Characterization of a novel mutation in the F8 promoter region associated with mild hemophilia A and resistance to DDAVP therapy.

We read with interest the recent letter by Dai et al. characterizing a patient with a C>T substitution at position –219 in the F8 promoter region resulting in mild hemophilia A [1]. We report here a novel mutation in the same region occurring in a family group including five patients (Fig. 1A) with factor VIII levels and a clinical phenotype compatible with mild hemophilia A. Four of them (patients IV:1, IV:2, IV:3, and IV:4) were tested with desmopressin (DDAVP; 30 μg/kg given subcutaneously) with no response (FVIII:C levels unchanged 60, 120 and 240 min after DDAVP administration). By contrast, the von Willebrand factor (VWF) response was preserved, with marked increases in VWF antigen and VWF ristocetin cofactor activity levels (Fig. 1B). Figure 1.  (A) Pedigree of mild hemophilia A family. Patients’ basal FVIII:C levels are as follows: IV:1, 22 IU dL–1; IV:2, 29 IU dL–1; IV:3, 28 IU dL–1; IV:4, 21 IU dL–1. (B) DDAVP test: means (± standard deviation) of FVIII:C, von Willebrand factor antigen (VWF:Ag) and von Willebrand factor ristocetin cofactor activity (VWF:RCo). Download figure to PowerPoint Mutation screening [2] by denaturing high-performance liquid chromatography analysis, followed by direct sequencing of all 26 exons and exon–intron junctions, failed to detect any variation from the normal control. Type 2N von Willebrand disease and a combined FV–FVIII deficiency were also ruled out. Patients were then treated with recombinant FVIII products during hemorrhagic episodes, without the development of inhibitors. However, a subsequent analysis of the F8 promoter performed by polymerase chain reaction followed by dideoxy sequencing identified a T>G substitution at position −257 (standard Human Genome Variation Society nomenclature [3]) in the four living probands (IV:1, IV:2, IV:3, and IV:4). The lack of this substitution in 100 normal chromosomes enables us to exclude the possibility that this base change is a common, non-pathogenic polymorphism. This mutation was found, in heterozygosis, also in the patients’ mothers (III:2 and III:5) but was not found in the patients’ fathers (III:1 and III:4) and in the unaffected son (IV:5), providing confirmation that the mutation is linked to the disease. Although several mutations in F8 have been reported so far [4], a minority of patients remain without a molecular characterization of the causative mutation. Mutations in the F8 promoter region have been described only rarely [1,5] and, to our knowledge, the mutation found in this family has not been previously reported. Interestingly, polymorphisms in the F8 promoter region have been recently found to influence FVIII synthesis. In our case, the −257T>G mutation is located in a promoter region that contains important positive cis-acting elements; it is well known that characterized binding sites for C/EBPα and C/EBPβ transcription factors and other transcription factors of unknown identity are located in this region [6]. Although a number of investigators have studied the hemostatic effects of DDVAP, the exact cellular mechanisms underlying the DDAVP-induced FVIII and VWF secretion are still to be fully elucidated. Indeed, although it is well known that VWF is synthesized and secreted from endothelial cells, where it is stored in intracytoplasmic Weibel–Palade bodies [7], the FVIII secretory pool is hypothesized to be localized in sinusoidal cells of the liver. The literature data suggest that the DDAVP-induced contemporary secretion of FVIII and VWF happens only if there is concomitant synthesis of these two hemostatic proteins [8]. Thus, the mutation found in our study, causing reduced FVIII synthesis (and probably depletion of the secretory pool), explains well the absence of response of FVIII but not of VWF. These findings, which could have important clinical and therapeutic implications, need to be confirmed by further reports on this or similar mutations.