Hemophagocytic lymphohistiocytosis (HLH) is a devastating, hyper-inflammatory condition that results in multi-organ failure and death. The systemic inflammation that characterizes the disease is the result of inappropriate and dysregulated activation of natural killer (NK) cells, CD8+ cytotoxic T-cells, and macrophages. The disease is classified as either primary (the result of inherited genetic mutations) or secondary (an inappropriate host response to infection, malignancy, or autoimmune disease). Patients with primary disease present early in childhood, whereas those with secondary disease present as adults with an associated acute illness, most commonly sepsis or a hematologic malignancy. Treatment is focused on immunosuppression coupled with cytotoxic chemotherapy, without which, large proportions of patients inevitably die.
HLH presents as two distinct clinical entities: primary or secondary
Accurate estimates of the prevalence and distribution of HLH within the population are difficult to obtain due to various factors, the most apparent of which is imprecise diagnostic criteria and the presence of multiple confounding medical illnesses at the time of diagnosis. The most comprehensive data with regards to primary HLH comes from a Swedish national registry that collected data from 1987-2006 and demonstrated a yearly incidence of roughly 1.5 per million. For children admitted to the intensive care unit (ICU), a retrospective review of 30,000 pediatric admissions at Texas Children’s Hospital estimated the incidence as high as 1 in 3,000.
In adults with secondary HLH, accurate estimates are even more difficult. Not only is the presentation often indistinguishable from the underlying trigger (sepsis, uncontrolled malignancy, autoimmune flare), but laboratory investigations to confirm HLH are infrequently undertaken. That being said, some estimates place it at 1 in 2000 for adult critical care admissions. Suspicion and institutional culture greatly affect the diagnostic workup and, therefore, the recorded incidence of the disease.
No clear racial or ethnic predilection has been observed in the literature, rather these categories tend to distribute in a fashion that approximates the surrounding geographic location of the study.  In the two largest epidemiologic studies in children, the gender distribution was approximately 1 to 1. In adults, however, there may be a slightly increased ratio of HLH in males compared to females of slightly less than 2 to 1.
HLH is characterized by a dysregulated innate immune system, specifically NK cells and CD 8+ cytotoxic T-cells. In a healthy immune system, NK cells and CD 8+ cytotoxic T-cells produce two cytolytic enzymes: perforin and granzyme. These proteins are packaged into granules that are discharged into the immunologic synapse between effector and target cell. Perforin forms destabilizing pores in the target cell's membrane allowing for the entrance of the strongly proteolytic granzyme as well as osmotic shifts, which result in target cell degradation. In patients with HLH, this process is disrupted either through specific genetic mutations or acquired through some highly immunogenic stimulus (a viral infected or malignant cell). The ineffective interaction between NK cells, CD 8+ cytotoxic T-cells, and their targets leads to a vicious cycle of inflammation. More and more cytotoxic cells are recruited but remain unable to rid the body of the pathologic antigen, and a massive increase in circulating cytokines occurs. This hypercytokinemia leads to widespread activation of macrophages and the resultant hemophagocytosis and excessive, organ-damaging, inflammation, which characterizes the disease.
The first genetic mutations identified in primary or familial HLH were located in the PRF1 gene, which encodes for perforin. Subsequently, several genetic mutations and associated syndromes have been identified with varying patterns of inheritability (recessive, autosomal, or X-linked). Although secondary HLH is classically understood as a purely acquired disorder, a large institutional series found genetic polymorphisms in primary HLH genes in 14% of adults, including in patients >70 years old.
Early research into HLH demonstrated high levels of serum soluble interferon 2 receptor (sIL-2R) in affected children. Additionally, levels correlated well with disease activity. (2786434) Since the discovery of sIL-2R, many other cytokines have been implicated in the pathophysiology of the disease, including interferon-gamma (INF-y), tumor necrosis factor (TNF), and IL-2. Limited success has been demonstrated in experiential studies of their related inhibitor antibodies.
Suspected cases of HLH typically undergo a biopsy of either lymph nodes, bone marrow (BM), or spleen. The historical histopathologic finding is hemophagocytosis in which macrophages are captured engulfing bone marrow cells. Although pathognomonic, this finding is not actually required for diagnosis. In fact, in patients who otherwise meet diagnostic criteria, pathologic review demonstrates hemophagocytosis present in only 67% of pediatric and 85% of adult cases.
Given the large clinical overlap between HLH and other systemic inflammatory conditions (severe infection, malignancy, and autoimmune disease) presenting symptoms are often non-specific signs of inflammation such as fever, malaise, and fatigue. Further blurring the distinction between HLH and other multi-organ inflammatory diseases is the fact that such disease processes often serve as the trigger for developing HLH. While children can have spontaneous HLH that is driven by genetic mutations, they can also develop HLH as the result of an antigenic stimulus or trigger, much in the same way it typically presents in adults. The most common trigger in children with primary HLH is an infection, particularly herpesvirus family infections, and even more specifically, Epstein Barr virus. In adults, the most common trigger is malignancy, accounting for about 45% of adult cases. Because of a relic of history rather than distinct pathophysiology, HLH associated with autoimmune diseases continues to be referred to as macrophage activation syndrome (MAS).
The physical exam is similarly protean. Many exam findings can be anticipated by remembering that HLH is a disease of hyper-reactive immune cells with particular attention paid to the reticuloendothelial system. To varying degrees, patients can present with cytopenias with associated bleeding, infection, or stressors related to extreme anemia (myocardial infarction, stroke, syncope) and with hepatosplenomegaly and/or diffuse adenopathy. Additionally, the severe systemic inflammation can present as altered mental status (including life-threatening meningoencephalitis), adult respiratory distress syndrome, acute liver failure, and acute renal failure.
HLH has had two major shifts in classification in its lifetime. The first major classification system was defined in the seminal trial known as HLH-94 (representing the year participants began enrolling), and the more recent definition is referred to as HLH-2004 (again representing the year participants were enrolled in the most recent therapeutic trial). In the last 5 years, attempts to redefine the diagnostic criteria have occurred but are not as validated as either of their predecessors. The original HLH-94 criteria were the following: 5 out of 5 of the criteria must be met to establish a diagnosis.
In the HLH-2004 criteria, 3 further laboratory findings were added to the list. Diagnosis is established by the presence of at least 5 out of the now combined 8 total criteria.
Notably, per the HLH-2004 criteria, a diagnosis no longer requires the presence of biopsy-proven hemophagocytosis.
The mainstay of treatment focuses on immunosuppression and cytotoxic therapy. Both of these are obviously contraindicated in patients with severe infection and underline the importance of arriving at a clear diagnosis prior to initiation. The landmark HLH-94 protocol was first developed and tested in an international collaborative study that enrolled and gathered data from 113 children. This protocol included a combination of dexamethasone, etoposide, cyclosporine A and intrathecal methotrexate (IT-MTX) in select patients followed by pre-planned bone marrow transplant (BMT). Outcomes of this trial demonstrated a 55% survival rate at 3 years and 5-year survival of 22%, both of which are quite favorable compared to the otherwise invariably rapid death in affected children.
The HLH-2004 protocol enrolled patients with both primary and secondary HLH and changed the previous protocol slightly by adding cyclosporine to the initial therapy cocktail as well as adding intrathecal steroids to IT-MTX for those with underlying neurologic dysfunction. Study analysis demonstrated no improvement in outcomes with the addition of upfront cyclosporine or intrathecal steroid therapy.
In both regimens, conditioning prior to BMT is achieved with reduced-intensity conditioning, such as fludarabine/busulfan.
Presenting symptoms are often non-specific or are thought to be accounted for in the context of known sepsis or malignancy. A higher degree of suspicion should be maintained in children who present with symptoms suggestive of Kawasaki disease or toxic shock syndrome. In adults, in which the disease is rarer, presentation is often obscured by another confirmed multi-organ system process such as malignancy, sepsis, or an autoimmune process. A high index of suspicion should be maintained in any patient who presents with multi-organ failure with associated cytopenias, coagulopathy, and failure to improve with standard treatment.
Clinicians will continue to refine the dosing, duration, and combination of immunosuppression and chemotherapy found in the HLH-2004 protocol. Survival will undoubtedly improve. However, progress will also (and is) being made in the realm of anti-cytokine agents. As mentioned previously, even the earliest of HLH research understood the central role of targeting hypercytokinemia. Various drugs have been developed or repurposed with specific cytokines or cytokine pathway targets, such as ruxolitinib (Janus kinase 1/2 inhibitor) and emapalumab (anti-TNF gamma).
Ruxolitinib, a medication currently approved for the treatment of myelofibrosis and polycythemia vera, has shown promise in murine models of HLH. Perforin deleted mice were treated with 1 mg/kg for 10 days and demonstrated improved survival, cytopenias, and serum levels of TNF-a and IL-6. Human trials have yet to been undertaken, however, given that it already has FDA approval for other hematologic diseases, such a transition could be expedited.
Phase II pediatric clinical trials have been performed with emapalumab, a novel human monoclonal anti-TNF gamma antibody. A trial of 13 children with refractory or worsening primary HLH, despite treatment, were administered the drug. Response (defined as a complete or partial improvement) occurred in 63% of treated children. 11 of 13 children were alive at 8 weeks.
Children affected by HLH will invariably die if left untreated, whereas adults can have spontaneous remission without recurrence. Admittedly, it is more difficult to obtain accurate estimates of adult fatality rates since it is often unclear whether deaths are the direct results of HLH or the underlying trigger (infection or malignancy). Retrospective studies have demonstrated a 3-year survival rate of 55% in children who received treatment with HLH-94. In children who survived to undergo BMT, 3-year survival rates reached 62%. Despite the fact that adults can have a spontaneous remission of HLH, the overall mortality rate remains quite high at roughly 41%. A recent multi-center case series in the United States demonstrated the poorest survival rates in patients with malignancy-associated HLH. The median time of survival was only 2.8 months (versus 10.7 months in those with non-malignancy-associated disease).
It seems redundant to speak of complications in a disease, which, in essence, is nothing but complications. HLH is, by definition, a multi-organ pathology with disease occurring indiscriminately as the result of widespread, dysregulated, immune system activity. Lungs can develop acute respiratory distress syndrome, hearts can be affected by myocarditis, and it’s myriad complications, kidneys can suffer microangiopathies, brains can succumb to fatal meningoencephalitis, livers can acutely fail, in fact, not a single major organ system is spared the devastating effects of this disease. The majority of patients who die from the disease die as the result of hemodynamic collapse.
Hemophagocytic lymphohistiocytosis is a devastating, hyper-inflammatory condition that results in multi-organ failure and death. Primary HLH occurs in children as the result of genetic mutations and secondary HLH in adults after exposure to a strongly immunogenic stimulus such as infection, malignancy, or autoimmune disease. Diagnostic criteria have limited sensitivity/specificity. However, aggressive treatment is required urgently in order to obtain favorable outcomes. Given the myriad of presenting symptoms and multi-organ involvement, patients with HLH are often treated in quaternary hospital ICUs with multiple subspecialties involved in care. A high index of suspicion, as well as good interprofessional communication, is required to establish a diagnosis and initiate appropriate treatment in a timely manner.
|||Al-Samkari H,Berliner N, Hemophagocytic Lymphohistiocytosis. Annual review of pathology. 2018 Jan 24 [PubMed PMID: 28934563]|
|||Morimoto A,Nakazawa Y,Ishii E, Hemophagocytic lymphohistiocytosis: Pathogenesis, diagnosis, and management. Pediatrics international : official journal of the Japan Pediatric Society. 2016 Sep [PubMed PMID: 27289085]|
|||Janka GE, Familial hemophagocytic lymphohistiocytosis. European journal of pediatrics. 1983 Jun-Jul [PubMed PMID: 6354720]|
|||[PubMed PMID: 26537747]|
|||[PubMed PMID: 26966089]|
|||[PubMed PMID: 26637718]|
|||[PubMed PMID: 24290661]|
|||[PubMed PMID: 3981309]|
|||[PubMed PMID: 25382070]|
|||[PubMed PMID: 19953651]|
|||[PubMed PMID: 24581757]|
|||[PubMed PMID: 2058392]|
|||[PubMed PMID: 9915956]|
|||[PubMed PMID: 21881043]|
|||[PubMed PMID: 27222478]|
|||[PubMed PMID: 28621800]|
|||[PubMed PMID: 25337215]|
|||[PubMed PMID: 12239144]|
|||[PubMed PMID: 28935695]|
|||[PubMed PMID: 8637226]|
|||[PubMed PMID: 16937360]|
|||[PubMed PMID: 20855862]|