Shelter from the cytokine storm: The changing role for rheumatology in COVID-19
As COVID-19, the viral ‘storm of the century’, continues to batter health care systems across the world, clinicians and researchers are scrambling to understand the virus and how to contain it. The process is imperfect; despite international research efforts, the data, as yet, have been simultaneously overwhelming and insufficient.
However, global research collaboration has shed light on certain clinical trends related to COVID-19, some of which indicate that one possible answer to this pandemic may lie in another ‘storm’. Several clinicians have reported a pattern of tempestuous inflammation in the most severe cases of COVID-19 that resembles cytokine storm syndrome, a phenomenon that can occur in certain rheumatologic diseases.
Often fatal in both the rheumatology and COVID-19 settings, involvement of the cytokine storm has set the stage for the rheumatology community to step onto the frontlines in the fight against the virus.
“Features of critically ill patients infected with COVID-19 suggest the presence of a cytokine storm syndrome resulting in acute respiratory distress syndrome and multi-organ failure,” W. Winn Chatham, MD, Louis W. Heck Clinical Scholar in Rheumatology in the division of clinical immunology and rheumatology at the University of Alabama at Birmingham, and Randy Q. Cron, MD, PhD, of Children’s of Alabama and the University of Alabama at Birmingham, wrote in a recent editorial in The Rheumatologist.
They were among the first to raise the red flag. “Indeed, many of the diagnostic criteria for cytokine storm syndrome are reported present in those COVID-19 infected individuals under intensive care.”
Two of those diagnostic criteria that are shared among various cytokine storms and COVID-19 are elevated ferritins and C-reactive protein (CRP). It should follow, then, that intensivists and hospitalists on the front lines with COVID-19 should obtain these lab results for the most severe patients and treat them accordingly.
Not so fast — cytokine storm syndrome is a big umbrella. Understanding the nuances of what falls under that umbrella is critical to making effective treatment decisions.
“Cytokine storm syndrome can be primary or genetic – referred to as the familial form – or it can be acquired,” Theresa Wampler Muskardin, MD, of the division of pediatric rheumatology at New York University Langone Health, told Healio Rheumatology.
The primary form is known as hemophagocytic lymphohistiocytosis (HLH), while secondary HLH may also be called macrophage activation syndrome (MAS). “It is important to understand that cytokine storm syndromes exist on a continuum, rather than being clearly defined, independent conditions,” Muskardin said. “This is why recognizing them can be so challenging. There can be so much overlap between different kinds of cytokine storm.”
It is for these reasons that clinicians should tread carefully when managing possible cytokine storm syndrome in the setting of COVID-19, according to Maximilian F. Konig, MD, of the division of rheumatology at The Johns Hopkins University School of Medicine.
“Patients with this severe form of COVID-19 should ideally be treated as part of ongoing randomized, controlled clinical trials which will ultimately allow us to identify the most appropriate treatments and when to apply them,” Konig said in an interview. “Many cytokine responses in severe COVID-19 mimic what we see in HLH or MAS. But whether these diseases are identical or fundamentally different from the cytokine storm syndrome in COVID-19, we still do not know.”
“It is important that we are careful about not over-simplifying and labeling what we see in COVID-19 as HLH or MAS,” he added. “The risk is that clinicians may haphazardly apply knowledge that may not directly translate to patients with severe COVID-19.”
Despite these concerns, early case series data have shown that certain drugs in the rheumatology armamentarium — including those that influence the IL-1, IL-6 and JAK/signal transducer and activator of transcription (STAT) pathways — have shown efficacy in COVID-19 patients with suspected cytokine storm syndrome. Whether these findings will bear out in randomized controlled trials remains to be seen.
The experts who spoke with Healio Rheumatology have paid close attention to these data and would like nothing more than to share their knowledge of the storm with those on the front lines. The goal would not be to cure or eradicate COVID-19 altogether, but to mitigate fatalities in this most severely impacted subset. But before that can happen, a deeper exploration of cytokine storm syndrome is essential.
Understanding the Storm
“Cytokine storm is defined by new onset hyperferritinemia in the setting of fever,” Joseph Carcillo, MD, associate professor in the departments of critical care medicine and pediatrics at the University of Pittsburgh School of Medicine,” told Healio Rheumatology in an interview. “Until recently, ferritin levels have only been checked to diagnose iron deficiency anemia when low; however, we now know that the new onset of very high ferritin levels reflects macrophage activation.”
New ferritin levels higher than 500 ng/mL are associated with macrophage activation, while those above 1,000 ng/mL are associated with death. “The risk for mortality increases in a stepwise fashion at 1,000, 3,000 and more than 10,000,” Carcillo said. “Certain organ systems, such as the reticuloendothelial system, are involved as the activation worsens including the liver, causing hepatobiliary dysfunction, elevated ALT and bilirubin, and damage to the blood vessels, leading to disseminated intravascular coagulation and elevated D-dimers.”
Other diagnostic criteria for cytokine storm syndrome include lymphopenia, prolonged prothrombin time, elevated lactate dehydrogenase (LDH), elevated IL-6 and elevated soluble CD25.
It is also documented that MAS is seen most commonly in the setting of lupus or systemic juvenile idiopathic arthritis (sJIA/Still’s disease), according to Muskardin. While COVID-19 has largely spared juvenile patients from severe acute disease, Muskardin highlighted this bit of information for a specific reason: “Rheumatologists are alert to the signs of cytokine storm because it can occur in some of our patients and, without treatment, the patients can die quickly. We’re trained to recognize the syndrome and intervene expeditiously.” she said.
An additional point is that cytokine storm syndrome may have different initiating pathways, depending on whether it is genetically- and/or viral-induced. “We know that the pathways culminate in cytotoxic dysfunction, and that T cells, NK cells and macrophages are recruited and create an over-exuberant inflammatory response,” Muskardin said.
It was such an inflammatory response that tipped off clinicians treating COVID-19 to the possibility that what they were seeing was, in fact, cytokine storm syndrome. But understanding where HLH or MAS ends and a COVID-related storm begins is another matter altogether.
Applying Knowledge to COVID-19
“What I see when I look at these patients with COVID-19 is that they are sick for a prolonged time, and right when they should be getting better, they suddenly get worse,” Michael B. Jordan, MD, member of the division of bone marrow transplantation and immune deficiency and the division of immunobiology at the University of Cincinnati, told Healio Rheumatology. “That delayed syndrome suggests either a secondary bacterial infection or an immune-mediated pathology.”
Jordan suggested that there has been little to no evidence of secondary infections in this subset of COVID-19 patients, which points to cytokine storm as the culprit.
It is important to understand that there is context for clinicians drawing associations between a viral infection and cytokine storm syndrome, according to Cron. “We had similar thoughts a decade ago with the Ebola outbreak, and this can also be seen in some patients who develop dengue fever,” he said. “Doctors have been trying to target the cytokine storm in these patients for a long time, so, it is not too crazy of an idea. We are attuned to it.”
Clinical summaries and case reports of COVID-19 data validated these hypotheses. “There is evidence that ferritins are elevated in COVID-19, although they are not screaming high like they are in other types of cytokine storm syndrome,” Chatham said. “Coagulopathy is also quite prevalent, although it is a lot more than we see in lupus or Still’s disease.”
With this in mind, Cron offered some practical advice for his colleagues in the ICU. “If you have a patient with COVID-19 who has a high fever and respiratory distress, elevated ferritin and/or CRP may be used as surrogate decision-making criteria to determine if the patient has cytokine storm syndrome,” he said. “On top of that, simple turnaround labs such as a complete blood count can also help to confirm that this is happening.”
In addition to lymphopenia or thrombocytopenia, elevation of ferritin, liver enzymes, and many cytokines is seen in COVID-19, similar to other cytokine storm syndromes, according to Konig. He also noted that D-dimer can be profoundly elevated and may reflect both the degree of inflammation and coagulopathy in COVID-19.
As a pediatric rheumatologist, Jordan has significant experience with the storms in HLH and MAS. But even he would be cautious in applying that knowledge to COVID-19. “Of course, there are some similarities between what I see and what is happening with the virus, like high ferritins or elevated AST,” he said. “But my overall interpretation is that few of these patients would adequately meet diagnostic criteria for HLH or MAS. We need to be certain of what we are dealing with.”
Konig added that personnel in intensive care should pay attention to all of this information — the similarities and the differences alike — for one key reason: “These are signals that may be uniquely associated with mortality,” he said.
Timing of Intervention
If there is another factor for ICU doctors to consider, it is when to intervene in the event of suspected cytokine storm syndrome. “We know from publications in China that people who end up dying of severe COVID-19 show changes in D-dimers and troponin early on,” Konig said. “They diverge within a day or two of being symptomatic: A window of opportunity to identify patients at risk early.”
Carcillo offered data supporting early intervention. “According to the clinical progression chart presented in a paper by Zhou and colleagues, fever developed at day 1, sepsis at day 10, admission to the ICU for acute respiratory distress syndrome at day 12 and death at 19 days,” he said. “Non-survivors also had secondary bacterial infections after day 14.”
It is worth noting that in this data set, hyperferritinemia at days 4 and 7 predicted non-survival long before development of sepsis and ICU admission, according to Carcillo. “This observation provides an opportunity to study whether therapies directed to reversing macrophage activation and cytokine storm syndrome when given at day 4 through 7 of COVID-19 illness to patients with hyperferritinemia could reduce progression to need for ICU care.”
Both Cron and Muskardin are adamant that intervention prior to ICU admission is of the essence. “The main goal of using immunomodulation in COVID-19 is to, ideally, prevent need for intensive care. However, such therapy should not be used in patients who are not likely to progress to severe disease,” Muskardin said. “We are telling most people to stay home if they do not have moderate to severe symptoms. So, if a patient is sick enough to be admitted, it is worthwhile to consider as early as possible after admission whether they have known risk factors for severe disease or other signs to warrant immunomodulatory treatment. This is when intervening can make a difference.”
For Cron, avoiding intubation should be a primary goal. “I would recommend taking the appropriate labs and acting within the first 24 hours of hospital admission,” he said. “Reducing the number of patients who are admitted to the ICU is going to have a number of benefits, including reducing the number of health care workers who are required to be there on the front lines.”
While Konig is sympathetic to this strategy, he believes there is much more work to be done to validate an early intervention approach. “In studies from China, among patients who ultimately died, there was very high percentage who developed secondary infection and sepsis,” he said. “If we were to immunosuppress patients early on with a drug that has a half-life of days or even weeks, you may put them at an increased risk of dying from infection.”
Somewhat reassuringly, according to findings from Gianfrancesco and colleagues, not all patients who are on long-term treatment with immunosuppressive drugs — for example, patients with rheumatic diseases or inflammatory bowel disease — have poor outcomes when developing COVID-19.
Despite his belief in early and swift action, Cron acknowledges the myriad quandaries faced by clinicians managing these patients. The first: treat or wait.
“No one wants to watch patients die without doing anything,” he said. “But, of course, we do not want to mistreat patients and do harm, either. This raises the conundrum of whether we wait for clinical trials to emerge, showing — conclusively — the safety and efficacy of medications in the IL-1 or IL-6 pathway, or whether we look at the anecdotal evidence that is emerging in real time, and try to treat them as best we can. I do not envy our colleagues in hard hit areas who have to make these choices every day.”
Despite these reservations, Cron suggested that a trend is emerging. “These doctors are not waiting,” he said. “This is an unprecedented time. People are dying. We should not just try any random idea that pops into our head, but if there is rationale for an idea — and I believe there is rationale for the treatments I am going to discuss — we should be aware of it.”
The other quandary faced by clinicians involves the choice to treat the virus or the cytokine storm, according to Cron. “Some of these agents, like the IFN or JAK/STAT inhibitors, may be treating the cytokine storm but making it harder to clear the virus,” he said. “You may have to choose a therapy that does not clear the virus quite as well but reduces the cytokine storm, so the patient does not die. You might need to make that choice.”
Konig explained the one reason why this choice is so difficult to make. “We do not have data as to when to strike the balance between not interfering with antiviral immune responses and suppressing the cytokine storm,” he said. “We are dealing with a different beast altogether, so we are learning in real-time.”
If there is good news, it is that the research community is in full swing investigating drugs that have shown some effectiveness in COVID-19. Carcillo combed through ClinicalTrials.gov and counted 11 studies investigating IL-6 inhibitors, 16 for hydroxychloroquine, five for steroids, and one each for NSAIDs, IL-1 inhibitors, IFN-gamma inhibitors and the JAK/STAT pathway in the COVID-19 space.
“There is unlikely to be a silver bullet, but it is likely that a combination of these therapies will be used, depending on time of disease, host response and underlying host chronic illnesses,” Carcillo said.
Looking at findings that have already emerged, two data sets involving tocilizumab (Actemra, Genentech) initially caught the attention of rheumatologists.
Xu and colleagues assessed the efficacy of tocilizumab as an additional component to routine therapy in a cohort of 20 patients who were severely impacted by COVID-19. Results showed improvements in both fever and other symptoms in all patients, with 15 of 20 showing a reduced oxygen intake and one requiring no oxygen at all.
Tocilizumab was associated with lymphocytes returning to normal in 52.6% of patients after 5 days of treatment, with significant decreases in CRP reported in 84.2% of the cohort. With minimal adverse reactions reported and 90.5% of the cohort discharged after 13.5 days, the researchers believe tocilizumab may be positioned as a potential therapeutic option for the most severely impacted patients.
Similarly, Luo and colleagues reported on a case series of 15 patients in China with COVID-19-associated cytokine storms. Eight of those patients were treated with both tocilizumab plus methylprednisolone, while five patients received tocilizumab twice or more during their stay. Three of four patients who received a single dose of tocilizumab died, while one patient survived but with CRP failing to return to normal range. In the 10 patients who responded, after an initial spike, tocilizumab ultimately decreased IL-6.
“Early on, these were the only data sets we had to work with,” Konig said.
While the findings are encouraging, the obvious issue voiced by many experts, including Chatham, is that the data sets are case series rather than randomized controlled trials. He also voiced one other issue with the IL-6 approach. “Sometimes there is significant liver injury,” he said.
Turning to the IL-1 pathway, Monteagudo and colleagues conducted a retrospective chart review of five patients with MAS who were treated with continuous IV anakinra (Kineret, SOBI) infusions up to 2,400 mg/day. Results showed both serologic and clinical response in four of the five patients. The researchers suggested that this approach should be considered for COVID-19 patients who present with the cytokine storm presentation.
Chatham described what seems to be the straightforward consensus on anakinra. “It has a pretty good safety profile,” Chatham said. “Also, it has a short half-life, so even if it is not effective, it clears from the system quickly.”
Muskardin suggested that the key challenges facing anakinra have nothing to do with safety or potential efficacy. “As yet, there is no FDA-approved protocol for use of this drug in cytokine storm syndromes, despite its standard-of-care use among pediatric rheumatologists in managing the cytokine storm syndrome , MAS. If it is not part of a COVID-19 clinical trial, it feels way too risky for physicians who are not familiar with this particular biologic’s use,” she said.
The word on anakinra is getting out, but slowly. Insurance coverage remains a problem, and the lack of clinical trials for the drug makes it nearly impossible for a patient to receive it by being registered for a study.
All of this has led to frustration among the experts who spoke with Healio Rheumatology. There is mounting evidence that cytokine storm syndrome is killing these patients, and mounting evidence that drugs in the rheumatology space can mitigate these fatalities, but the word is not yet out.
A Neglected Resource
If there is one other area linking cytokine storm syndrome to COVID-19 that the broader medical community should consider, it is the genetic background of HLH and MAS, according to Jordan. “HLH is usually genetic,” he said. “It is traceable to cytotoxic function of T cells.”
A deep dive into genetic studies of HLH and MAS could offer insight into pathways of immune deficiency that may be present in the subset of COVID-19 patients who experience a cytokine storm. “There are families of genes found in patients with HLH, and in those who develop MAS in sJIA,” Jordan said. “Some patients have bi-allele mutations, others have single allele mutations that may or may not occur in a different context in COVID-19. This information could be valuable.”
Muskardin insists that familiarity with cytokine storm syndromes is critical to saving lives during COVID-19. “I preemptively spoke to colleagues about this, anticipating that we would likely get called for consultation regarding management of cytokine storm or use of biologics in COVID-19, and offered my expertise in managing patients with forms of cytokine storm syndrome and use of IL-6 and IL-1 inhibitors,” she said. “Perhaps understandably, rheumatologists are not the first or second specialists called upon during a pandemic. However, as a comparatively small group of providers who have relevant expertise and want to do what we can to help this effort, we’re being underutilized.”
This should not deter the rheumatology community from springing to action, according to Muskardin. “We need to insert ourselves into the groups of people who are designing protocols to help manage this subset of patients with COVID-19,” she said. “We have valuable input to give. While it is often unwise to give unsolicited advice, in this situation, folks may not realize that they should ask us, and concern for patients should outweigh any other possible reasons.”
- Cron RQ, Chatham WW. Don’t Forget the Host: COVID-19 Cytokine Storm. The Rheumatologist. Available at: https://www.the-rheumatologist.org/article/dont-forget-the-host-covid-19-cytokine-storm/. Published March 16, 2020. Accessed May 11, 2020.
- Gianfrancesco, M et al. Ann Rheum Dis. 2020;doi:10.1136/annrheumdis-2020-217871.
- Luo, P. et al. J Med Virol. 2020;doi:10.1002/jmv.25801.
- Monteagudo LA, et al. ACR Open Rheumatol. 2020;doi:10.1002/acr2.11135.
- Muskardin TLW. ACR Open Rheumatol. 2020;doi:10.1002/acr2.11140.
- Xu, X. et al. ChinaXiv. 2020;doi:ChinaXiv:202003.00026.
- Zhou F, et al. Lancet. 2020; doi:10.1016/S0140-6736(20)30566-3.
- For more information:
- Joseph Carcillo, MD, can be reached at 4401 Penn Ave., Pittsburgh, PA 15224; email: firstname.lastname@example.org.
- W. Winn Chatham, MD, can be reached at FOT 858, 1720 2nd Ave., South Birmingham, AL 35294-3408; email: email@example.com.
- Randy Q. Cron, MD, PhD, can be reached at 1825 University Blvd., Shelby Building, 306 UAB Birmingham, AL 35294; email: firstname.lastname@example.org.
- Michael B. Jordan, MD, can be reached at 240 Albert Sabin Way, ML 7038 Cincinnati, OH 45229-3039; email: email@example.com.
- Maximilian F. Konig, MD, can be reached at 5200 Eastern Ave., Mason F. Lord Building CT, Suite 4100, Baltimore, MD 21224; email: firstname.lastname@example.org.
- Theresa Wampler Muskardin, MD, can be reached at 333 East 38th St., 4th Floor New York, NY 10016; email: email@example.com.