17 (12%) of 144 placebo-treated patients met the composite endpoint (odds ratio 0·91, 95% CI 0·48–1·73; p=0·77), but the median time to recovery was 1 day faster in the ruxolitinib group, although this difference was not statistically significant (hazard ratio 1·10, 95% CI 0·89–1·36). Han and colleagues concluded that ruxolitinib showed no benefit for the overall study population and that a larger clinical trial is necessary to show potential benefit in subgroups of patients with COVID-19 who showed potential improvement with ruxolitinib.

The RUXCOVID trial is another failed attempt to treat the hyperinflammation associated with COVID-19 with immunomodulatory therapy. These results differ from a trial that showed some COVID-19 survival benefit for another JAK1/2 inhibitor, baricitinib, when used in combination with the antiviral remdesivir. Less than 8% of patients with COVID-19 in the RUXCOVID trial received remdesivir, which might have contributed to this disparity in findings. Additionally, only slightly more than half of the patients in the RUXCOVID study received dexamethasone, which is now standard of care for patients admitted to hospital with COVID-19, and might allow for anti-cytokine approaches to be beneficial.3, 4 Another JAK inhibitor, tofacitinib, showed a COVID-19 survival benefit in conjunction with standard of care (89% of patients received glucocorticoids). Thus, JAK inhibitors, which inhibit signalling from multiple cytokines and are intermediate between glucocorticoid-induced broad immunosuppression and targeted cytokine approaches, appear to have a role in treating patients admitted to hospital with COVID-19.

So, why is that some cytokine inhibitors, including different JAK inhibitors, show benefit for patients admitted to hospital with COVID-19, and others do not? Trial design, including patient selection, different standard of care regimens, and timing of therapies, could be critical. The lack of benefit of ruxolitinib for patients with COVID-19 in the RUXCOVID trial is somewhat surprising, given the ability of ruxolitinib to benefit a wide variety of patients with cytokine storm syndrome. Therefore, the combination of glucocorticoids and ruxolitinib could be crucial to optimise therapy for patients with cytokine storm syndrome. Moreover, as the optimal inflammatory features of COVID-19 were relatively unknown at the time of the RUXCOVID trial, patient selection did not include features such as high C-reactive protein, hyper-ferritinaemia, or elevated pro-inflammatory cytokine concentrations, but rather relied strictly on clinical criteria. Patients with COVID-19 with hyperinflammatory states are more likely to benefit from anti-cytokine therapies. Interestingly, ruxolitinib lowered IL-2RA levels, a marker of hyperinflammation, in the RUXCOVID trial.

Although some laboratory markers (eg, IL-2RA, ferritin, D-dimers, C-reactive protein, and IL-6) of more standard varieties of cytokine storm syndrome are elevated in patients with COVID-19, the degree of elevation is often modest, and only subsets of patients with COVID-19 meet traditional criteria for cytokine storm syndrome. Unsurprisingly, established treatment approaches used for cytokine storm syndrome before the SARS-CoV-2 pandemic are notably less effective in treating the imperfect cytokine storm associated with COVID-19. Although there has been clear progress in dampening the cytokine storm associated with COVID-19, prevention of developing COVID-19 through highly effective and safe vaccines remains a priority.

I report grants and personal fees from Sobi, and personal fees from Novartis and Pfizer, outside the submitted work.