COVID-19 and the Slippery Road of Cytokine Storms Save

Early reports on the COVID-19 pandemic allude to a cytokine storm or hyper-inflammation, especially in patients who succumbed to the illness. These conclusions are based on elevated levels of ferritin and other inflammatory markers, drawing comparisons to rare diseases of uncontrolled systemic inflammation such as hemophagocytic lymphohistiocytosis (HLH), and macrophage activation syndrome (MAS).

HLH is a rare congenital immune disorder resulting in uncontrolled, systemic activation of macrophages. The activated macrophages in turn attack vital organs such as blood, liver, and brain.

A similar phenomenon occurs in patients with underlying rheumatologic diseases, and is referred to as MAS. Clinical manifestations of HLH and MAS include fever, splenomegaly, pancytopenia, hepatitis, hypofibrinogenemia, and elevated inflammatory markers, such as ferritin and soluble CD25 (IL-2 receptor alpha). Left untreated, HLH/MAS can be rapidly fatal, making early diagnosis and treatment important. The hallmark of these disorders is a surge of inflammatory cytokines that is unwarranted or disproportionate to a stimulus. The ensuing inflammation is then the main culprit responsible for organ damage, and anti-inflammatory treatments are thus life-saving.

Inflammation in response to an infection, though, is normal, necessary, and protective. Elevated ferritin, no matter how high, is however not sufficient to distinguish between disproportionate and appropriate levels of inflammation. The high ferritin observed in COVID19 patients might reflect a suitable inflammatory response to the virus, and not the hyper-inflammation of HLH/MAS. A significant proportion of patients requiring ICU care exhibit high levels of ferritin, including those with ARDS. Besides being a marker of inflammation, ferritin is also released by dying cells. Thus, higher levels of serum ferritin in sicker patients might simply indicate more organ damage.

In addition to ferritin, the diagnosis of HLH/MAS is suspected, though not confirmed (as is mistakenly believed), by the fulfillment of entry criteria for the HLH-2004 clinical trial (any five of fever, splenomegaly, cytopenia, low fibrinogen, high triglycerides, elevated sCD25, reduced NK-cell function, hemophagocytosis), or by a high H-score (which utilizes similar features).

The HLH-2004 criteria or high H-score are not the definition of HLH, nor are these specific to or diagnostic of HLH/MAS, and were meant to be applied in the absence of other conditions such as infections or malignancy. Fever and leukopenia are reported in a significant majority of COVID19 patients, even in the absence of high ferritin, and are thus likely attributable to the viral infection itself. This highlights the substantial confusion around HLH/MAS, especially in adults.

Over the last few years, several patients (children and adults) were referred to us after receiving treatment for HLH because their condition fulfilled the HLH-2004 criteria. We instead found them to suffer from malignancies (lymphoma, leukemia) or infections (histoplasmosis, HSV, adenovirus, and others). Unfortunately, by the time we discovered the real disorder, the malignancy or infection had inflicted irreversible damage to vital organs, and most of the patients did not survive.

We now have a better understanding of HLH, and have more reliable assays to quickly and accurately diagnose it, so we don't rely solely on the non-specific HLH-2004 diagnostic criteria. Alarmingly however, with increasing frequency, using these outdated and non-specific criteria, not only children, but adults are being diagnosed with and being treated for HLH, all without any evidence that the disease entity occurs in adults, and without any indication that pediatric treatment is effective or even safe in adults. Even in children, lymphoma and infection are more common than HLH, and are thus likely to be misdiagnosed by the blind application of these criteria.

Treatment for HLH/MAS consists of corticosteroids and Etoposide, which alleviate inflammation due to any condition, but can actually make infections worse. Newer treatments for MAS/HLH like anakinra, tocilizumab, ruxolitinib, and emapalumab block inflammatory cytokine signaling, and might be safer, but we don't have enough evidence yet that these newer treatments are safe in the presence of an active infection.

When HLH is diagnosed and treated in patients who actually suffer from leukemia or lymphoma, the problem is worse, since HLH-directed therapy can induce partial remissions, giving a false sense of security. Once treatment is initiated, diagnosing the underlying leukemia or lymphoma becomes near impossible. The inflammation that accompanies infections or malignancies is labeled "secondary HLH," to differentiate it from the congenital disorder. This only furthers the confusion, since the label of HLH becomes front and center, a diagnosis in and of itself. In addition to incorrect treatment, the diagnosis of HLH in these situations leads to premature closure of the search for the underlying disease that is causing an illness that resembles HLH.

Unquestionably, certain infections are associated with excessive inflammation that adds insult to injury, and anti-inflammatory treatments like corticosteroids are indeed helpful in these situations. Hyper-inflammation as a cause of morbidity and mortality in COVID-19 is certainly possible, but we don't have enough evidence yet to label the inflammation as disproportionate solely on the basis of hyperferritinemia, let alone blame it as a cause or contributor to the pathology of the illness.

In COVID-19, we are facing an unprecedented health crisis, and need a solution, quickly. As described by Nobel laureate Daniel Kahneman, fast thinking is however not always rational, and we jump to conclusions that appear quickly and easily, without considering the true likelihood or probability of the conclusion. The first diagnosis arising in the clinician's mind that readily fits the patient's condition becomes the most likely one, relegating or even discarding other more probable ones as unlikely, inconceivable, or even absurd. In the absence of a known explanation for the patient's illness, we search for disease patterns that are familiar, and become vulnerable to seeing such patterns even when they don't exist. No sooner do we fall for the illusion, we get anchored to it. Any additional evidence is then perceived through the lens of the accepted diagnosis or explanation.

We desperately need to find effective treatments for COVID-19. However, we will need to wait on data from controlled clinical studies to decide on safe and effective therapies, to avoid making lethal errors. Simply having an elevated ferritin or even fulfilling the HLH-2004 criteria is not sufficient to diagnose a patient with hyper-inflammation or HLH. Adults diagnosed with HLH usually have something else – something potentially treatable, but something deadly.

Ashish Kumar, MD, PhD, is professor of pediatrics and director of the Langerhans Cell Histiocytosis Center and Pediatric Hematology/Oncology Fellowship Program at Cincinnati Children's Hospital

References

1. Zhou F, Yu T, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 2020;395(10229):1054-1062.

2. Ruan Q, Yang K, et al. Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China. Intensive Care Med 2020.

3. Mehta P, McAuley DF, et al. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 2020;395(10229):1033-1034.

4. Michael B. Jordan M. Emergence of Targeted Therapy for Hemophagocytic Lymphohistiocytosis. March-April 2018.

5. Lachmann G, Knaak C, et al. Hyperferritinemia in Critically Ill Patients. Crit Care Med 2020; 48(4):459-465.

6. Connelly KG, Moss M, et al. Serum ferritin as a predictor of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 155(1):21-25.

7. Sharkey RA, Donnelly SC, et al. Initial serum ferritin levels in patients with multiple trauma and the subsequent development of acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 159(5 Pt 1):1506-1509.

8. Kell DB, Pretorius E. Serum ferritin is an important inflammatory disease marker, as it is mainly a leakage product from damaged cells. Metallomics 2014; 6(4):748-773.

9. Henter JI, Horne A, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007;48(2):124-131.

10. Fardet L, Galicier L, et al. Development and validation of the HScore, a score for the diagnosis of reactive hemophagocytic syndrome. Arthritis Rheumatol 2014;66(9):2613-2620.

11. Gurunathan A, Boucher AA, et al. Limitations of HLH-2004 criteria in distinguishing malignancy-associated hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2018;65(12):e27400.

12. Kahneman D. Thinking, Fast and Slow. Farrar, Straus and Giroux 2011