Division of Clinical Immunology and Rheumatology, Department of Internal Medicine, Post Graduate Institute of Medical Education and Research, Chandigarh, Chandigarh, India
With 12 million cases and half a million deaths (as of 10 July 2020), the corona virus disease (COVID-19) has paralysed healthcare systems the world over. The observations by Potere et al and mechanistic insights by Capecchi et al merit a discussion of important considerations that should be borne in mind before using tocilizumab in severe COVID-19. The available evidence concerning its use, although promising, is preliminary and probably underpowered to detect important safety/efficacy signals.
Patients with COVID-19 are not free from the risk of a bacterial infection, which could be secondary (nosocomial) or co-primary (community acquired). Concomitant bacterial infections have been reported in 10%–20% of COVID-19 cases and ~50% of COVID-19-related deaths, consistent with, and reminiscent of bacterial infections being the the most common cause of mortality in previous influenza pandemics. This risk is higher in old age, presence of comorbidities and those requiring invasive mechanical ventilation (all probable candidates for use of tocilizumab). A hesitancy towards the use of CT and fiberoptic bronchoscopy at many centres to reduce COVID-19 transmission makes differentiation of COVID-19 acute respiratory distress syndrome (ARDS) from bacterial pneumonia challenging. Suppression of C reactive protein, leucocytosis and fever by tocilizumab render these parameters unusable for diagnosing an underlying bacterial infection. Inadvertent use of tocilizumab in the latter setting may be catastrophic.
To complicate things further, an increased risk of serious infections with tocilizumab (initially deemed unlikely with short-term use) has recently been demonstrated in COVID-19. A 13% higher risk of new infections was seen with tocilizumab when added to standard-of-care, in the largest cohort study available until date. Two deaths due to septic shock were seen in another cohort of 100 patients treated with tocilizumab. Candidaemia with candidal endophthalmitis and endocarditis was reported in 3 of the 43 severe COVID-19 patients treated with tocilizumab in Italy.
Gastrointestinal involvement in COVID-19 is fairly common (pooled prevalence=15%), especially in patients with severe disease. Considering the compromised mucosal integrity in these patients because of intestinal hypoperfusion, and extrapolating the experience from rheumatic diseases, there is a realistic possibility of bowel perforation with the use of tocilizumab in COVID-19. Unsurprisingly, two such cases have already been reported.
Most safety data for tocilizumab is derived from autoimmune rheumatic diseases, where routine screening, especially for viral hepatitis, dyslipidaemia, diverticular disease and latent infections including tuberculosis (TB), precedes its use. Whether short-term emergent use, as in COVID-19, warrants a similar or abbreviated screening is still unknown. The narrow window of intervention in severe COVID-19 necessitates prompt administration, often without screening, which could be potentially lethal. Mortality due to herpes simplex virus-1 reactivation-related liver failure was recently reported with the use of tocilizumab in COVID-19, along with hepatitis B reactivation. There have also been reports of acute severe hypertriglyceridaemia (with subsequent pancreatitis) with tocilizumab, although confounded by concomitant use of lopinavir–ritonavir.
Although data from randomised controlled trials (RCTs) and registries in non-endemic countries suggests a low risk for reactivation of TB with tocilizumab, data from high TB-endemic settings is unclear. Should empirical isoniazid prophylaxis be given to COVID-19 patients with a history of TB or those living in high-endemic countries? Should latent TB testing be done before administration using the interferon gamma release assay (considering tubercular skin testing will take 48–72 hours)? Or, considering short-duration use (one to three doses), are these concerns misplaced? There is no data to back any of these statements yet.
Finally, the cytokine release syndrome (CRS) of severe COVID-19 has many key players apart from interleukin 6 (IL-6), including tumour necrosis factor- alpha, granulocyte colony stimulating factor, interferon gamma and IL-1 beta. In such a multipronged pathogenetic model, should tocilizumab be considered only in those patients with elevated serum IL-6 levels, as suggested by the Chinese guidelines? Since measuring IL-6 is difficult in routine clinical practice, could measurement of serum CRP be a reliable, cost-effective surrogate? A better understanding of COVID-CRS pathogenesis is required.
The use of tocilizumab in COVID-19 outside of an RCT setting thus needs to be rationalised. Its short-term (one to three doses) use is not free from serious adverse events, and a vigilant monitoring is mandatory. The efficacy data, although promising, are preliminary. Results of ongoing RCTs are eagerly awaited.