Discussion

Tocilizumab is a humanised monoclonal antibody against the IL-6 receptor. It is licensed for use in patients with rheumatoid arthritis.4

In patients with severe COVID-19, there is secondary haemophagocytic lymphohistiocytosis (sHLH). This is characterised by cytokine storm (a surge of IL-2, IL-6, IL-7, tumour necrosis factor α, granulocyte-colony stimulating factor) and multiorgan failure.3 A tool to aid diagnosis is the ‘Hscore’. This score takes into account any underlying immunosuppressing disorders, temperature, presence of hepatomegaly/splenomegaly, AST, ferritin, triglyceride, any cytopaenia and haemophagocytosis on bone marrow aspirate. Our patient had a score of at least 169 at the time of the infusion, translating to a strong possibility of sHLH.5 This is relevant as 50% of patients with sHLH develop acute respiratory distress syndrome (ARDS).

The Berlin definition for severe ARDS is an acute deterioration (1 week or less) with bilateral chest X-ray changes and partial pressure of oxygen (PaO₂):fraction of inspired oxygen (FiO₂) ratio <100 mm Hg with positive end expiratory pressure (PEEP) >5 mm Hg. Severe ARDS is associated with a mortality of 45% even in the absence of COVID-19. Since we were managing our patient within the confines of a ward and not intensive care unit (ICU), we were unable to supply PEEP and so our patient does not meet the criteria for ARDS, but he certainly had severe COVID-19-related acute respiratory distress. This greatly improved post-tocilizumab infusion with rapid reduction in inspired oxygen and increase in SpO₂, and we propose that this is due to IL-6 blockade from tocilizumab.

To the best of our knowledge, we have described the first successful treatment of severe COVID-19 pneumonia and concomitant hyperinflammatory syndrome with tocilizumab in the UK. Tocilizumab is potentially crucial in blocking the cytokine storm and associated ARDS which might develop in a subset of patients with COVID-19. Based on the described case, we speculate that tocilizumab and other anti-IL-6 drugs including sarilumab6 should be administered as soon as possible in patients who develop an abrupt clinical deterioration 7–10 days from the initial onset of symptoms and have a high Hscore (169 or above), corresponding to a sensitivity of 93%, specificity of 86% and accurate classification of 90% of patients with hyperinflammatory syndrome.5 Early tocilizumab treatment could potentially (1) reduce the risk of requiring non-invasive or invasive ventilation; (2) offer a chance of survival to people who are unfit for escalation or have refused to be ventilated; and (3) potentially increase the chance of survival in some patients who are already ventilated but fail to improve with supportive treatment. However, in ICU patients the established degree of inflammatory lung damage might be too extensive to be reversible and additional complications might be present, including deranged coagulation function, disseminated intravascular coagulation and pulmonary embolism. Also, tocilizumab treatment should be avoided in people with severe active infections, including tuberculosis and hepatitis B. Finally, severe and sometimes fatal infections have been reported after tocilizumab treatment, although tocilizumab did not significantly increase the risk of infection in severely immunocompromised patients with haematological malignancy.7

The present case report is relevant for two main reasons. First, a unique peculiarity of this case is that the patient received tocilizumab treatment in the predexamethasone era, as no data on dexamethasone had been made available from the RECOVERY (Randomized Evaluation of COVID-19 Therapy) researchers at the time, or indeed on any other medications, including remdesivir. Hence, the very favourable effects of tocilizumab in this patient have not been potentially enhanced, or indeed completely explained, by other concomitant treatments, including dexamethasone or remdesivir.

Second, and most importantly, this case report shows a clear-cut temporal association between tocilizumab administration and rapid improvement in the clinical condition in a patient with clear hyperinflammatory syndrome, as highlighted by an Hscore of 169. This is a crucial point, as two randomised controlled trials aimed at evaluating the role of tocilizumab in severe COVID-19 pneumonia, the RECOVERY8 and the COVACTA (Study to Evaluate the Safety and Efficacy of Tocilizumab in Patients With Severe COVID-19 Pneumonia)9 trials, did not include the Hscore among the inclusion criteria. Indeed, under the overwhelming pressure of the urgent need for effective treatments against the COVID-19 pandemic, both studies decided to adopt much broader and somewhat aspecific inclusion criteria (for instance, C reactive protein ≥75 mg/L and oxygen saturation <92% on room air or requiring oxygen in the RECOVERY trial). As pointed out by Furlow,10 this issue might explain the disappointing preliminary results of the COVACTA trial, still unpublished but announced by Hoffmann-La Roche on 29 July 2020, which show that tocilizumab failed to meet its primary endpoint of improved clinical status and to improve mortality, although patients receiving tocilizumab had shorter hospitalisation time compared with the placebo arm. Tocilizumab continues to be evaluated by the RECOVERY trial, which enrolled over 850 patients, almost twice the size of the COVACTA trial, and might confirm or refute the results of COVACTA. Independently from the results of the RECOVERY trial, which suffers from the same limitations as COVACTA (lack of stratification by clinical signs of hyperinflammatory syndrome), we genuinely hope that the current case report could encourage researchers to perform post-hoc analyses of the above studies in due course or, even better, to design new randomised controlled trials which include Hscore as one of the key inclusion criteria.