Immunotherapy of Familial Hemophagocytic Lymphohistiocytosis With Antithymocyte Globulins: A Single-Center Retrospective Report of 38 Patients
OBJECTIVES. Familial hemophagocytic lymphohistiocytosis is a genetically determined condition that is characterized by unremitting CD8 T lymphocyte and macrophage activation and leads to death in the absence of therapy. On the basis of the immunologic pathophysiology of familial hemophagocytic lymphohistiocytosis, we propose a therapy with a combination of antithymocyte globulins with corticosteroids, cyclosporin A, and intrathecal injections of methotrexate.
METHODS. We retrospectively analyzed the outcome of antithymocyte globulin–based therapy that was performed in 38 consecutive patients who had familial hemophagocytic lymphohistiocytosis and were treated in a single center between 1991 and 2005. Overall, they received 45 courses of antithymocyte globulin (5–10 mg/kg per day for 5 days).
RESULTS. This regimen was associated with infections after 10 of 45 courses of antithymocyte globulin. There were 6 events after 11 antithymocyte globulin courses given as second-line therapy against 4 after 34 antithymocyte globulin courses in patients who were treated primarily with antithymocyte globulin. Antithymocyte globulin administration led to rapid and complete response of familial hemophagocytic lymphohistiocytosis in 73% of cases, partial response in 24%, and no response only once. When hematopoietic stem cell transplantation was performed early after complete or partial response induction, it led to a high rate of cure, in 16 of 19 cases. Overall survival was 21 of 38 with 4 toxic deaths.
CONCLUSION. Antithymocyte globulin based immunotherapy of familial hemophagocytic lymphohistiocytosis is efficient and carries an acceptable toxicity when used as a first treatment of familial hemophagocytic lymphohistiocytosis.
Familial hemophagocytic lymphohistiocytosis (FHLH) is a rare, genetically determined condition caused by gene mutations that impair T/NK granule-dependent lymphocyte cytotoxic activity.1–5 Mutations of perforin- (PRF1), hMunc13.4- (UNC13D), and, more recently, Syntaxin 11- (STX11) encoding genes have been incriminated in FHLH.5–7 The condition is characterized typically by an early onset of hepatosplenomegaly, fever, neurologic symptoms, blood coagulation disorders, and the infiltration of many organs, including the central nervous system (CNS), by polyclonally activated CD8 T lymphocytes and activated macrophages.1,8,9 Hypercytokinemia is a hallmark of the disease and is responsible for the main clinical features.10–12 The central role in the pathophysiology of the disease played by CD8 T cell activation, interferon γ overproduction. and macrophage activation was recently emphasized in perforin-deficient mice that were infected with the lymphocytic choriomeningitis virus.13
FHLH is lethal in the absence of treatment. Remission can be achieved by chemotherapy with etoposide (VP-16) combined with corticosteroids, cyclosporin A, and intrathecal methotrexate.14–17 Only allogeneic hematopoietic stem cell transplantation (HSCT), however, can cure the disease.18–22 An immunotherapeutic approach, based on the pathophysiologic features of the condition, has been developed using antithymocyte globulins (ATGs) in association with corticosteroids and cyclosporin A, combined with intrathecal methotrexate.3,13,23 A preliminary report described its efficacy in achieving remission of FHLH in 6 treated patients.24 We now report a retrospective analysis of the tolerance and efficacy of this regimen administered to 38 consecutive patients who had FHLH and were treated in 1 center between 1991 and 2005.
Between February 1991 and June 2005, 38 consecutive patients received a diagnosis of FHLH in the Pediatric Hemato-Immunology Unit of Necker-Enfants Malades Hospital (Paris, France). They received ATG therapy in association with a combination of corticosteroids and cyclosporin A, together with intrathecal methotrexate and corticosteroids, as a treatment of the acute phase. The first 6 cases have been reported.24 The criteria for FHLH diagnosis were those defined by the Histiocyte Society.23 CNS involvement was determined by clinical examination, cerebrospinal fluid analysis (pleocytosis, elevated protein level), and neuroradiologic findings.8 Patients' characteristics are described in Table 1. Genetic studies were performed in 29 (76%) patients. In 4, no mutation was found despite a comprehensive study of the currently known molecular defects that are associated with FHLH. Biallelic mutations of the perforin gene and the hMunc13.4 (UNC13D) gene were found in 14 and 11 cases, respectively, whereas no mutation in syntaxin 11 gene was detected. In the remaining 9 patients, diagnosis of FHLH was based on family history and/or parental consanguinity and/or disease severity and relapses in the absence of secondary HLH. X-linked lymphoproliferative disorder was excluded in boys by signaling lymphocyte-activation molecule–associated protein gene analysis.25
Forty-five courses of ATG (rabbit ATG; Genzyme SAS, Saint-Germain-en-Laye, France) were administered to the 38 patients. The total dosage of ATG was 50 mg/kg (n = 39) or 25 mg/kg (n = 6) according to the severity of disease, over 5 consecutive days. ATG was diluted in 5% glucose serum and infused with increasing infusion speed when well tolerated, over 5 to 8 hours. Methylprednisolone (4 mg/kg per day) was given with ATG for 5 days and then progressively tapered. Intrathecal methotrexate and corticosteroids were given at various dosages, depending on the patient's age (8 mg for those younger than 1 year; 10 mg for those between 1 and 3 years; 12 mg for those older than 3 years). Three to 5 injections were given for prophylaxis at a 1- to 2-week interval according to CNS disease severity. During the early phase of disease, the patients also received infusions of fibrinogen, irradiated packed red cells, and platelets. Fluid intake was restricted, and broad-spectrum systemic antibiotics were given. During the maintenance phase, until an HSCT was performed, cyclosporin A was given by intravenous route to reach a plasma concentration of 150 ng/mL. In the meantime, the children received intravenous immunoglobulins every 3 to 4 weeks and oral trimethoprim-sulfamethoxazole.
ATG was used as first-line therapy in 28 patients. Because of partial efficacy or relapse, 6 received a second course of ATG. Ten received ATG as a second-line treatment for the acute phase of FHLH, after other therapies. These were corticosteroids with cyclosporin A (n = 6), corticosteroids with VP-16 (n = 2), or corticosteroids with polychemotherapy (n = 2) administered elsewhere before referral to us. One of these 10 patients had 2 courses of ATG (Table 1). Seven experienced a relapse of FHLH, and 3 had refractory disease.
A complete response (CR) was defined by complete disappearance of clinical manifestations, including organomegaly, as well as of biological signs of HLH (blood cell counts, levels of triglycerides, fibrinogen, ferritin, liver enzymes, and HLA DR+ CD8+ T <5% of CD8+ T cells). A partial response (PR) was defined by a significant but incomplete improvement of clinical and/or biological manifestations. As indicated in Table 1, clinical manifestations included mainly fever, hepatosplenomegaly, neurologic symptoms, and bleeding. Biological manifestations included cytopenia, hypertriglyceridemia, hypofibrinemia, and hyperferritinemia; high blood levels of liver enzymes, pleocytosis, and hyperproteinorachy; and excess HLA DR+ /CD8+ T cells in the blood and/or cerebrospinal fluid.
A retrospective analysis of the tolerance and efficacy of ATG-based therapy was performed. The end point for analysis was November 1, 2005. The relationships between variables were tested using the χ2 test or the Fisher's exact test, where appropriate. A statistically significant difference was considered at P ≤ .05.
Tolerance of ATG
In 20 of 45 ATG courses, immediate adverse effects (days 1–5) were observed. They consisted of episodes of fever or chills during infusion (Table 2). Under the appropriate symptomatic therapy, they all were rapidly and completely reversible. No interruption of treatment was required. As shown in Table 2, transient cytopenia that consisted of isolated severe neutropenia (<300/μL) and lasted 3 to 18 days (median: 8 days) was seen in 4 cases, and transient pancytopenia was seen in 2. Neurologic signs were seen in 2 patients, 1 with seizures and the other with pyramidal irritation. They rapidly reversed without sequelae, as with the other noted adverse effects.
Infections occurred on 10 (22%) occasions after ATG with a median time of 15 days after first infusion (Table 3). Bacterial, fungal, and viral infections were encountered. In 4 cases, the infections were lethal and were caused either by disseminated fungal infection or, in 1 patient, by Epstein-Barr Virus (EBV)-induced B lymphoproliferative disorder (BLPD) associated with the disease. Infections were more frequent in patients who received ATG as a second-line treatment (6 episodes in 11 courses vs 4 episodes in 34 courses for patients who were treated with ATG as first-line therapy; P < .01). No other factor (eg, age at onset or at diagnosis, interval between first symptom of FHLH and ATG therapy, neurologic manifestations of FHLH, dosage of ATG administered) influenced tolerance of ATG therapy (data not shown). Reactivation of EBV associated with rapidly progressive multifocal lymphoproliferative disorder was observed in 3 patients who either had previously received prolonged immunosuppression by corticosteroids and cyclosporin A or had received 2 ATG courses. EBV-BLPD events occurred 2 to 35 days after the ATG course, very likely indicating the effect of T cell lymphocytopenia in this complication. In 1 case, anti-CD20 therapy (Rituximab; Roche, Nevilly-sur-Seine, France) was effective in controlling BLPD. Thus, ATG therapy after immunosuppressive therapy with corticosteroids and/or other immunosuppressive agents (n = 10), ATG (n = 6), or both (n = 1) carried a significant risk for EBV-BLPD (3 [17.5%] of 17), whereas a single ATG course had a low risk (0 of 28 EBV-BLPD).
Efficacy of ATG Therapy
Overall, ATG therapy led to a CR in 33 (73%) of 45 cases, a PR in 11 (24%) of 45 cases, and no response in 1 case (Table 4). CR was achieved in a median time of 8 days (range: 4–15 days). First-line ATG therapy induced CR in 23 (82%) of 28 cases after a first course, whereas second-line ATG therapy was effective in 5 (50%) of 10 cases after a first course. The difference does not reach significance. Of note, a second course of ATG seemed to be as efficient as a first course by inducing CR in 5 (71%) of 7 cases vs 28 (74%) of 38 cases. Molecular diagnosis of FHLH, age at diagnosis, ATG dosage, and the interval between the first symptom of HLH and ATG therapy did not influence ATG efficacy (data not shown). In contrast, in patients with signs of neurologic disease (n = 19), the probability of achieving CR was 58% (11 of 19) compared with 89% (17 of 19) in those without overt neurologic disease (P = .05). In patients who did not receive a transplant shortly after ATG therapy, median duration of CR was 1.3 months with considerable variability (range: 0.5–18 months).
Once a response was achieved, the practice from 1991 to 1995 was to perform HSCT only for patients with a healthy HLA genoidentical donor and give the others maintenance therapy (cyclosporin A and reduced dosage of corticosteroids). It was considered that the risk/benefit rate was too low for patients for whom no evidence of genetic HLH was provided and who lacked an HLA-identical donor. In this instance, a doubt about the indication of HSCT combined with its known toxicity outweighed the potential chance of cure.
Under this policy, relapse occurred in 10 patients, including CNS disorder in 5. and eventually led to death of 9 (Fig 1). After 1996, an HSCT from a genoidentical, unrelated, or haploidentical donor was performed on patients with CR or PR. Overall, median time between onset of therapy and HSCT was 6 weeks (range: 4–32 weeks). It is remarkable that 16 of 19 patients who were treated by ATG as first-line therapy and underwent HSCT are alive and well, whereas only 4 of 8 patients who received HSCT after second-line ATG therapy have been cured (Fig 1). As shown in Fig 1, the option to undergo an HSCT for patients who relapsed after first-line ATG was only 3 in 10. This included 2 patients in whom there was insufficient control of FHLH by ATG, rather than excessive delay in undergoing HSCT.
This report describes the outcome for 38 patients who had FHLH and received combined immunotherapy based on ATG, steroids, and maintenance cyclosporin A associated with intrathecal methotrexate injection to achieve remission. It extends the preliminary report of Stephan et al24 showing its efficacy in a first series of 6 patients. Overall tolerance of ATG therapy can be regarded as acceptable. All patients had received the full 5-day course, and 7 (18%) received 2 courses. Immediate toxicity was limited to fever and discomfort. Transient neutropenia was infrequent and possibly related to other factors, such as FHLH disease and/or the effect of intrathecal methotrexate injections, which, in patients with HLH, frequently result in detectable methotrexate concentrations in blood (data not shown). This contrasts with the consequences of the VP-16–based therapy of FHLH reported often to be associated with neutropenia in the first 2 months of therapy.15 ATG nevertheless induces a rapid and profound T cell lymphopenia,24 which may be a concern in favoring opportunistic infections. The occurrence of infectious complications was indeed significant in this cohort, including 3 cases of EBV-BLPD. Preemptive therapy with anti-CD20 antibody, as reported for cases of X-linked lymphoproliferative syndrome, could be considered for patients who had FHLH and showed evidence of EBV replication to reduce the risk for EBV-BLPD.26 Infectious complications led to or were associated with the death of 4 patients. Of note, however, most of the severe infections occurred in patients who received other immunosuppressive or cytotoxic therapies before ATG (corticosteroids, cyclosporin A, and VP-16) or were treated with 2 courses of ATG. It thus seems that ATG associated with steroids, cyclosporin A, and intrathecal methotrexate for initial therapy carries an acceptable toxicity risk. Duration of neutropenia when observed was much shorter and thus induced a lower risk to predispose to bacterial and fungal infections as compared with chemotherapy of FHLH. Moreover, there are no detectable long-term adverse effects, a possible difference from VP-16 use, which can induce secondary acute myeloblastic leukemia, as reported in 3 patients who were treated according to the HLH 94 protocol.15,27,28 Overall, the efficacy of ATG in inducing remission of FHLH was confirmed in this series of patients who were treated during a 14-year period, because there was a response in all but 1 patient and a CR in 73%. Risk factor analysis shows that first-line ATG therapy had a higher chance of inducing CR (82%), whereas second-line ATG therapy was fully efficient in only half of the cases. The only other risk factor was the presence of FHLH neurologic signs that indicated a poorer response. This may be related to the fact that ATG does not cross the blood-brain barrier efficiently and that presence of a neurologic disease possibly reflects a more severe FHLH. No other factors (age at onset and interval between onset and therapy) could be found to influence efficacy. What remains undetermined is the optimal ATG regimen. Because of a slightly less severe disease presentation, ATG dosage was reduced to half (5 mg/kg per day for 5 days) in 6 patients. Efficacy seemed similar, but this was possibly because disease was less severe. In any case, efficacy can be regarded as satisfactory compared with the HLH 94 protocol because in the latter group, 50% of patients still had active disease after 2 months of therapy.15 This compares with 26% (10 of 38) in our series. Obviously, in the absence of a direct comparative study, no firm conclusion can be drawn.
It is worth noting that ATG used in combination with corticosteroids as a rescue treatment, either because a first ATG course was partially or only transiently efficient or because another therapy was not efficient, may still lead to an FHLH remission. However, the overall efficacy in this context was reduced and the toxicity was higher than after first-line ATG. In total, 30 patients who received ATG therapy underwent HSCT. When performed without delay, as performed since 1996, it led to a high cure rate (76%). Striking is the observation that patients who were treated by HSCT shortly after ATG was administered as first-line therapy had a very good outcome: 16 of 19 are alive and well. These results obviously need to be confirmed in larger series. It is also possible that use of ATG in the CR preceding HSCT contributes to overall improved survival by reducing also graft-versus-host disease occurrence.18 Additional refractory relapses prevented HSCT in patients who did not immediately undergo the treatment once a remission was achieved. It is likely that more patients would have been treated by HSCT if the guideline adopted in 1996 had been used since 1991.
Altogether, this study shows that an immunotherapy regimen based on ATG combined with corticosteroids and intrathecal methotrexate injection is an alternative to the HLH 2004 protocol based on VP-16 and intrathecal methotrexate injection for the treatment of FHLH. Choice of VP-16 was based initially at a time when cause was uncertain, because of its ability to be cytotoxic for phagocytic cells that were thought to be primarily involved in the pathogenesis of FHLH.14 Another rationale for VP-16 use was then provided because VP-16 induces cell apoptosis, a mean to counterbalance the defect in apoptosis that is observed in FHLH.29 As stated, in the absence of direct comparison within the same study, conclusions can be only tentative. Nevertheless, the rationale for an immunotherapy for FHLH is strongly based on the primary role of defective cytolytic T cells in the pathogenesis of the disease.3,13 This is particularly emphasized by the observation that anti–T-cell antibody (and even anti-CD8 T-cell antibody) therapy can control experimental FHLH in lymphochoriomeningitis virus–infected, perforin-deficient mice.13 This crucial experimental result, together with the clinical evidence for the role of T cells in the disease pathogenesis, including increased plasma levels of γ interferon10 as well as detection of intrahepatic γ interferon and interleukin-2 receptor,11 constitute a rationale for additional improvement of immunologic FHLH therapy. A more focused anti–T-cell effector treatment might improve efficacy and reduce toxicity. Indeed, therapeutic strategies of FHLH aimed at reducing toxicity of the primary treatment could ultimately improve the outcome of allogeneic HSCT, the necessary curative step.
This study shows that immunotherapy of FHLH, based on ATG, is an efficient treatment that is associated with significant but acceptable toxicity and provides a means to prepare patients for curative HSCT. Additional improvements are required in ATG dosage determination, and progress toward a more focused immunotherapy can be envisaged to reduce further the toxicity of FHLH therapy.
This work was supported by INSERM.
We thank the medical and nursery staff of the Department of Pediatric Immuno-Hematology of Necker Enfants Malades Hospital for the excellent care of patients. We thank Ms Sifouane for skillful secretarial assistance.
- Accepted January 29, 2007.
- Address correspondence to Alain Fischer, MD, PhD, Unité d'Immunologie et Hématologie Pédiatrique, Hôpital Necker-Enfants Malades, 149 rue de Sèvres, 75015 Paris, France. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- Schneider EM, Lorenz I, Muller-Rosenberger M, Steinbach G, Kron M, Janka-Schaub GE. Hemophagocytic lymphohistiocytosis is associated with deficiencies of cellular cytolysis but normal expression of transcripts relevant to killer-cell-induced apoptosis. Blood.2002;100 :2891– 2898
- ↵Stepp SE, Dufourcq-Lagelouse R, Le Deist F, et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science.1999;286 :1957– 1959
- ↵zur Stadt U, Schmidt S, Kasper B, et al. Linkage of familial hemophagocytic lymphohistiocytosis (FHL) type-4 to chromosome 6q24 and identification of mutations in syntaxin 11. Hum Mol Genet.2005;14 :827– 834
- ↵Haddad E, Sulis ML, Jabado N, Blanche S, Fischer A, Tardieu M. Frequency and severity of central nervous system lesions in hemophagocytic lymphohistiocytosis. Blood.1997;89 :794– 800
- ↵Henter JI, Elinder G, Soder O, Hansson M, Andersson B, Andersson U. Hypercytokinemia in familial hemophagocytic lymphohistiocytosis. Blood.1991;78 :2918– 2922
- ↵Billiau AD, Roskams T, Van Damme-Lombaerts R, Matthys P, Wouters C. Macrophage activation syndrome: characteristic findings on liver biopsy illustrating the key role of activated, IFN-gamma-producing lymphocytes and IL-6- and TNF-alpha-producing macrophages. Blood.2005;105 :1648– 1651
- ↵Jordan MB, Hildeman D, Kappler J, Marrack P. An animal model of hemophagocytic lymphohistiocytosis (HLH): CD8+ T cells and interferon gamma are essential for the disorder. Blood.2004;104 :735– 743
- ↵Henter JI, Samuelsson-Horne A, Arico M, et al. Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunochemotherapy and bone marrow transplantation. Blood.2002;100 :2367– 2373
- Fischer A, Virelizier JL, Arenzana-Seisdedos F, Perez N, Nezelof C, Griscelli C. Treatment of four patients with erythrophagocytic lymphohistiocytosis by a combination of epipodophyllotoxin, steroids, intrathecal methotrexate, and cranial irradiation. Pediatrics.1985;76 :263– 268
- ↵Ouachee-Chardin M, Elie C, de Saint Basile G, et al. Hematopoietic stem cell transplantation in hemophagocytic lymphohistiocytosis (HLH): a single report of 48 patients. Pediatrics.2006;117(4) . Available at: www.pediatrics.org/cgi/content/full/117/4/e743
- Blanche S, Caniglia M, Girault D, Landman J, Griscelli C, Fischer A. Treatment of hemophagocytic lymphohistiocytosis with chemotherapy and bone marrow transplantation: a single-center study of 22 cases. Blood.1991;78 :51– 54
- ↵Jabado N, de Graeff-Meeder ER, Cavazzana-Calvo M, et al. Treatment of familial hemophagocytic lymphohistiocytosis with bone marrow transplantation from HLA genetically nonidentical donors. Blood.1997;90 :4743– 4748
- ↵Stephan JL, Donadieu J, Ledeist F, Blanche S, Griscelli C, Fischer A. Treatment of familial hemophagocytic lymphohistiocytosis with antithymocyte globulins, steroids, and cyclosporin A. Blood.1993;82 :2319– 2323
- ↵Arico M, Imashuku S, Clementi R, et al. Hemophagocytic lymphohistiocytosis due to germline mutations in SH2D1A, the X-linked lymphoproliferative disease gene. Blood.2001;97 :1131– 1133
- ↵Milone MC, Tsai DE, Hodinka RL, et al. Treatment of primary Epstein-Barr virus infection in patients with X-linked lymphoproliferative disease using B-cell-directed therapy. Blood.2005;105 :994– 996
- Copyright © 2007 by the American Academy of Pediatrics