Haemophagocytic lymphohistiocytosis (HLH) is characterised by fever, hepatosplenomegaly, central nervous system symptoms, cytopenia, coagulopathy, and lipid changes because of pathological immune activation, hypercytokinaemia and organ infiltration by phagocytosing histiocytes. Despite being an aggressive disease, effective treatment does exist. A treatment protocol was firstly designed in 19941 and later revised in 2004 by the HLH Study Group of the Histiocyte Society.2 Since the implementation of the treatment protocol of HLH 1994, its 5-year survival rate has improved from around 20% to more than 50%.1

Notably, HLH is comprised of two different conditions: familial/primary HLH (FHL) and secondary HLH (Table 1),3 with the former being an autosomal recessive condition. Five mutations that lead to FHL (Nos 1-5) have now been identified and the underlying genetic defect described in four. They are: PRF1, UNC13D, STX11, and STXBP2. For genetic defect in FHL1, the potential gene locus has been identified but not the specific genetic defect. The perforin gene (PRF1) mutation is the first genetic defect described in FHL (FHL2) and accounts for 20% to 50% of all affected FHL families identified in a Japanese study.4 Perforin is a soluble, pore-forming cytolytic protein synthesised in cytotoxic lymphocytes. This molecule plays a crucial role in regulating the access of granzymes to the cytosol of target cells, where they cleave key substrates to initiate apoptotic cell death. It is sequestered, along with granzyme serine proteases, in secretory cytotoxic granules. The PRF1 mutation results in reduction of perforin protein production and its cytotoxic function. This in turn impairs the control of lymphocyte homeostasis during immune responses and leads to hypercytokinaemia and continued expansion of populations of histiocytes and activated cytotoxic lymphocytes.5 Patients without an identifiable genetic cause but with a clear familial history of HLH are also classified as having FHL.

The differentiation of primary and secondary HLH is notoriously difficult. The only definite way is by genetic study to find the causative mutation. In Asia, the perforin gene mutation has mostly been identified in HLH patients in Japan.6 7 The case reported here is the first and the only HLH perforin gene mutation identified in Hong Kong.

The patient was the first child in a non-consanguineous family, with no history of previous unexplained deaths in the parents’ families. The child presented at 34 days of life with fever, hepatosplenomegaly, and pancytopenia in June 2009. The ferritin level was markedly increased to 18 513 (reference range [RR], 29-333) pmol/L and there was hypofibrinogenaemia with a level of 0.54 (RR, 1.85-3.83) g/L. Bone marrow examination showed features of haemophagocytosis. The diagnostic criteria for HLH were therefore met. Initial magnetic resonance imaging (MRI) of the brain and cerebrospinal fluid examination were normal. Virology investigations including serology for Epstein-Barr virus (EBV), human herpesvirus 6, herpes simplex virus, and cytomegalovirus were all negative. The patient was treated according to HLH 2004 protocol with dexamethasone, cyclosporine A, and etoposide. Her clinical condition deteriorated with severe metabolic acidosis and she underwent haemodialysis. She experienced persistent neutropenia after the first dose of etoposide. Repeat bone marrow examinations showed markedly depressed granulopoiesis with residual haemophagocytic activity. Further doses of etoposide were therefore withheld while dexamethasone and cyclosporine A were continued. The first course of chemotherapy was stopped after the 11th week of treatment. However, the patient had a relapse of HLH 3 weeks after stopping chemotherapy (manifesting as fever and hepatosplenomegaly). Repeat bone marrow examination confirmed the presence of haemophagocytic activity, for which treatment with dexamethasone, etoposide, and cyclosporine A was restarted. She developed progressive metabolic acidosis8 that was once again treated by haemodialysis. Her condition then became stabilised.

Since this patient presented at early age and had a recurrence after cessation of chemotherapy, she was suspected to suffer from FHL, for which the ultimate treatment is haematopoietic stem cell transplant (HSCT). Search for a related or unrelated donor was started while the patient continued to receive chemotherapy.

While waiting for the HSCT, the patient developed tremors of the lower limbs, and bilateral ankle clonus, limb spasticity and intermittent squints (with no definite visual fixation) were noted. The developmental age regressed from 8 to 3 months, whilst brain MRI revealed diffuse parenchymal and leptomeningeal enhancing lesions suggestive of lymphohistiocytic infiltration. Cerebrospinal fluid also showed presence of pleocytosis and a lymphohistiocytic infiltrate. The patient was diagnosed to have central nervous system involvement by HLH. Three doses of intrathecal chemotherapy with methotrexate 6 mg and hydrocortisone 8 mg were given over a 10-day period.

The patient received an unrelated double-unit cord blood transplant after conditioning with oral busulphan (23 mg/kg), etoposide (30 mg/kg), cyclophosphamide (120 mg/kg), and thymoglobulin (7.5 mg/kg). She had a neutropenic fever on post-transplant day 12. Donor cell engraftment was achieved on post-transplant day 16. Regrettably, she developed veno-occlusive disease causing hyperbilirubinaemia, fluid retention, progressive hepatosplenomegaly, and respiratory distress. The maximum bilirubin level was 209 μmol/L. Despite intensive care unit treatment, intubation, and positive pressure ventilation, the patient developed respiratory failure and died on day 45 after cord blood transplant. The parents refused a full postmortem. A postmortem liver biopsy showed marked sinusoidal dilatation and congestion with atrophy of central hepatocytes. These features were compatible with sinusoidal obstruction due to veno-occlusive disease.

Genetic analysis of the patient and the parents’ blood was performed, with the coding region of the perforin gene in exons 2 and 3 amplification by a polymerase chain reaction. This revealed a heterozygous one base pair deletion (65 delC) in exon 2 in the patient and her father. There was another mutation 853-855 del AAG in exon 3 of patient and her mother. Collectively, the patient had compound heterozygous mutations of the perforin gene, namely 853-855 del AAG and 65delC.

Previously we reported our seven consecutive cases of HLH encountered from 1991 to 2006.9 Since then, there had been four other patients. Apart from showing haemophagocytosis in bone marrow or lymph node or both (Table 2), all eleven patients had fever, splenomegaly, and cytopenia in at least two cell lines. They also had markedly elevated ferritin levels with or without a raised fasting triglyceride level or hypofibrinogenaemia. The standard diagnostic criteria for HLH were met in all patients. Another teaching hospital in Hong Kong reported nine cases from 1991 to 2006.10 The overall survival of all 20 patients was 58%.

Interestingly, EBV infection was confirmed (by immunoglobulin M vs EBV or EBV DNA) in 11 (55%) of the 20 patients. The overall survival of EBV-related HLH was 55%. In eight patients, their secondary HLH was related to malignant histiocytosis, Still’s disease, and anaplastic large cell lymphoma. Two patients had X-linked lymphoproliferative disorders.10 They all had primary EBV infection and one had died. The other patient received a mismatched unrelated cord blood transplant and had a full recovery without recurrence.

Four patients in our unit were investigated for perforin gene mutations; only one whom had an abnormality (compound heterozygous mutations in the gene). Hence this patient is the first reported case in Hong Kong with a perforin gene mutation causing FHL.

In Asia, most perforin gene mutations of HLH patients have been reported from Japan. Ueda et al6 reported five of 14 HLH patients with perforin gene abnormalities. The 1090-1091delCT and 207delC mutations of the perforin gene were frequently present in Japanese HLH patients. Ueda et al4 also reported a collaborative study which did not show PRF1 gene mutations from Korea (n=4), Malaysia (n=3), Hong Kong (n=2), Australia (n=1), and Taiwan (n=1). Lee et al11 from Taiwan reported 26 HLH patients; none of whom had PRF1, Mun12-4, STX11, or SH2D1A mutations. There was only one case report of a heterozygous PRF1 mutation (Arg390stop) in a young Taiwanese girl who presented with a panniculitis-like T-cell lymphoma and subsequently endured fatal HLH.12

Among the 20 patients in Hong Kong, only one had a PRF1 gene mutation and two had X-linked lymphoproliferative disorders. However, the actual rate of genetic abnormalities in HLH patients remains unknown as not all patients with HLH had genetic testing. This is partially due to inconsistent protocols for genetic investigations in this disease entity and inadequate laboratory support for genetic tests. Clearly, revision of the local clinical and laboratory service protocol is warranted, and more importantly, an international multicentre collaboration to improve immunological assessment and genetic analysis of HLH patients should be promoted.

1. Henter JI, Arico M, Egeler RM, et al. HLH-94: a treatment protocol for hemophagocytic lymphohistiocytosis. HLH study Group of the Histiocyte Society. Med Pediatr Oncol 1997;28:342-7. CrossRef

2. Henter JI, Home A, Aricó M, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007;48:124-31. CrossRef

3. Freeman HR, Ramanan AV. Review of haemophagocytic lymphohistiocytosis. Arch Dis Child 2011;96:688-93. CrossRef

4. Ueda I, Ghim T, Peng LH, et al. Proceedings of the 2nd Congress Asian Society for Pediatric Research; 2006 Dec 8-10; Yokohama, Japan.

5. Usmani GH, Woda BA, Newburger PE. Advances in understanding the pathogenesis of HLH. Br J Haematol 2013;161:609-22. CrossRef

6. Ueda I, Morimoto A, Inaba T, et al. Characteristic perforin gene mutations of haemophagocytic lymphohistiocytosis patients in Japan. Br J Haematol 2003;121:503-10. CrossRef

7. Ueda I, Kurokawa Y, Koike K, et al. Late-onset cases of familial hemophagocytic lymphohistiocytosis with missense perforin gene mutations. Am J Hematol 2007;82:427-32. CrossRef

8. Hui WF, Luk CW, Chan WK, Miu TY, Yuen HL. Severe lactic acidosis in an infant with haemophagocytic lymphohistiocytosis. Hong Kong J Paediatr (New Series) 2012;17:183-9.

9. Chan JS, Shing MM, Lee V, Li CK, Yuen P. Haemophagocytic lymphohistiocytosis in Hong Kong children. Hong Kong Med J 2008;14:308-13.

10. Ho MH, Cheuk DK, Lee TL, Ha SY, Lau YL. Haemophagocytic lymphohistiocytosis in Hong Kong children have a wider clinical spectrum. Hong Kong Med J 2008;14:503-4.

11. Lee WI, Chen SH, Hung IJ, et al. Clinical aspects, immunologic assessment, and genetic analysis in Taiwanese children with hemophagocytic lymphohistiocytosis. Pediatr Infect Dis J 2009;28:30-4. CrossRef

12. Chen RL, Hsu YH, Ueda I, et al. Cytophagic histiocytic panniculitis with fatal haemophagocytic lymphohistiocytosis in a paediatric patient with perforin gene mutation. J Clin Pathol 2007;60:1168-9. CrossRef