High-stakes heterogeneity in COVID-19.

As all rheumatologists know, one size rarely fits all. An insufficient understanding of patient heterogeneity and its implications for prognosis and therapeutic response has contributed to numerous failed trials and insufficient treatment options in many rheumatic diseases. The same is true for COVID-19, for which understanding patient heterogeneity is a high-stakes endeavour. Such understanding could help clinicians to treat patients earlier and with the most effective therapies, potentially making the difference between life and death.

Most people infected with SARS-CoV-2 develop only mild symptoms or have no symptoms at all. A minority develop severe disease, however, and require hospitalisation; a small percentage of those develop life-threatening hyperinflammation. Thus far, treatment of these patients has been mostly reactionary, based on supportive care and close monitoring, with escalation of treatment in those whose conditions deteriorate. An ability to determine which patients are most at risk of developing or progressing to severe disease would facilitate a more proactive approach.

Cytokine storm syndromes in the context of rheumatic disorders have informed treatment of patients with COVID-19 and ongoing trials. However, as we understand more about the immune response to SARS-CoV-2, disparities between patients start to outweigh commonalities. In this issue of The Lancet Rheumatology, Jessica Manson and colleagues delineate a high-risk hyperinflammatory (COV-HI) phenotype in a cohort of UK patients who were admitted to hospital for COVID-19. Defined by two readily available clinical variables—C-reactive protein and ferritin—the COV-HI phenotype was associated with poor patient outcomes and predicted (by at least a day) need for increased respiratory support in nearly 75% of patients. Initial data from the COVID–ImmunoPhenotype (COVID-IP) project revealed that concentrations of a trio of circulating cytokines and chemokines—interferon gamma-induced protein 10 (IP-10; also known as CXCL10), interleukin (IL)-10, and IL-6—predicted subsequent clinical progression among patients admitted to hospital with COVID-19 (of varying severity).

A clearer understanding of patient heterogeneity might not only help clinicians to anticipate which patients are likely to require more aggressive treatment but also, in time, which therapies are most likely to benefit an individual patient. Inhibitors of cytokine pathways, including IL-1 (anakinra), IL-6 (tocilizumab, sarilumab), and GM-CSF (mavrilimumab), have shown some benefit in observational cohorts, although conclusive evidence awaits results of ongoing randomised trials. In one study, tocilizumab treatment was associated with reduced mortality in a cohort of 764 patients with severe COVID-19 requiring support in the intensive care unit, particularly in those with evidence of hyperinflammation. However, we still have a long way to go in predicting which patients might respond to specific cytokine inhibitors or to broader immunosuppression.

Emerging data from observational studies can also help to inform trial design. Both timing of treatment and patient stratification (eg, by evidence for hyperinflammation) could be key to determining trial success or failure; indeed, an absence of stratification for inflammatory phenotype might have contributed to the failure of two high-profile trials of tocilizumab in COVID-19. Post-hoc analysis of these negative trials will be essential to tease out whether particular subsets of patients might have benefitted from treatment.

Clues to patient heterogeneity and variability of clinical disease trajectories in patients with COVID-19 are also emerging from immunological studies. T cell responses might be key for long-term immunity; indeed, studies have shown that many people test positive for SARS-CoV-2-specific T cells, including some who tested negative for antibodies. Early vaccine trials have shown evidence for robust antiviral T cell responses, and preliminary data from a French case series showed that the T cell-boosting cytokine IL-7 safely increased T cell numbers in 12 critically ill patients with COVID-19 and severe lymphopenia. A trial of IL-7 is currently underway in the UK. A COVID-19 immune cell signature has been described, in which altered T cell phenotypes correlated with disease severity, and a paucity of T cells in blood better predicted length of hospital stay than did inflammatory markers such as IL-6.

There is no one-size-fits all scenario for COVID-19, but we have already made progress towards understanding heterogeneity and the importance of patient stratification for improving outcomes. With evidence suggesting that SARS-CoV-2 is unlikely to be eliminated any time soon, it is crucial to continue deciphering pieces of the COVID-19 heterogeneity puzzle.