Cytolytic pathway proteins mutated in MAS. A diagram of the immunologic synapse between a cytolytic (to the left) lymphocyte and an APC (to the right) is portrayed. Proteins involved in the cytolytic pathway that can be found mutated in MAS and HLH include Rab27a, Munc13-4, syntaxin 11, Munc18-2, and perforin (Reproduced with permission from Ravelli et al. )
These mutations cause a severe impairment of cytotoxic function of NK cells and cytolytic T lymphocytes. However, the mechanisms that link the defect in cytotoxic cell functions with the expansion and proliferation of activated macrophages are not clear. One possible explanation is that deficient cytolytic activity may lead to decreased ability to control some infections. NK cells and cytotoxic T lymphocytes may fail to kill infected cells and, thus, to remove the source of antigenic stimulation. Continued antigen stimulation might lead, in turn, to persistent antigen-driven activation and proliferation of T cells accompanied by rising production of cytokines that stimulate macrophages. Abnormal cytotoxic cells may also be unable to provide proper apoptotic signals for removal of activated macrophages and T cells during the final stage of some immune responses, which contributes to persistent expansion of T cells and macrophages secreting pro-inflammatory cytokines [32–34]. As a result of continuous stimulation with pro-inflammatory cytokines, particularly interferon gamma (IFN-γ), macrophages become hemophagocytic [35, 36]. Activated T cells and histiocytes infiltrate all tissues and lead to tissue necrosis and organ failure. Most recently, delayed killing by defective NK cell and CD8 T cells deficient in perforin or granzyme B has been shown to prolong the interaction between the lytic cell and the target cell, resulting in a pro-inflammatory cytokine storm .
Recent evidence has been provided that, as in FHLH, patients with sJIA-associated MAS may also have functional defects in the exosome degranulation pathway [38–40]. The same biallelic mutations in the MUNC13-4 gene reported in FHLH have been found in monoallelic form in some patients with MAS . With the assumption that MAS predisposition in sJIA could be attributed to rare gene sequence variants affecting the cytolytic pathway, Kaufman et al. performed whole-exome sequencing in 14 patients with sJIA and MAS and in their parents . The authors identified rare variants not only in genes associated with primary hemophagocytic syndromes (LYST, MUNC13-4, and STXBP2) but also in a large number of genes encoding proteins involved in intracellular vesicle transport. According to these findings, it is reasonable to hypothesize the existence of some genetic overlap between MAS and FHLH. Similarly, there appears to be a high percentage of patients with single-copy mutations in cytolytic pathway/FHLH-associated genes in children with MAS both associated with sJIA as well as other disorders . Indeed, recent evidence suggests that some of these monoallelic HLH gene mutations partially disrupt cytolytic activity of NK cells and CD8 T cells via a dominant-negative effect and, thus, predispose individuals to developing MAS [43, 44].
In spite of the compelling evidence for a pathogenetic role of defects in cytotoxic cell function in FHLH, in many instances of sJIA-associated MAS no such defects have been identified, or have been found to have only variable penetrance. This has led to a search for alternative causative mechanisms. Recently, a murine model of MAS induced by repeated stimulation of toll-like receptor (TLR)-9 provided important insight . It suggested that situations of repeated activation of TLR-9 may replicate the environment that allows MAS to develop in a genetically wild-type host. Intriguingly, certain aspects of the disease in this model were IFN-γ dependent, establishing a connection to FHLH . A prominent role of IFN-γ in HLH was previously demonstrated by the observation that in perforin-deficient mice only the antibody directed to IFN-γ, and not to other cytokines, prolonged survival and prevented the development of histiocytic infiltrates and cytopenia .
Recent findings suggest that MAS, or at least some of its forms, could be considered as part of the monogenic auto-inflammatory disease spectrum. A 7-year-old European female with recurrent episodes characterized by leucopenia, chronic anemia, thrombocytopenia, elevation of acute-phase reactants and transaminases, hypertriglyceridemia, and hyperferritinemia (all typically MAS biomarkers) has been recently reported . The girl was found to have a mutation in the nucleotide-binding domain of the inflammasome component, NLRC4, associated with high levels of the inflammasome-dependent cytokines interleukin (IL)-1β and IL-18. The NK cell function was normal and the genetic testing for HLH and periodic fever syndromes were negative. The administration of the recombinant human IL-1 receptor antagonist, anakinra, led to clinical improvement, although serum levels of IL-18 did not normalize after treatment.
The end result of this interaction between genetic predisposition and defective cytotoxic and cytolytic pathway is a pro-inflammatory cytokine “storm,” induced by cytokines secreted by macrophages [IL-1β, tumor necrosis factor (TNF), IL-6, and IL-18] and T lymphocytes (IFN-γ and IL-2) and independent of the different pathogenic triggering factors . The implication of these cytokines in the pathogenesis of MAS provides the rationale for the modulation of their axes in the management of the syndrome (Fig. 22.2).
Schematic representation of the pathophysiology of MAS. MAS can develop in the setting of active SJIA, which is associated with increased cytokine levels, including IL-1, IL-6, IL-18, and TNF-α. MAS can also be triggered by viral infections, through the recognition of pathogen-associated molecular patterns (PAMPs) by toll-like receptors (TLRs), which, in turn, induces hypersecretion of inflammatory cytokines. The increased secretion of IFNγ leads to activation of macrophages that acquire apro-inflammatory phenotype and release high levels of chemokines and cytokines
Role of Cytokines
Several biomarkers reflect the degree of activation and expansion of T cells and macrophages and can help to identify subclinical MAS in patients with sJIA.
The role of IL-1 in sJIA has been well described, but its involvement in the pathogenesis of MAS is not clearly defined. Patients with primary HLH present with high levels of IL-1 , but the progression of the disease seems to be independent of the IL-1 signaling pathway . In addition, patients with MAS do not present the typical clinical features of IL-1-mediated diseases, such as cryopyrin-associated periodic fever syndromes.
IL-6 is another pivotal cytokine of the acute-phase response and it is thought to play an important role in the development of MAS in the presence of an infectious trigger . A study of hepatic biopsies in patients with various types of HLH, including MAS, revealed extensive infiltration of the liver by hemophagocytic macrophages secreting IL-6 . In IL-6 transgenic mice, prolonged exposure to high levels of IL-6 seemed to lower the threshold needed to trigger an episode of MAS . This is perhaps because in addition to amplifying a pro-inflammatory environment , IL-6 may lower the cytolytic capacity of NK cells. Nevertheless, the role of IL-6 in the pathogenesis of MAS remains poorly understood.
TNF is associated with the pathogenesis of the inflammatory process in chronic arthritis, but recent findings were less than enthusiastic for its major impact in the development of MAS. Shimizu et al. analyzed the serum concentration of TNF receptor types I (sTNF-RI) and II (sTNF-RII) in five serum samples of patients with sJIA-associated MAS . The authors found a rise in the concentrations of both receptors after remission of MAS, which suggests that the presence of high levels of TNF-α during the clinically silent phase of MAS may be correlated with a generic cellular activation.
The serum levels of soluble IL-2 receptor α (sIL-2Rα, also known as CD25) and soluble CD163 (sCD163, also known as scavenger receptor cysteine-rich type 1 protein, M130), which reflect the degree of activation and expansion of T cells and phagocytic macrophages, respectively, are valuable diagnostic markers for MAS and may help to identify patients with subclinical forms [18, 19, 53]. In seven patients evaluated in the acute phase of sJIA-associated MAS, the levels of these two biomarkers were comparable to those previously reported in patients with primary and secondary forms of HLH and significantly higher than those detected in 16 patients with new-onset, untreated sJIA . In patients with MAS, the levels of both biomarkers returned to the normal range after resolution of the acute phase . Five of the patients with sJIA had sIL-2Rα and CD163 levels comparable to those found in patients with acute MAS. These patients also had low platelet counts and high ferritin levels (both of which can be present in MAS), and two of them later developed full-blown MAS .
Another interesting observation that came out from recent studies regards the crucial regulatory role of IL-10 in controlling the MAS process. Mice given repeated TLR-9 stimulation, coupled with blockade of the IL-10 receptor, developed a more fulminant disease. Notably, polymorphisms of IL-10 associated with decreased function of the cytokine have also been associated with sJIA [54, 55]. It has been suggested that the combined immunologic abnormality of hyperactive TLR/IL-1β signaling and decreased IL-10 function may result in a predisposition to MAS . These findings may lead to the speculation that patients with occult or subclinical MAS may have better IL-10 function than those developing full-blown MAS. In the former patients, the enhanced IL-10 function would help to maintain a state of relative quiescence.
It has been further suggested that the hyper-production of IL-18 (which strongly induces Th-1 responses and IFN-γ production and enhances NK cell cytotoxicity) and an imbalance between the levels of biologically active free IL-18 and those of its natural inhibitor (the IL-18 binding protein) may play a role in secondary hemophagocytic syndromes, including MAS [57, 58]. Two distinct subsets of patients with sJIA have been defined on the basis of their serum cytokine profiles as either IL-6-dominant or IL-18-dominant . The IL-6-dominant subset had a significantly greater number of active joints and higher serum levels of MMP-3 than the IL-18-dominant subset, whereas the IL-18-dominant subset had an increased risk of developing MAS. This finding led the authors to suggest that two subsets of patients with sJIA, one prone to arthritis and the other to MAS, can be identified on the basis of their serum IL-6 and IL-18 levels.
In 28 patients with sJIA (including seven who developed MAS), serum levels of follistatin-related protein 1, a glycoprotein overexpressed in certain inflammatory diseases, were markedly elevated during acute MAS and returned to normal following treatment . Patients with newly diagnosed sJIA who had elevated serum levels of this protein had dysregulated expression of genes involved in innate immunity, erythropoiesis and NK cell function, suggesting a pattern of gene expression similar to that of MAS. Indeed, two of these patients ultimately developed MAS. On the basis of these findings, the authors speculated that elevated serum levels of follistatin-related protein 1 at the onset of sJIA might predict progression to overt MAS. In the same study, the ferritin-ESR ratio revealed superior sensitivity and specificity of ferritin levels alone for the differentiation of overt MAS from new-onset sJIA.
In summary, high levels of circulating cytokines and cytokine inhibitors in patients with MAS have been reported by many studies. These include cytokines derived from lymphocytes, such as IFN-γ and IL-2, as well as cytokines of monocyte and macrophage origin, including IL-1, TNF, IL-6, and IL-18. The huge hypersecretion of pro-inflammatory cytokines that characterizes MAS has led some authors to use term “cytokine storm” to define the pathophysiology of the syndrome.
Clinical, Laboratory, and Histopathologic Features
The clinical presentation of MAS is generally acute and occasionally, dramatic, requiring the admission of the patient to the ICU. The onset of the syndrome is usually heralded by the sudden occurrence of non-remitting high fever, profound drop in all three blood cell lines (leukopenia, anemia, and thrombocytopenia), liver enlargement, splenomegaly and generalized lymphadenopathy, and increase in serum liver enzymes. There is often an abnormal coagulation profile, with prolongation of prothrombin and partial thromboplastin times, hypofibrinogenemia, detectable fibrin degradation products, and increase in D-dimers. As a result, patients may have purpura, easy bruising, and mucosal bleeding. High concentrations of triglycerides and lactic dehydrogenase and low sodium levels are observed frequently. The acute phase of MAS is usually marked by a sharp rise of ferritin, often above 5000–10,000 ng/ml. Measurement of serum ferritin level may assist in the diagnosis of MAS and is a useful indicator of disease activity, therapy response, and prognosis . Central nervous system dysfunction is seen in around one-third of cases and may cause lethargy, irritability, disorientation, headache, seizures, or coma. Renal, pulmonary, and cardiac involvement may develop in the sickest patients, who progress to develop multi-organ failure.
In children with sJIA, the clinical picture of MAS may mimic sepsis or a flare of the underlying disease. However, the pattern of non-remitting fever is different from the remitting high-spiking fever typical of sJIA flare. Moreover, patients may show a paradoxical improvement in the underlying inflammatory disease at the onset of MAS, with disappearance of signs and symptoms of arthritis and a fall in the ESR. The latter phenomenon is mainly related to the hypofibrinogenemia secondary to fibrinogen consumption and liver dysfunction and helps explain the utility of the ferritin-ESR ratio. A characteristic feature of the syndrome may be seen on bone marrow examination, which reveals numerous morphologically benign macrophages exhibiting hemophagocytic activity (Fig. 22.3). Such cells may infiltrate the lymph nodes and spleen as well as many other organs in the body and may be responsible for several clinical manifestations of the syndrome. However, in patients with MAS, the bone marrow aspirate does not always show hemophagocytosis (present in approximately 60 % of patients), and, in a mouse model of fulminant MAS, hemophagocytosis was only revealed when blocking IL-10, a finding that suggested that IL-10 responsiveness can change the spectrum of disease severity, including the presence of hemophagocytosis . Therefore, failure to reveal hemophagocytosis does not exclude the diagnosis of MAS. It is still unclear whether MAS is a discrete clinical event or whether it represents the most severe end of the spectrum of disease activity in sJIA. Since some sJIA patients develop MAS repeatedly and others never develop MAS, there may be a genetic predisposition (perhaps cytolytic defects) for up to half of the sJIA patients to be at risk for MAS . Table 22.1 shows the main clinical and laboratory features of MAS complicating sJIA. The comparison of the features of active sJIA and MAS is presented in Table 22.2.
Bone marrow aspirate showing macrophage hemophagocytosis in a patient with systemic juvenile idiopathic arthritis and macrophage activation syndrome
Main features of macrophage activation syndrome
Central nervous system dysfunction
Abnormal liver function tests
Decreased erythrocyte sedimentation rate
Macrophage hemophagocytosis in the bone marrow
Comparison of clinical and laboratory features of active systemic juvenile idiopathic arthritis (sJIA) and macrophage activation syndrome (MAS)
Petechial or purpuric
White blood cells and neutrophils
Normal or ↓
Erythrocyte sedimentation rate
Normal or ↓
Normal or ↑
Normal or ↑
Partial thromboplastin time
Normal or ↑
Soluble interleukin-2 receptor α
Normal or ↑
Normal or ↑
Although most instances of MAS lack an identifiable precipitating factor, the syndrome has been associated with a number of triggers, including a flare of the underlying disease; toxicity of aspirin or other nonsteroidal anti-inflammatory drugs; viral, bacterial, or fungal infections; a second injection of gold salts; or a side effect of second-line or biologic medications .
MAS episodes are reported most frequently during a flare of active underlying sJIA, and, in about 20 % of patients, the syndrome occurs at the time of sJIA onset . However, it is well known that MAS may be incited by an infection, particularly by members of the herpesvirus family. Epstein-Barr virus is the most common causative viral agent, but virtually any infectious agent can precipitate the development of MAS.
In several reports, the trigger of MAS has been related to modifications in drug therapy. A young girl with sJIA was described who developed MAS 24 h after the first methotrexate administration . The shortness of the time interval between MTX dosing and onset of MAS and the characteristics of clinical symptoms (particularly the intense and generalized itching) argued for a hypersensitivity or idiosyncratic reaction, a mechanism similar to that hypothesized in the pathogenesis of MAS secondary to gold salt injections. Considering the large number of medications associated with MAS, it is reasonable to suppose a lack of direct causality in many cases. Additional therapies are often required in patients with severe underlying rheumatic disease, which means that the susceptibility to MAS may be increased by the exacerbation of disease activity, rather than by medication toxicity. In the recent years, instances of MAS in sJIA patients during treatment with biologic medications have been described. However, the role of these drugs in the induction of MAS is controversial, as published experience for the use of IL-1 blocking therapy for the treatment of refractory sJIA-associated MAS has been favorable. In fact, increasing the dose of IL-1 blockade typically helps to control MAS in children with sJIA .
Severe episodes of MAS have been observed among patients who had undergone autologous bone marrow transplantation for sJIA refractory to conventional therapies [64, 65]. Although in most of these cases an infectious trigger for MAS was identified, it was hypothesized that the complication was due to stringent T cell depletion, with resultant inadequate control of macrophage activation. After an adaptation of the protocol, consisting in less profound T cell depletion, better control of systemic disease before transplantation, and slow tapering of corticosteroids after the procedure, no further cases of MAS have been reported.
MAS can be difficult to distinguish from several conditions. However, an early diagnosis is essential to select the appropriate therapeutic interventions in a timely manner. Some of the MAS features, such as lymphadenopathy, splenomegaly, and hyperferritinemia, are common manifestations of active sJIA and may not differentiate MAS from a conventional sJIA flare. Because many features are shared between MAS and cSLE (e.g., cytopenia, splenomegaly, fever), MAS is also probably an underdiagnosed complication of cSLE . When a patient with cSLE presents with unexplained fever and cytopenia, MAS should be considered and investigations, including measurement of ferritin level, should be carried out.
Because MAS bears close clinical similarities with FHLH or acquired virus-associated HLH (VA-HLH), its differentiation from these conditions may be challenging, particularly when MAS develops at onset of sJIA in the absence of arthritis. Onset at a very young age, positive family history, and more profound cytopenias are clues to differentiate FHLH from MAS. Notably, a recent comparative analysis showed that neutrophil count and CRP were significantly higher in patients with MAS than in patients with FHLH/VA-HLH. Furthermore, a soluble CD25 level <79,000 U/l indicated MAS, rather than FHLH/VA-HLH .
Other clinical entities associated with hepatic dysfunction, coagulopathy, cytopenias, or encephalopathy must also be considered in the differential diagnosis of MAS. In some sJIA patients, the combination of hepatic dysfunction with encephalopathy may be due to Reye syndrome: a viral prodrome, unexplained vomiting, behavioral changes and increasing of serum aminotransferase, ammonia level, and prothrombin time, coupled with minimal change in serum bilirubin, are all important features of this illness . The hemorrhagic symptoms seen in MAS may resemble those seen in thrombotic thrombocytopenic purpura (TTP), but the microangiopathic anemia with the finding of fragmented red blood cells in the peripheral blood smears, which is characteristic of TTP, is not typically seen in MAS. However, it should be taken into account that TTP may occasionally complicate the most severe instances of MAS. It is also important to note that MAS is not a diagnosis of exclusion but rather an end common pathway/complication of many hyper-inflammatory states.
Intracellular bacterial infections and zoonotic diseases frequently presenting with hepatosplenomegaly and leukopenia, such as Brucella, Rickettsia spp, and visceral leishmaniasis, bear a close resemblance to MAS . The clinical and laboratory picture of visceral leishmaniasis may be indistinguishable from that of MAS patients. Leishmania donovani and Leishmania infantum infections by themselves can cause secondary forms of HLH. A bone marrow aspirate, with a microscopic direct examination of Leishmania spp in the hematopoietic cells and in vitro culture of bone marrow aspirates, helps to make the correct diagnosis. Throughout the world, it is also important to consider epidemics caused by hemorrhagic fever viruses (e.g., dengue, Crimean-Congo, and possibly Ebola) as causes of potentially treatable MAS .
The difficulties in making the diagnosis, the recent therapeutic progress, and the advances in understanding MAS pathophysiology highlight the need for diagnostic tools and well-established diagnostic guidelines. Diagnostic criteria would also be useful for research purposes and use in literature reports.
The recognition that MAS shares many similarities with HLH has led some clinicians to advocate the use of the HLH diagnostic guidelines (Table 22.3) to diagnose this syndrome in children with sJIA . However, HLH criteria are not necessarily suitable for use in identifying MAS in sJIA. Due to the prominent inflammatory nature of sJIA, the occurrence of a relative drop in white blood cell count, platelets, or fibrinogen rather than the absolute decrease required by the HLH criteria may be more useful to make an early diagnosis. Furthermore, the threshold level for hyperferritinemia required for the diagnosis of HLH (500 ng/ml) may not discriminate MAS from a flare of sJIA. It is well known that many patients with active sJIA, in the absence of MAS, have ferritin levels above than threshold . In the acute phase of MAS, ferritin levels may peak to greater than 5000 ng/ml. Other HLH criteria that are not readily applicable to MAS are the demonstration of low or absent natural killer cell activity or soluble IL-2 receptor α chain above normal limits for age. These tests are not routinely performed in most pediatric rheumatology centers.
HLH-2004 diagnostic guidelines
1. Molecular diagnosis
2. Diagnostic criteria
Cytopenia (affecting two of three lineages in the peripheral blood):
Hemoglobin <90 g/l
Platelets <100 × 109/l
Neutrophils <1.0 × 109/l
Hypertriglyceridemia and/or hypofibrinogenemia:
Triglycerides ≥265 mg/dl
Fibrinogen ≤1.5 g/l
Hemophagocytosis in bone marrow, spleen, or lymph nodes
Low or absent NK cell activity
Ferritin ≥500 ng/ml
Soluble CD25 ≥2400 U/ml
In 2005, preliminary diagnostic guidelines for MAS complicating sJIA were published  (Table 22.4). These guidelines have the merit of being data-driven and not merely based on expert consensus. However, the study that led to their development had several limitations, including the lack of several laboratory measurements in a number of patients and insufficient data for some of the laboratory parameters evaluated. In addition, the criteria have yet to be validated.
Preliminary diagnostic guidelines for macrophage activation syndrome (MAS) complicating systemic juvenile idiopathic arthritis (sJIA)
1. Decreased platelet count (≤262 × 109/l)
2. Elevated levels of aspartate aminotransferase (>59 U/l)
3. Decreased white blood cell count (≤4.0 × 109/l)
4. Hypofibrinogenemia (≤2.5 g/l)
1. Central nervous system dysfunction (irritability, disorientation, lethargy, headache, seizures, coma)
2. Hemorrhages (purpura, easy bruising, mucosal bleeding)
3. Hepatomegaly (≥3 cm below the costal arch)
Evidence of macrophage hemophagocytosis in the bone marrow aspirate
The diagnosis of macrophage activation syndrome requires the presence of at least two laboratory criteria or the presence of at least one laboratory criterion and one clinical criterion. A bone marrow aspirate for the demonstration of hemophagocytosis may be required only in doubtful cases
A recent study compared the performance of HLH-2004 diagnostic guidelines and preliminary MAS diagnostic guidelines in differentiating sJIA-associated MAS from two confusable conditions, represented by active sJIA without MAS and febrile systemic infections, in a large patient sample collected in a multinational survey . The authors found that the preliminary MAS guidelines achieve the best trade-off between sensitivity and specificity in differentiating MAS from sJIA flare and febrile systemic infections. The poorer performance of the HLH-2004 guidelines was mainly explained by the low frequency of the items cytopenia and hypofibrinogenemia in patients with MAS, mainly due to their too stringent thresholds. It is well known that patients with sJIA often have increased white blood cell and platelet counts as well as increased serum levels of fibrinogen as part of the underlying inflammatory process. Thus, the occurrence of a relative decline in these laboratory parameters, rather than the absolute decrease required by the HLH-2004 guidelines, can be more useful to make an early diagnosis of MAS.