The aetiology of anti-factor VIII (FVIII) autoantibody formation in acquired haemophilia remains unknown. We hypothesised that encounter of antigenically different, allogeneic FVIII may challenge inhibitor formation after presentation on MHC class II. Eighteen consecutive cases with acquired haemophilia were enrolled (9 females, 9 males). A control group comprised 50 male and 50 female healthy blood donors. The coding region of the FVIII gene and the HLA-DRB1 genotype were studied. The presentation of foreign FVIII variants on the patient's MHC class II alleles was predicted using SYFPEITHI algorithm. A rare FVIII variant (E2004K) was found in 1 patient with acquired haemophilia after massive transfusion; the 2004K allele was predicted to be presented on the patient's HLA-DRB1*0101. Moreover, distribution of a polymorphism (D1241E) was significantly skewed comparing patients and controls. 3 of 3 patients with transfusion-associated disease carried 1241D in homozygous or hemizygous form and were predicted to present 1241E (foreign), but not 1241D (self), on their HLA-DRB1*0301. Therefore, encounter of 1241E may result in the presentation of a new T cell epitope in these patients. The same conditions were not found in any patient with acquired haemophilia of other causes. The expected frequency in the general Caucasoid population undergoing transfusion is 3 to 4 %. In conclusion, encounter of variant allogeneic FVIII presented on a suitable MHC background could be a risk factor for inhibitor formation.Response to Reviewers: Dear Professor Eichinger, Thank you for considering our manuscript for publication in Annals of Hematology. We would also like to thank the reviewer for valuable comments. We have addressed the comments as follows:
*Only 3 cases of transfusion-associated acquired hemophilia: Indeed, acquired hemophilia associated with transfusion has rarely been reported. In our series, 3 cases among 18 consecutive patients are even more than expected from the literature. Our study suggest that immune recognition of certain FVIII variants could play a role in the pathogenesis, but formal proof would require much higher numbers of transfusion-associated cases that, in all probability, will never be collected.
*Transfusion in the other cases: No, transfusion was not documented in any other case of our series. Medical histories were taken by the authors or other well-trained physicians of the department.We are sure that an association with transfusion would not have been missed in these patients. We have acknowledged this fact in the Methods section (p. 4, first para).
*Frequency of 1241E in blood donations: Based on the known allelic distribution of the 1241E allele and the known sex distribution of blood donors, we determined the frequency of 1241E in singledonor blood products as 21 % (p. 4, last para).
*Transfused units and interval between transfusion and acquired hemophilia: all transfusionassociated cases received multiple transfusions. Time between first transfusion and acquired hemophilia ranged...