The perfect storm

In this issue of Blood, Tsoukas et al report that 2 distinct defects in immune pathways, cytotoxicity, and autoinflammation can synergize to produce a spontaneous hemophagocytic lymphohistiocytosis (HLH). This observation significantly advances our pathogenic model for understanding how and why HLH presents in certain patients. This in turn opens new avenues for addressing these pathways diagnostically and therapeutically.¹ 

HLH is a clinical syndrome that results in severe inflammation, multiorgan failure, and often death; it is characterized by excessive production of inflammatory cytokines, such as interferon γ. Our understanding of the pathophysiology of HLH, a prototypical cytokine storm, has allowed for the development of rational treatment protocols that have made dramatic advances in improving patient outcomes. In 1999, it was recognized that HLH was associated with defects in perforin function in hematopoietic cells.²  Thus, bone marrow transplantation became the obvious path to effecting a clinical cure. Subsequently, other genetic defects were discovered, with the common theme that they all affect perforin function, leading to defective killing. In 2004, it was first reported in Blood that defective perforin cytotoxicity resulted in the inability of CD8 T cells to kill their targets.³  This leads to excessive activation of CD8 T cells and causes them to overproduce interferon γ, which in turns leads to the cytokine toxicity. Fifteen years later, the first anti–interferon γ neutralizing antibody was approved for use in HLH.

At the same time that perforin deficiency was being recognized as a driver of HLH in the world of hematology/oncology, rheumatologists were seeing a similar cytokine storm syndrome they labeled macrophage activation syndrome (MAS). MAS bears a striking similarity to HLH, but these patients did not have the profound genetic defects in perforin pathways associated with HLH. Instead, the disease seemed to be associated with autoinflammation, particularly with elevations in the interleukin-1 (IL-1) family member cytokine IL-18.^4,5  Because IL-18 is an interferon γ–driving cytokine, this allowed for the retention of a similar pathogenic model centered on interferon γ; however, instead of perforin defects being the proximal mediator, the upstream event was overproduction of IL-18, either by genetic (NLRC4, XIAP, CDC42) causes or for as yet unknown reasons.

Although these 2 means to an interferon γ cytokine storm have been largely considered separately in studies, Tsoukas et al now show that these 2 pathways synergize to produce an even more fulminant storm. Perforin deficiency is typically insufficient to produce spontaneous HLH, requiring an infection to set off the disease. Similarly, transgenic overproduction of IL-18 in mice does not produce MAS. However, when a mouse is generated that has both defects, spontaneous HLH occurs, driven by interferon γ and CD8 T cells. Perhaps most striking, even mice that are only haploinsufficient for perforin, but also overproduce IL-18, develop hyperinflammatory disease.

It has been increasingly recognized that patients with unexplained HLH/MAS may have an increased burden of heterozygous genetic defects in cellular cytoxicity genes. However, such defects are reasonably common in healthy populations and thus insufficient to explain the development of HLH by themselves. The work of Tsoukas et al now provides a model that can add clarity to this apparent conundrum, suggesting that in the setting of heterozygous defects in perforin, excess IL-18 can tip the precarious balance of the immune system from homeostasis toward a cytokine storm. IL-18 is a clinically measurable cytokine; therefore, this information could become important diagnostically to determine HLH/MAS status and may have therapeutic implications, because interferon γ blockade is already available, and IL-18 blockade has been described in single patients⁶  and has an ongoing clinical trial (registered at www.clinicaltrials.gov as #NCT03113760).

It is interesting to speculate that there may be other cytokines or defects that can combine in the same synergistic manner to contribute to total HLH/MAS risk. Other IL-1 family members have been implicated, including IL-1β⁷  and IL-33.⁸  Perhaps heretofore unrecognized immune pathways also play a role in contributing. Continuing to add to the model of independently small insults that can synergize to produce large pathologic effects in a cytokine storm will be an important area of investigation in driving both diagnostic and therapeutic improvements in the care of HLH/MAS.