We read with interest the Letter from Siang Kow et al.1 commenting on our systematic review2 and their discussion of the preliminary evidence from recent randomized controlled trials (RCTs) on the efficacy and safety of tocilizumab for COVID-19. We concur with the authors that it may indeed be time to divert some of our attention from IL-6 to other mediators of inflammation in COVID-19 patients. In fact, previous pharmacological attempts to modulate the inflammatory response in patients with ARDS and sepsis have repeatedly proven unsuccessful. It is therefore reasonable to also question whether suppressing the pathophysiological inflammatory response, or blocking a single mediator for that matter, will be beneficial for patients with COVID-19.

The authors commented on the absence of significant difference in mortality between patients who received tocilizumab (Actemra/RoActemra) or placebo in the industry funded COVACTA (NCT04320615 - https://www.roche.com/dam/jcr:6d8de90d-2e31-43c8-b4e1-0a24a2675015/en/29072020-mr-covacta.pdf) and EMPACTA (NCT04372186 - https://www.roche.com/media/releases/med-cor-2020-09-18.htm) trials.

The results of these trials confirm that findings from nonrandomized trials should be interpreted with caution and that such caution is warranted particularly during public health emergencies when large numbers of patients may subsequently receive redundant treatments. As discussed by the authors in the context of tocilizumab and exemplified also by the hydroxychloroquine landslide,3 experimental drugs are not always harmless, particularly when indiscriminately used. Patient safety should always be prioritized, which is why experimental drugs must be administered within the framework of registered RCTs that are accompanied by appropriate monitoring and regulation.

Research methodology may have also contributed to the negative findings of the above-mentioned RCTs. One example of a potential determinant of outcome in relation to treatment is the timing of administration in respect to the clinical phase of the disease.4 Another is the treatment dose. Case mix may also have diluted the results; there may be sub-populations of COVID-19 patients who do actually benefit from receiving tocilizumab. Hopefully the full reports of the COVACTA and EMBACTA trials will shed some light on these questions and more. These analyses combined with additional data from the interventional tocilizumab arm of the RECOVERY trial (www.recoverytrial.net) may yet change our perspective on this drug. To summarize, although oft repeated, the following rhetoric is simply the truth: more (high quality) research is urgently needed.