Genotype–phenotype study of familial haemophagocytic lymphohistiocytosis due to perforin mutations
- A Trizzino1,
- U zur Stadt2,
- I Ueda3,
- K Risma4,
- G Janka2,
- E Ishii5,
- K Beutel2,
- J Sumegi4,
- S Cannella1,
- D Pende6,
- A Mian7,
- J-I Henter8,
- G Griffiths9,
- A Santoro1,10,
- A Filipovich4,
- M Aricò1,
- for the Histiocyte Society HLH Study group
- 1Pediatric Hematology Oncology, Ospedale dei Bambini ”G. Di Cristina”, ARNAS Civico, Palermo, Italy
- 2Department of Pediatric Hematology Oncology, University Medical Center Hamburg Eppendorf, Hamburg, Germany
- 3Department of Pediatrics, Kyoto Prefectural University of Medicine, Kyoto, Japan
- 4Cincinnati Children's Hospital Medical Center, OH, USA
- 5Department of Pediatrics, Ehime University Graduate School of Medicine, Ehime, Japan
- 6Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy
- 7Arkansas Children Hospital, Arkansa, USA
- 8Childhood Cancer Research Unit, Department of Woman and Child Health, Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden
- 9Cambridge Institute for Medical Research, Addenbrooke's Hospital, Cambridge, UK
- 10Ematologia I, A.O. Cervello, Palermo, Italy
- M Aricò, Pediatric Hematology Oncology, Ospedale dei Bambini ”G. Di Cristina”, Via Benedettini 1, 90134 Palermo, Italy; arico{at}ospedalecivicopa.org
- Received 29 June 2007
- Revised 16 August 2007
- Accepted 19 August 2007
- Published Online First 14 September 2007
Abstract
Background: PRF1 gene mutations are associated with familial haemophagocytic lymphohistiocytosis type 2 (FHL2). Genotype–phenotype analysis, previously hampered by limited numbers of patients, was for the first time performed by data pooling from five large centres worldwide.
Patients and methods: Members of the Histiocyte Society were asked to report cases of FHL2 on specific forms. Data were pooled in a common database and analysed.
Results: The 124 patients had 63 different mutations (including 15 novel mutations): 11 nonsense, 10 frameshift, 38 missense and 4 in-frame deletions. Some mutations were found more commonly: 1122 G→A (W374X), associated with Turkish origin, in 32 patients; 50delT (L17fsX22) associated with African/African American origin, in 21 patients; and 1090-91delCT (L364fsX), in 7 Japanese patients. Flow cytometry showed that perforin expression was absent in 40, reduced in 6 and normal in 4 patients. Patients presented at a median age of 3 months (quartiles: 2, 3 and 13 months), always with fever, splenomegaly and thrombocytopenia. NK activity was absent in 36 (51%), ≤2% in 18 (26%), 3–≤5% in 10 (14%), >5% in 4 (6%), “reduced” in 2 (3%) (not reported, n = 54). Nonsense mutations were significantly associated with younger age at onset (p<0.001) and absent natural killer activity (p = 0.008).
Conclusion: PRF1 mutations are spread over the functional domains. Specific mutations are strongly associated with Turkish, African American and Japanese ethnic groups. Later onset and residual cytotoxic function are observed in patients with at least one missense mutation.
Footnotes
-
Competing interests: None declared.








