Len and Cassie Calabrese COVID-19 Corner Issue #19
The Challenges of doing successful trials of immunomodulators in COVID-19
Since the beginning of the pandemic it has been apparent that patients with serious and advanced disease experience an inflammatory phase of the illness that the vast majority of individuals who recover from the infection do not. (Figure 1 below)
Adapted from Leonard H. Calabrese Cytokine storm and the prospects for immunotherapy with COVID-19, Cleveland Clinic Journal of Medicine Date: 30-Jun-2020 DOI: 10.3949/ccjm.87a.ccc008
This “phase three” has been referred to as cytokine release syndrome or cytokine storm and with this modeling in mind a myriad of trials with immune based therapies with varying rationales (and lack of rationales) were launched. As of October 28th there are now 3578 clinical trials registered investigating hundreds of drugs and therapies.
If we reflect over the past 9 months, the study of COVID-19 late stage disease was marked by initial enthusiasm, multiple positive studies of non-randomized design and finally multiple negative studies of rigorous RCT design. We thought we would reflect on this phenomenon in this issue of COVID Corner.
I think that IL-6 is a particularly interesting and informative chapter in this evolving story. In April we wrote an article in the Cleveland Clinic Journal of Medicine entitled: “Cytokine storm and the prospects for immunotherapy with COVID-19”. At that time there were only a few small trials in the public domain, all of which were positive, demonstrating more rapid disease resolution, less need for ventilator support and improved survival. By April/May IL-6 targeting was being utilized in virtually every ICU in every major institution in the country as well as widely used abroad. The article was designed to help familiarize intensivists and others managing COVID-19 patients who were not familiar with our drugs. We ended with this comment: “The pandemic has created major ethical and practical questions about patient selection and the use of open label versus use in the context of a randomized clinical trials. Pandemics such as this one suggest that we cannot wait for perfect data from randomized trials when there are reasonable data available that suggest potential efficacy thus this review is an attempt to provide practical information as why, how and when to target IL-6 with a vigilance for safety. We eagerly await the ability to define the standard of care which will ultimately only be determined by properly preformed trials.”
We clearly stand by this appraisal made at a time of desperation but times have changed and the data are now clear. IL-6 targeting may be effective but it is not a home run for COVID-19. Gupta and colleagues have just recently reported the results of a large non-randomized multicenter cohort study that included 3924 patients that found the risk of in-hospital death was estimated to be lower (28% vs 37%) with tocilizumab treatment in the first 2 days of ICU admission compared with no early use of tocilizumab. While clearly good news it is not definitive as due to its non-randomized design it was susceptible to numerous confounders. In an accompanying editorial (Parr JB. Time to Reassess Tocilizumab’s Role in COVID-19 Pneumonia. JAMA Intern Med. Published online October 20, 2020. doi:10.1001/jamainternmed.2020.6557), Parr compares these data to all published RCTs with this drug, none of which met their primary end points and concluded that the picture is far from clear. He also points out that the NIH and Infectious Disease Society of America now conclude that IL-6 inhibitors should only be given in the context of clinical trials. And hot off the press is the well-designed and executed RCT from our colleague John Stone published this week in the NEJM (Oct 21), https://www.nejm.org/doi/full/10.1056/NEJMoa2028836?query=featured_coronavirus which was negative in all primary endpoints and we are even more cautionary. Add this to the fact that the large Phase 3 RCTs of both tocilizumab and sarilumab were stopped for futility leading to a dampening of enthusiasm for a drug class that was expected to work and work dramatically. We urge you to read the Parr editorial which elegantly discusses the basis to critique such trials and is noted below.
What could have happened? We posit four factors that are affecting all trials right now which should be kept in mind when examining all subsequent trials of immunomodulatory agents (and antivirals for that matter).
- It’s hard to improve on success. Early on in the epidemic when no one knew anything about management, mortality rates were astronomical and approached 50% in some ICUs. Over time, our critical care colleagues figured out a lot on how to manage this disease with more judicious use of ventilator support, prone ventilation, anticoagulation, etc., and now in some units the mortality is not only less than 10% but closer to 5%. In one IL-6 study from Italy by our colleague Carlos Salvarani, the mortality in the control group was 1.6%! Add to this that in the past month or two, all patients sick enough to be admitted to the hospital are now started on remdesivir and then dexamethasone when hypoxic. While this complicates analysis of all new therapies this is good for the patient!
- Timing is everything. If you buy the three act model it is clear that in Acts 1 and 2 we do everything we can to promote and boost our integrated adaptive and innate immune responses. Accordingly anything immunosuppressive given too early works against us. This is not as simple as it sounds as glucocorticoids given late may be good but bad if given early. This paradigm is not so clear with other drugs targeting specific cytokines (i.e. TNF, IL-6, IL-17, IL-23) for which we have less granular knowledge of their precise role in viral infections. This is even more important for agents like interferon which theoretically will benefit us in Acts 1 and2 but could be harmful in Act 3 where they may promote inflammation. Determining when the right time to block IL-6 is a work in progress.
- Chronologic time is not immunologic time. Unfortunately most trials merely enter patients based on how many days since symptom onset and what stage of illness they are using a 7 point ordinal scale from asymptomatic to being dead. Studies now starting to come out that we have previously referred to as BIG DATA show that some patients in Act 3 are profoundly immunosuppressed while others are literally on fire. A further breakdown reveals that the inflammation mat be driven by lymphocytes, monocytes and may be alternatively triggered by virus (i.e. PAMPS) or danger associated antigens from tissue necrosis (i.e. DAMPS). Entering patients on the basis of time and ordinal scale disease does not address this.
- New clinical subsets are also to be accommodated. We have discussed the newly described syndromes of MIS-C in children and MIS-An in adults which, while relatively rare, likely represent distinct immune endotypes and should be studied separately. The final frontier to us is to tease out the majority of people in Act 3 where virtually all have advanced pulmonary disease which are clearly heterogeneous. Many have secondary bacterial infections and others have advanced immuno-coagulopathy. Others may be PAMP driven, underlying their diffuse alveolar damage, while others may have uncontrolled viral replication. Since we have no biomarkers for any of these scenarios we must ask ourselves, is reasonable expect to hit a home run with any single therapy?
Moving ahead we look forward to trial results to analyze post hoc searching for biomarkers and dissecting immune endotypes. Next we will be using these new tools to enter select patients within clinical strata to make COVID-19 more personalized. We’re getting closer.
Next issue- the Immune therapeutics!
Cassandra Calabrese, DO, Prabalini Rajendram, MD, Gretchen L. Sacha, PharmD and Leonard Calabrese, DO Practical aspects of targeting IL-6 in COVID-19 diseaseCleveland Clinic Journal of Medicine October 2020, DOI: https://doi.org/10.3949/ccjm.87a.ccc018
Gupta S, Wang W, Hayek SS, et al. Association Between Early Treatment With Tocilizumab and Mortality Among Critically Ill Patients With COVID-19. JAMA Intern Med. Published online October 20, 2020. doi:10.1 https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2772185?resultClick=1001/jamainternmed.2020.6252
Parr JB. Time to Reassess Tocilizumab’s Role in COVID-19 Pneumonia. JAMA Intern Med. Published online October 20, 2020. doi:10.1001/jamainternmed.2020.6557
Have questions, or some topics you want us to dig into? Email us please
Len and Cassie