- Email: [email protected]
Cytotoxic therapy for severe avian influenza A (H5N1) infection
Hypothesis
Cytotoxic therapy for severe avian influenza A (H5N1) infection Jan-Inge Henter, Chun-Bong Chow, Chi-Wai Leung, Yu-Lung Lau Lancet 2006; 367: 870–73 Published Online March 2, 2006 DOI:10.1016/S0140-6736(06) 68232-9 Childhood Cancer Research Unit, Department of Woman and Child Health, Karolinska Institutet, Karolinska University Hospital Q6:05, SE-171 76 Stockholm, Sweden (Prof J-I Henter MD); Department of Paediatrics and Adolescent Medicine, Princess Margaret Hospital, Hong Kong (C-B Chow FRCPCH, C-W Leung FRCPCH); and Department of Paediatrics and Adolescent Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong (Prof Y-L Lau MD) Correspondence to: Jan-Inge Henter [email protected]
For WHO information on mortality rates see http://www.who.int/csr/disease/ avian_influenza/en/
870
The mortality rate in documented avian influenza A virus subtype H5N1 infection is still high, which is currently reported by WHO at about 50%. Post-mortem analyses in affected patients have revealed haemophagocytosis similar to that found in patients with haemophagocytic lymphohistiocytosis (HLH); such haemophagocytosis could be a very prominent post-mortem feature in H5N1 infection. There are also clinical similarities between H5N1 infection and HLH, such as massive hypercytokinaemia, cytopenia, and acute encephalitis. Importantly, patients with another severe viral infection that may be complicated by secondary HLH, severe Epstein-Barr-virus-associated HLH, have significantly better survival if specific HLH therapy (including the cytotoxic and pro-apoptotic drug etoposide) is initiated early, with survival reported to rise from about 50% to 90%. With this notable improvement in survival, specific HLH treatment, including cytotoxic therapy, could be considered in patients with severe avian influenza A infection complicated by secondary HLH. The first outbreak of documented infection of avian influenza A virus subtype H5N1 in human beings caused disease in 18 patients and resulted in six deaths (33%) in Hong Kong in 1997.1,2 The mortality rate remains high, and as of Feb 9, 2006, 166 confirmed human cases of H5N1 influenza worldwide have been reported by WHO, of whom 88 (53%) have died. Treatment of avian influenza A H5N1 infection includes antiviral therapy, such as ribavirin and inhibitors of influenza neuraminidase (oseltamivir and zanamivir), which are sometimes used in combination with corticosteroids.3 The presence of residue Asp31 in the M2 protein of the H5N1 virus invariably confers resistance to amantadine treatment, and sequence analyses have revealed that Asp31 is present in avian viruses (isolated from both birds and human beings) isolated in Thailand and Vietnam, whereas the viruses isolated in mainland China and Indonesia generally still respond to amantadine.4,5 Early use of the neuraminidase inhibitors could reduce the disease duration as well as the risk for complications.6 However, recently, H5N1 viruses with an aminoacid substitution in neuraminidase (thereby conferring high-level resistance to oseltamivir treatment) were isolated from two of eight Vietnamese patients during oseltamivir treatment and both patients died despite early initiation of treatment in one patient.7 Other options to reduce human mortality include precautions such as elimination of poultry in affected areas and poultry vaccination. Massive vaccination campaigns have been initiated in China, and ever since Hong Kong began the screening of all imported poultry from China for H5N1 vaccination and laboratory evidence of H5N1 infection, Hong Kong has remained free of H5N1 transmission to human beings.8 Presently, there are no vaccinations for human beings. Despite precautions, this lethal pathogen might cause a severe pandemic. Poultry vaccination, a core part of an H5N1 control strategy, is difficult to implement well. Moreover, human-to-human transmission of H5N1 has
been suggested, although the transmission rate presently is low, if any.9 Recent data of the complete genome of the 1918 influenza virus confirm phylogenetic studies suggesting that the 1918 virus was an entirely avian-like virus that adapted to human beings.10 Furthermore, current influenza A H5N1 viruses have been suggested to have evolved into more virulent forms since 1997.5,9 Therefore, since the mortality from H5N1 infection is high, and since there is a concern that the virus could cause a pandemic, novel treatments for human beings are warranted.
H5N1 infection and haemophagocytic lymphohistiocytosis (HLH) Patients with H5N1 infection have symptoms similar to those found in HLH, an often rapidly fatal disease for which survival rates recently have improved greatly.11 Investigators from Hong Kong have reported that two patients with fatal H5N1 disease in 1997 had a reactive haemophagocytic syndrome as the most prominent feature.12 Moreover, these patients had high amounts of soluble interleukin-2 receptor, interleukin 6, and interferon . Similarly, HLH is associated with a massive hypercytokinaemia, and the earliest reference to this term in PubMed is actually to HLH.13 Reactive haemophagocytosis has also subsequently been reported in H5N1-infected patients from Hong Kong in 2003 and from Thailand in 2004.4,14 The cause of death in patients with HLH is often a sepsis-related condition with multiorgan failure, which is also seen in patients with H5N1 infection. Bicytopenia is another common complication.3,4,14,15 Many individuals with HLH develop encephalitis, which is not rare as a cause of death;11,16,17 acute encephalitis was reported as the cause of death in two Vietnamese siblings with fatal H5N1 infection.18 Increased protein content in the cerebrospinal fluid is another common finding in patients with HLH and patients with H5N1 infection, although the mechanisms causing the encephalopathy in HLH could be different to those in influenza.15,18 In brief, there are many clinical www.thelancet.com Vol 367 March 11, 2006
Hypothesis
similarities between H5N1 infection and HLH; both disorders are also characterised by a malignant—ie, often fatal—course of a prominent non-malignant inflammatory reaction, and patients with H5N1 could develop secondary HLH.
Can HLH chemotherapy be used to treat H5N1 infection? HLH is a general term for the disorder’s primary forms (genetically inherited; such as familial HLH) and secondary forms (mostly infection-associated). In 1994, a treatment protocol was developed for HLH (HLH-94), which was based on corticosteroids, etoposide, cyclosporine A, and intrathecal methotrexate, followed by stem cell transplantation for patients with the inherited form of the disease.11 The protocol has been successful both for familial, autosomal recessive HLH and secondary HLH.11 Importantly, Epstein-Barr-virusassociated HLH (EBV-HLH) is linked with a high mortality, especially in east Asia, but the mortality is reduced significantly if HLH treatment is initiated early, in children as well as young adults.11,19–21 Moreover, it has been reported that a patient with a haemophagocytic syndrome associated with hepatitis-B-virus infection responded to etoposide, became apyrexic, and also later became negative for the virus on lamivudine treatment.22 Thus, although cytotoxic treatment (etoposide) for influenza might seem to be a substantial jump in therapeutic thinking, such therapy could still be reasonable. HLH treatment has shown to be very efficient against EBV-HLH. Obviously, the treatment should be combined with antiviral therapy and relevant supportive therapy. We think that such a combination is worth considering, especially since the mortality recorded in H5N1 infections remains high.
The therapeutic mechanism? Patients with the familial form of HLH have shown to have a lack of apoptosis-triggering within the immune system.23 Subsequent molecular studies first revealed mutations in the gene encoding perforin (PRF),24 which is important for the initiation of apoptosis in target cells, and later in two genes regulating intracellular transport and exocytosis of perforin-containing granule (UNC13D, encoding the protein MUNC13-4; and STX11, encoding syntaxin 11).25,26 These mutations result in low or absent activity of cytotoxic T cells and natural killer cells. Also, patients with EBV-HLH often have impaired cytotoxic capacity during active disease, but the underlying mechanism is not known in detail. Importantly, the cytotoxic drug etoposide is well known as an excellent inducer of apoptosis, and dexamethasone has pro-apoptotic effects on mononuclear cells.27,28 Therefore, a therapeutic mechanism for these drugs could be to downregulate the inflammatory response by apoptosis-triggering of cells within the immune system. www.thelancet.com Vol 367 March 11, 2006
Although corticosteroids and etoposide could hamper important functions in viral defence, HLH-94 treatment with these drugs have been shown to greatly increase survival in severely sick patients with Epstein-Barr virus infection,19–21 and could be effective in severely sick patients with H5N1 infection also.
Testing the hypothesis Results of the HLH-94 treatment have been published and have shown the therapy to be effective in secondary HLH;11 therefore, it could be ethically justified to initially use 8 weeks of this treatment for patients with H5N1 infection and secondary HLH. Antiviral therapy should also be used and the viral load monitored. Although the HLH therapy could be tested in animal models relevant for avian influenza, such as primates or mice,29,30 HLH-94 has been used successfully in human beings affected by severe virus infections for more than 10 years;11,19–21 thus, we thought well-controlled use of the treatment in people with severe H5N1 infection would be reasonable and relevant in the current context.10 The first 8 weeks of HLH-94 treatment include etoposide and dexamethasone, and intrathecal methotrexate for selected patients with progressive neurological symptoms after 2 weeks of therapy or with abnormal cerebrospinal fluid that has not improved at that time.11 The Histiocyte Society has launched a revised protocol, HLH-2004, with use of cyclosporine A upfront in addition to etoposide and dexamethasone. However, we would not presently recommend cyclosporine A upfront in patients with avian influenza, since renal problems are frequent during H5N1 infection, and we therefore suggest investigating the published HLH-94 protocol as a first step.11 Etoposide could be considered to be initially used once (instead of twice) weekly for the first 2 weeks, because of the risk of myelotoxic effects. Additionally, we would suggest a reduction of every etoposide dose from 150 mg/m2 to 100 mg/m2 in patients aged 15 years or more (and further reduce the dose to 50 mg/m2 in middleaged and elderly patients), since children seem to tolerate the drug better than adults. The figure shows the 10 mg/m2 5 mg/m2
Daily dexamethasone dose (mg/m2)
2·5 mg/m2 1·25 mg/m2
Intravenous etoposide dose (150 mg/m2 per dose)
Week
1
2
3
4
5
6
7
8
9
Figure: Suggested therapy for patients with avian influenza A (H5N1) infection and secondary HLH This protocol is a modification of the HLH-94 protocol.11 Suggested doses are for children only.
871
Hypothesis
suggested treatment protocol for children with H5N1 and HLH. Similarly, the dose of dexamethasone can be reduced in adults. It may be too early to suggest intrathecal therapy for patients with H5N1 infection, since most patients with HLH respond favourably (with respect to neurological findings) to the combination of systemic dexamethasone and etoposide and do not receive intrathecal therapy,11 and dexamethasone is known to pass the blood-brain-barrier well. Whether treatment needs to be exactly 8 weeks as suggested above remains to be elucidated. Since EpsteinBarr virus is a DNA virus that is incorporated into the host (giving rise to extended illness), and H5N1 is an RNA virus (resulting in a short sharp illness), a treatment duration shorter than 8 weeks could be sufficient. Some patients with mild, early H5N1 disease could respond to mild therapy (eg, corticosteroids, intravenous immunoglobulins, or both), but results from EBV-HLH studies clearly indicate that early introduction of etoposide benefits patients with EBV-HLH.11,19–21 The HLH-2004 protocol presently states that the treatment is suggested for patients with severe, persistent, or recurrent disease (or with familial HLH) and accordingly, we suggest HLH therapy for patients with H5N1 and severe or persistent secondary HLH. The diagnostic criteria for HLH in HLH94 included fever, splenomegaly, bicytopenia, hypertriglyceridaemia or hypofibrinogenaemia (or both), and haemophagocytosis.15 For the HLH-2004 protocol, the criteria were revised, with three new criteria (ferritin 500 g/L, low natural killer cell activity, and soluble interleukin-2 receptor 2400 U/mL). Five of the eight criteria in total are now needed to make a diagnosis of HLH.31 Since the regions of the authors (Sweden and Hong Kong) are both free from human H5N1 infection, we cannot test this therapeutic hypothesis ourselves but welcome others to do so. We urge that each case in which HLH therapy is used is reported to the HLH Study Centre at the Karolinska Institute in Stockholm, Sweden. Obviously, we cannot guarantee any therapeutic efficacy, but it would be unfortunate if this therapy was effective but never attempted. Importantly, we want to emphasise that this hypothesis must not be misinterpreted as an instruction to treat influenza-like symptoms with etoposide. In addition to the reporting of all treated patients to the HLH Study Centre, we would welcome WHO to consider a platform for the undertaking of clinical trials based on a modified HLH protocol (including corticosteroids and etoposide) in addition to supportive and antiviral therapy. Contributors C-B Chow, supported by C-W Leung and Y-L Lau, prepared a management protocol for H5N1 and considered HLH treatment. J-I Henter supported this idea and developed it further. J-I Henter wrote the manuscript and the final version was approved by all the authors. Conflict of interest statement We declare that we have no conflict of interest.
872
References 1 Yuen KY, Chan PKS, Peiris M, et al. Clinical features and rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus. Lancet 1998; 351: 467–71. 2 Claas EC, Osterhaus AD, van Beek R, et al. Human influenza A H5N1 virus related to a highly pathogenic avian influenza virus. Lancet 1998; 351: 472–77. 3 Hien TT, Liem NT, Dung NT, et al. Avian influenza A (H5N1) in 10 patients in Vietnam. N Engl J Med 2004; 350: 1179–88. 4 Peiris JSM, Yu WC, Leung CW, et al. Re-emergence of fatal human influenza A subtype H5N1 disease. Lancet 2004; 363: 617–19. 5 Li KS, Guan Y, Wang J, et al. Genesis of a highly pathogenic and potentially pandemic H5N1 influenza virus in eastern Asia. Nature 2004; 430: 209–13. 6 Moscona A. Neuraminidase inhibitors for influenza. N Engl J Med 2005; 353: 1363–73. 7 de Jong MD, Tran TT, Truong HK, et al. Oseltamivir resistance during treatment of influenza A (H5N1) infection. N Engl J Med 2005; 353: 2667–72. 8 Normile D. Avian influenza. China will attempt largest-ever animal vaccination campaign. Science 2005; 310: 1256–57. 9 Ungchusak K, Auewarakul P, Dowell SF, et al. Probable person-toperson transmission of avian influenza A (H5N1). N Engl J Med 2005; 352: 333–40. 10 Taubenberger JK, Reid AH, Lourens RM, Wang R, Jin G, Fanning TG. Characterization of the 1918 influenza polymerase genes. Nature 2005; 437: 889–93. 11 Henter J-I, Samuelsson-Horne AC, Arico M, et al. Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunochemotherapy and bone marrow transplantation. Blood 2002; 100: 2367–73. 12 To KF, Chan PK, Chan KF, et al. Pathology of fatal human infection associated with avian influenza A H5N1 virus. J Med Virol 2001; 63: 242–46. 13 Henter J-I, Söder O, Hansson M, Elinder G, Andersson B, Andersson U. Hypercytokinemia in familial hemophagocytic lymphohistiocytosis. Blood 1991; 78: 2918–22. 14 Chokephaibulkit K, Uiprasertkul M, Puthavathana P, et al. A child with avian influenza A (H5N1) infection. Pediatr Infect Dis J 2005; 24: 162–66. 15 Henter J-I, Elinder G, Öst Å. Diagnostic guidelines for hemophagocytic lymphohistiocytosis. Semin Oncol 1991; 18: 29–33. 16 Henter J-I, Elinder G. Cerebromeningeal hemophagocytic lymphohistiocytosis. Lancet 1992; 339: 104–07. 17 Henter J-I, Nennesmo I. Neuropathological findings and neurological symptoms in 23 children with hemophagocytic lymphohistiocytosis. J Pediatr 1997; 130: 358–65. 18 de Jong MD, Cam BV, Qui PT, et al. Fatal avian influenza A (H5N1) in a child presenting with diarrhea followed by coma. N Engl J Med 2005; 352: 686–91. 19 Imashuku S, Hibi S, Ohara T, et al. Effective control of EpsteinBarr virus-related hemophagocytic lymphohistiocytosis with immunochemotherapy. Blood 1999; 93: 1869–74. 20 Imashuku S, Kuriyama K, Teramura T, et al. Requirement for etoposide in the treatment of Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. J Clin Oncol 2001; 19: 2665–73. 21 Imashuku S, Kuriyama K, Sakai R, et al. Treatment of EpsteinBarr virus-associated hemophagocytic lymphohistiocytosis (EBV-HLH) in young adults. Med Pediatr Oncol 2003; 41: 103–09. 22 Aleem A, Al Amoudi S, Al-Mashhadani S, Siddiqi N. Haemophagocytic syndrome associated with hepatitis-B virus infection responding to etoposide. Clin Lab Haematol 2005; 27: 395–98. 23 Fadeel B, Orrenius S, Henter J-I. Induction of apoptosis and caspase activation in cells obtained from familial haemophagocytic lymphohistiocytosis patients. Br J Haematol 1999; 106: 406–15. 24 Stepp SE, Dufourcq-Lagelouse R, Le Deist F, et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science 1999; 286: 1957–59.
www.thelancet.com Vol 367 March 11, 2006
Hypothesis
25
26
27
28
Feldmann J, Callebaut I, Raposo G, et al. Munc13-4 is essential for cytolytic granules fusion and is mutated in a form of familial hemophagocytic lymphohistiocytosis (FHL3). Cell 2003; 115: 461–73. 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: 827–34. Fadeel B, Orrenius S, Henter J-I. Familial hemophagocytic lymphohistiocytosis: too little cell death may seriously damage your health. Leuk Lymphoma 2001; 42: 13–20. Rimmelzwann GF, Kuiken T, van Amerongen G, Bestebroer TM, Fouchier RA, Osterhaus AD. Pathogenesis of influenza A (H5N1) virus infection in a primate model. J Virol 2001; 75: 6687–91.
www.thelancet.com Vol 367 March 11, 2006
29
30
31
Lipatov AS, Webby RJ, Goverkova EA, Krauss S, Webster RG. Efficacy of H5 influenza vaccines produced by reverse genetics in a lethal mouse model. J Infect Dis 2005; 191: 1216–20. Rivas Totino PR, Pratt Riccio EK, Corte-Real S, Daniel-Ribeiro CT, Ferreira-da-Cruz MD. Dexamethasone has pro-apoptotic effects on non-activated fresh peripheral blood mononuclear cells. Cell Biol Int 2005; published online Nov 2. DOI:10.1016/j.cellbi. 2005.09.002. Henter J-I, Tondini C, Pritchard J. Histiocytic syndromes. Crit Rev Oncol Hematol 2004; 50: 157–74.
873
Cytotoxic therapy for severe avian influenza A (H5N1) infection Jan-Inge Henter, Chun-Bong Chow, Chi-Wai Leung, Yu-Lung Lau Lancet 2006; 367: 870–73 Published Online March 2, 2006 DOI:10.1016/S0140-6736(06) 68232-9 Childhood Cancer Research Unit, Department of Woman and Child Health, Karolinska Institutet, Karolinska University Hospital Q6:05, SE-171 76 Stockholm, Sweden (Prof J-I Henter MD); Department of Paediatrics and Adolescent Medicine, Princess Margaret Hospital, Hong Kong (C-B Chow FRCPCH, C-W Leung FRCPCH); and Department of Paediatrics and Adolescent Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong (Prof Y-L Lau MD) Correspondence to: Jan-Inge Henter [email protected]
For WHO information on mortality rates see http://www.who.int/csr/disease/ avian_influenza/en/
870
The mortality rate in documented avian influenza A virus subtype H5N1 infection is still high, which is currently reported by WHO at about 50%. Post-mortem analyses in affected patients have revealed haemophagocytosis similar to that found in patients with haemophagocytic lymphohistiocytosis (HLH); such haemophagocytosis could be a very prominent post-mortem feature in H5N1 infection. There are also clinical similarities between H5N1 infection and HLH, such as massive hypercytokinaemia, cytopenia, and acute encephalitis. Importantly, patients with another severe viral infection that may be complicated by secondary HLH, severe Epstein-Barr-virus-associated HLH, have significantly better survival if specific HLH therapy (including the cytotoxic and pro-apoptotic drug etoposide) is initiated early, with survival reported to rise from about 50% to 90%. With this notable improvement in survival, specific HLH treatment, including cytotoxic therapy, could be considered in patients with severe avian influenza A infection complicated by secondary HLH. The first outbreak of documented infection of avian influenza A virus subtype H5N1 in human beings caused disease in 18 patients and resulted in six deaths (33%) in Hong Kong in 1997.1,2 The mortality rate remains high, and as of Feb 9, 2006, 166 confirmed human cases of H5N1 influenza worldwide have been reported by WHO, of whom 88 (53%) have died. Treatment of avian influenza A H5N1 infection includes antiviral therapy, such as ribavirin and inhibitors of influenza neuraminidase (oseltamivir and zanamivir), which are sometimes used in combination with corticosteroids.3 The presence of residue Asp31 in the M2 protein of the H5N1 virus invariably confers resistance to amantadine treatment, and sequence analyses have revealed that Asp31 is present in avian viruses (isolated from both birds and human beings) isolated in Thailand and Vietnam, whereas the viruses isolated in mainland China and Indonesia generally still respond to amantadine.4,5 Early use of the neuraminidase inhibitors could reduce the disease duration as well as the risk for complications.6 However, recently, H5N1 viruses with an aminoacid substitution in neuraminidase (thereby conferring high-level resistance to oseltamivir treatment) were isolated from two of eight Vietnamese patients during oseltamivir treatment and both patients died despite early initiation of treatment in one patient.7 Other options to reduce human mortality include precautions such as elimination of poultry in affected areas and poultry vaccination. Massive vaccination campaigns have been initiated in China, and ever since Hong Kong began the screening of all imported poultry from China for H5N1 vaccination and laboratory evidence of H5N1 infection, Hong Kong has remained free of H5N1 transmission to human beings.8 Presently, there are no vaccinations for human beings. Despite precautions, this lethal pathogen might cause a severe pandemic. Poultry vaccination, a core part of an H5N1 control strategy, is difficult to implement well. Moreover, human-to-human transmission of H5N1 has
been suggested, although the transmission rate presently is low, if any.9 Recent data of the complete genome of the 1918 influenza virus confirm phylogenetic studies suggesting that the 1918 virus was an entirely avian-like virus that adapted to human beings.10 Furthermore, current influenza A H5N1 viruses have been suggested to have evolved into more virulent forms since 1997.5,9 Therefore, since the mortality from H5N1 infection is high, and since there is a concern that the virus could cause a pandemic, novel treatments for human beings are warranted.
H5N1 infection and haemophagocytic lymphohistiocytosis (HLH) Patients with H5N1 infection have symptoms similar to those found in HLH, an often rapidly fatal disease for which survival rates recently have improved greatly.11 Investigators from Hong Kong have reported that two patients with fatal H5N1 disease in 1997 had a reactive haemophagocytic syndrome as the most prominent feature.12 Moreover, these patients had high amounts of soluble interleukin-2 receptor, interleukin 6, and interferon . Similarly, HLH is associated with a massive hypercytokinaemia, and the earliest reference to this term in PubMed is actually to HLH.13 Reactive haemophagocytosis has also subsequently been reported in H5N1-infected patients from Hong Kong in 2003 and from Thailand in 2004.4,14 The cause of death in patients with HLH is often a sepsis-related condition with multiorgan failure, which is also seen in patients with H5N1 infection. Bicytopenia is another common complication.3,4,14,15 Many individuals with HLH develop encephalitis, which is not rare as a cause of death;11,16,17 acute encephalitis was reported as the cause of death in two Vietnamese siblings with fatal H5N1 infection.18 Increased protein content in the cerebrospinal fluid is another common finding in patients with HLH and patients with H5N1 infection, although the mechanisms causing the encephalopathy in HLH could be different to those in influenza.15,18 In brief, there are many clinical www.thelancet.com Vol 367 March 11, 2006
Hypothesis
similarities between H5N1 infection and HLH; both disorders are also characterised by a malignant—ie, often fatal—course of a prominent non-malignant inflammatory reaction, and patients with H5N1 could develop secondary HLH.
Can HLH chemotherapy be used to treat H5N1 infection? HLH is a general term for the disorder’s primary forms (genetically inherited; such as familial HLH) and secondary forms (mostly infection-associated). In 1994, a treatment protocol was developed for HLH (HLH-94), which was based on corticosteroids, etoposide, cyclosporine A, and intrathecal methotrexate, followed by stem cell transplantation for patients with the inherited form of the disease.11 The protocol has been successful both for familial, autosomal recessive HLH and secondary HLH.11 Importantly, Epstein-Barr-virusassociated HLH (EBV-HLH) is linked with a high mortality, especially in east Asia, but the mortality is reduced significantly if HLH treatment is initiated early, in children as well as young adults.11,19–21 Moreover, it has been reported that a patient with a haemophagocytic syndrome associated with hepatitis-B-virus infection responded to etoposide, became apyrexic, and also later became negative for the virus on lamivudine treatment.22 Thus, although cytotoxic treatment (etoposide) for influenza might seem to be a substantial jump in therapeutic thinking, such therapy could still be reasonable. HLH treatment has shown to be very efficient against EBV-HLH. Obviously, the treatment should be combined with antiviral therapy and relevant supportive therapy. We think that such a combination is worth considering, especially since the mortality recorded in H5N1 infections remains high.
The therapeutic mechanism? Patients with the familial form of HLH have shown to have a lack of apoptosis-triggering within the immune system.23 Subsequent molecular studies first revealed mutations in the gene encoding perforin (PRF),24 which is important for the initiation of apoptosis in target cells, and later in two genes regulating intracellular transport and exocytosis of perforin-containing granule (UNC13D, encoding the protein MUNC13-4; and STX11, encoding syntaxin 11).25,26 These mutations result in low or absent activity of cytotoxic T cells and natural killer cells. Also, patients with EBV-HLH often have impaired cytotoxic capacity during active disease, but the underlying mechanism is not known in detail. Importantly, the cytotoxic drug etoposide is well known as an excellent inducer of apoptosis, and dexamethasone has pro-apoptotic effects on mononuclear cells.27,28 Therefore, a therapeutic mechanism for these drugs could be to downregulate the inflammatory response by apoptosis-triggering of cells within the immune system. www.thelancet.com Vol 367 March 11, 2006
Although corticosteroids and etoposide could hamper important functions in viral defence, HLH-94 treatment with these drugs have been shown to greatly increase survival in severely sick patients with Epstein-Barr virus infection,19–21 and could be effective in severely sick patients with H5N1 infection also.
Testing the hypothesis Results of the HLH-94 treatment have been published and have shown the therapy to be effective in secondary HLH;11 therefore, it could be ethically justified to initially use 8 weeks of this treatment for patients with H5N1 infection and secondary HLH. Antiviral therapy should also be used and the viral load monitored. Although the HLH therapy could be tested in animal models relevant for avian influenza, such as primates or mice,29,30 HLH-94 has been used successfully in human beings affected by severe virus infections for more than 10 years;11,19–21 thus, we thought well-controlled use of the treatment in people with severe H5N1 infection would be reasonable and relevant in the current context.10 The first 8 weeks of HLH-94 treatment include etoposide and dexamethasone, and intrathecal methotrexate for selected patients with progressive neurological symptoms after 2 weeks of therapy or with abnormal cerebrospinal fluid that has not improved at that time.11 The Histiocyte Society has launched a revised protocol, HLH-2004, with use of cyclosporine A upfront in addition to etoposide and dexamethasone. However, we would not presently recommend cyclosporine A upfront in patients with avian influenza, since renal problems are frequent during H5N1 infection, and we therefore suggest investigating the published HLH-94 protocol as a first step.11 Etoposide could be considered to be initially used once (instead of twice) weekly for the first 2 weeks, because of the risk of myelotoxic effects. Additionally, we would suggest a reduction of every etoposide dose from 150 mg/m2 to 100 mg/m2 in patients aged 15 years or more (and further reduce the dose to 50 mg/m2 in middleaged and elderly patients), since children seem to tolerate the drug better than adults. The figure shows the 10 mg/m2 5 mg/m2
Daily dexamethasone dose (mg/m2)
2·5 mg/m2 1·25 mg/m2
Intravenous etoposide dose (150 mg/m2 per dose)
Week
1
2
3
4
5
6
7
8
9
Figure: Suggested therapy for patients with avian influenza A (H5N1) infection and secondary HLH This protocol is a modification of the HLH-94 protocol.11 Suggested doses are for children only.
871
Hypothesis
suggested treatment protocol for children with H5N1 and HLH. Similarly, the dose of dexamethasone can be reduced in adults. It may be too early to suggest intrathecal therapy for patients with H5N1 infection, since most patients with HLH respond favourably (with respect to neurological findings) to the combination of systemic dexamethasone and etoposide and do not receive intrathecal therapy,11 and dexamethasone is known to pass the blood-brain-barrier well. Whether treatment needs to be exactly 8 weeks as suggested above remains to be elucidated. Since EpsteinBarr virus is a DNA virus that is incorporated into the host (giving rise to extended illness), and H5N1 is an RNA virus (resulting in a short sharp illness), a treatment duration shorter than 8 weeks could be sufficient. Some patients with mild, early H5N1 disease could respond to mild therapy (eg, corticosteroids, intravenous immunoglobulins, or both), but results from EBV-HLH studies clearly indicate that early introduction of etoposide benefits patients with EBV-HLH.11,19–21 The HLH-2004 protocol presently states that the treatment is suggested for patients with severe, persistent, or recurrent disease (or with familial HLH) and accordingly, we suggest HLH therapy for patients with H5N1 and severe or persistent secondary HLH. The diagnostic criteria for HLH in HLH94 included fever, splenomegaly, bicytopenia, hypertriglyceridaemia or hypofibrinogenaemia (or both), and haemophagocytosis.15 For the HLH-2004 protocol, the criteria were revised, with three new criteria (ferritin 500 g/L, low natural killer cell activity, and soluble interleukin-2 receptor 2400 U/mL). Five of the eight criteria in total are now needed to make a diagnosis of HLH.31 Since the regions of the authors (Sweden and Hong Kong) are both free from human H5N1 infection, we cannot test this therapeutic hypothesis ourselves but welcome others to do so. We urge that each case in which HLH therapy is used is reported to the HLH Study Centre at the Karolinska Institute in Stockholm, Sweden. Obviously, we cannot guarantee any therapeutic efficacy, but it would be unfortunate if this therapy was effective but never attempted. Importantly, we want to emphasise that this hypothesis must not be misinterpreted as an instruction to treat influenza-like symptoms with etoposide. In addition to the reporting of all treated patients to the HLH Study Centre, we would welcome WHO to consider a platform for the undertaking of clinical trials based on a modified HLH protocol (including corticosteroids and etoposide) in addition to supportive and antiviral therapy. Contributors C-B Chow, supported by C-W Leung and Y-L Lau, prepared a management protocol for H5N1 and considered HLH treatment. J-I Henter supported this idea and developed it further. J-I Henter wrote the manuscript and the final version was approved by all the authors. Conflict of interest statement We declare that we have no conflict of interest.
872
References 1 Yuen KY, Chan PKS, Peiris M, et al. Clinical features and rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus. Lancet 1998; 351: 467–71. 2 Claas EC, Osterhaus AD, van Beek R, et al. Human influenza A H5N1 virus related to a highly pathogenic avian influenza virus. Lancet 1998; 351: 472–77. 3 Hien TT, Liem NT, Dung NT, et al. Avian influenza A (H5N1) in 10 patients in Vietnam. N Engl J Med 2004; 350: 1179–88. 4 Peiris JSM, Yu WC, Leung CW, et al. Re-emergence of fatal human influenza A subtype H5N1 disease. Lancet 2004; 363: 617–19. 5 Li KS, Guan Y, Wang J, et al. Genesis of a highly pathogenic and potentially pandemic H5N1 influenza virus in eastern Asia. Nature 2004; 430: 209–13. 6 Moscona A. Neuraminidase inhibitors for influenza. N Engl J Med 2005; 353: 1363–73. 7 de Jong MD, Tran TT, Truong HK, et al. Oseltamivir resistance during treatment of influenza A (H5N1) infection. N Engl J Med 2005; 353: 2667–72. 8 Normile D. Avian influenza. China will attempt largest-ever animal vaccination campaign. Science 2005; 310: 1256–57. 9 Ungchusak K, Auewarakul P, Dowell SF, et al. Probable person-toperson transmission of avian influenza A (H5N1). N Engl J Med 2005; 352: 333–40. 10 Taubenberger JK, Reid AH, Lourens RM, Wang R, Jin G, Fanning TG. Characterization of the 1918 influenza polymerase genes. Nature 2005; 437: 889–93. 11 Henter J-I, Samuelsson-Horne AC, Arico M, et al. Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunochemotherapy and bone marrow transplantation. Blood 2002; 100: 2367–73. 12 To KF, Chan PK, Chan KF, et al. Pathology of fatal human infection associated with avian influenza A H5N1 virus. J Med Virol 2001; 63: 242–46. 13 Henter J-I, Söder O, Hansson M, Elinder G, Andersson B, Andersson U. Hypercytokinemia in familial hemophagocytic lymphohistiocytosis. Blood 1991; 78: 2918–22. 14 Chokephaibulkit K, Uiprasertkul M, Puthavathana P, et al. A child with avian influenza A (H5N1) infection. Pediatr Infect Dis J 2005; 24: 162–66. 15 Henter J-I, Elinder G, Öst Å. Diagnostic guidelines for hemophagocytic lymphohistiocytosis. Semin Oncol 1991; 18: 29–33. 16 Henter J-I, Elinder G. Cerebromeningeal hemophagocytic lymphohistiocytosis. Lancet 1992; 339: 104–07. 17 Henter J-I, Nennesmo I. Neuropathological findings and neurological symptoms in 23 children with hemophagocytic lymphohistiocytosis. J Pediatr 1997; 130: 358–65. 18 de Jong MD, Cam BV, Qui PT, et al. Fatal avian influenza A (H5N1) in a child presenting with diarrhea followed by coma. N Engl J Med 2005; 352: 686–91. 19 Imashuku S, Hibi S, Ohara T, et al. Effective control of EpsteinBarr virus-related hemophagocytic lymphohistiocytosis with immunochemotherapy. Blood 1999; 93: 1869–74. 20 Imashuku S, Kuriyama K, Teramura T, et al. Requirement for etoposide in the treatment of Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. J Clin Oncol 2001; 19: 2665–73. 21 Imashuku S, Kuriyama K, Sakai R, et al. Treatment of EpsteinBarr virus-associated hemophagocytic lymphohistiocytosis (EBV-HLH) in young adults. Med Pediatr Oncol 2003; 41: 103–09. 22 Aleem A, Al Amoudi S, Al-Mashhadani S, Siddiqi N. Haemophagocytic syndrome associated with hepatitis-B virus infection responding to etoposide. Clin Lab Haematol 2005; 27: 395–98. 23 Fadeel B, Orrenius S, Henter J-I. Induction of apoptosis and caspase activation in cells obtained from familial haemophagocytic lymphohistiocytosis patients. Br J Haematol 1999; 106: 406–15. 24 Stepp SE, Dufourcq-Lagelouse R, Le Deist F, et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science 1999; 286: 1957–59.
www.thelancet.com Vol 367 March 11, 2006
Hypothesis
25
26
27
28
Feldmann J, Callebaut I, Raposo G, et al. Munc13-4 is essential for cytolytic granules fusion and is mutated in a form of familial hemophagocytic lymphohistiocytosis (FHL3). Cell 2003; 115: 461–73. 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: 827–34. Fadeel B, Orrenius S, Henter J-I. Familial hemophagocytic lymphohistiocytosis: too little cell death may seriously damage your health. Leuk Lymphoma 2001; 42: 13–20. Rimmelzwann GF, Kuiken T, van Amerongen G, Bestebroer TM, Fouchier RA, Osterhaus AD. Pathogenesis of influenza A (H5N1) virus infection in a primate model. J Virol 2001; 75: 6687–91.
www.thelancet.com Vol 367 March 11, 2006
29
30
31
Lipatov AS, Webby RJ, Goverkova EA, Krauss S, Webster RG. Efficacy of H5 influenza vaccines produced by reverse genetics in a lethal mouse model. J Infect Dis 2005; 191: 1216–20. Rivas Totino PR, Pratt Riccio EK, Corte-Real S, Daniel-Ribeiro CT, Ferreira-da-Cruz MD. Dexamethasone has pro-apoptotic effects on non-activated fresh peripheral blood mononuclear cells. Cell Biol Int 2005; published online Nov 2. DOI:10.1016/j.cellbi. 2005.09.002. Henter J-I, Tondini C, Pritchard J. Histiocytic syndromes. Crit Rev Oncol Hematol 2004; 50: 157–74.
873