Successful Allogenic Haematopoietic Stem Cell Transplantation in a Boy with Hemophilia A and MDS-RAEB

Hemophilia A is an X-linked bleeding disorder resulting from deficiency of clotting factor VIII (FVIII) due to mutations in the FVIII gene. It affects approximately one in 5 000 males (Bolton-Maggs PH et al., Lancet 2003; 361: 1801-1809). Modern medical management has improved the life expectancy of patients with haemophilia. This might go along with a higher cumulative risk to experience leukemia or other neoplasia later in their life. Incidence of HIV and Hepatitis C are still elevated among adult haemophilic patients and constitutes an increased risk for secondary malignancies (Dunn AL. Hemophilia 2010; 16: 427–436). In this regard an increasing number of patients with hemophilia A or other bleeding disorders patients might need HSCT in the upcoming years. The management of their bleeding risk will be a major clinical concern (Graf L et al., Hamostaseologie 2012; 32: 56–62). The liver is considered the major production site of FVIII. However, a number of preclinical studies reported a contribution of bone-marrow derived cells to FVIII-production (Follenzi A et al., Blood 2012; 119: 5532–5542). So far only little is known about the clinical relevance of this observation.

We describe a 4-year-old boy with mild congenital hemophilia A, who underwent allogeneic haematopoietic stem cell transplantation (HSCT) for myelodysplastic syndrome. In the second year of life, the boy first presented with easy bruising, prolonged partial thromboplastin time (PTT) (56 s) and marked thrombocytopenia (28 000/µl). Because of an unremarkable bone marrow (BM) analysis and a recent upper airway infection, initially the bleeding symptoms were interpreted as an idiopathic thrombocytopenic purpura and resolved spontaneously. At the age of 4, the boy presented again with increased bleeding symptoms and prolonged PTT. Further hemostaseological workup led to the diagnosis of mild Haemophilia A with a residual FVIII-activity of 18% in absence of FVIII-inhibitors. Molecular analysis revealed a novel missense mutation in exon 15 (c.5341T>C) of the FVIII gene predicting an amino acid exchange of Tyr1762His. Additionally, the patient had marked thrombocytopenia, anaemia and monocytosis. In vitro bleeding time (PFA 100) was prolonged and platelet α- and δ-granula secretion was impaired. BM examination revealed MDS-RAEB and monosomy 7 and the patient consequently qualified for allogeneic HSCT. In order to establish a regimen to prevent bleeding episodes during HSCT the efficacy of 1-Desamino-8-D-Arginin-Vasopressin (DDAVP) was first evaluated. After DDAVP-administration FVIII-activity increased to a maximum of 42% but fell back to baseline levels within only 6 h. We considered this response as inefficient especially with regard to possible side effects of repeated DDAVP-administration and opted for an intensified substitution treatment with human FVIII. Throughout the pre-transplant period, the patient was substituted with human FVIII prior to any invasive procedure. For the first BM aspiration he had received 500 IU (27.5 IU/kg) at 0, 12, 24, 48 and 72 h but still developed bleedings. For the second BM puncture he received 2000 IE (110 IU/kg) before and 3×1 000 IE (55 IU/kg)/d for 5 days and did not experience any further bleeding signs. Prior to the insertion of tunnelled central venous catheter the patient received 1 000 IU FVIII (55 U/kg), which led to an immediate increase of FVIII-activity to 90% and remaining activity of 35% after 6 h. On the first post-op day the patient received 4×500 IU of FVIII/d and 3×500 IU of FVIII/d until day 5 after surgery. This regimen prevented any bleeding complications and allowed maintaining FVIII-levels steadily between 25–50%.

The patient received a conditioning regimen with Busulfan, Cyclophosphamide, Melphalan and ATG and was transplanted with BM (4.7×10⁸ MNC/kg and 3.0×10⁶ CD34⁺ cells/kg) from an unrelated 9/10 HLA-identical donor. GvHD prophylaxis consisted of methotrexate and cyclosporine A. Based on the evaluation of the FVIII-activity kinetics, the patient was pre-emptively substituted with 2×500 IU F-VIII/day. This regimen allowed maintaining FVIII-activity levels steadily between 25 and 50%. Furthermore he was transfused with platelet concentrates every 48 h to maintain platelets>30 000 G/l from the day of his first ATG administration until platelet recovery. The patient did not present with any bleeding symptoms throughout the entire period of haematological aplasia. The preemptive FVIII- and platelet substitution was stopped upon successful platelet recovery on day+25. Up until then he had needed 15 platelet transfusions and 55 FVIII substitutions, FVIII-activity first declined after discontinuation of the FVIII- substitutions and reached a baseline of 25% after 6 days. In the following, the FVIII-activity spontaneously increased to a maximal level of 45%, remained elevated during 4 weeks, but then declined to pre-HSCT levels 2 months later. The elevated FVIII-activity was associated with increased Von-Willebrand factor (VWF) -activity and VWF-antigen. Otherwise, hematopoietic recovery occurred in time, the patient did neither experience severe treatment related toxicity nor GvHD. The clinical course was complicated by an adenovirus infection that resolved spontaneously. In regard to the MDS-RAEB, blood cell counts and HbF have normalized. At the last follow-up the patient was well and in stable hematological remission. F-VIII-activity was 19.6% and he did not present any bleeding symptoms.

So far, 4 children and 1 adult patient with hemophilia have been reported to have undergone HSCT (summarized in [Table 1]). Among these, one 2-year old patient with hemophilia A and high FVIII-inhibitor titer was transplanted to eradicate the inhibitor. Although hematological regeneration occurred rapidly and despite an intensive substitution regimen with coagulation factors, this patient presented severe bleeds throughout the entire HSCT period and eventually developed fatal septicemia (Uprichard J et al., Hemophilia 2010; 16: 143–147). Another patient with moderate hemophilia A and aplastic anemia received FVIII-substitutions after HSCT only upon clinical requirement and developed bleedings and at the catheter implantation site (Ostronoff M et al., Bone marrow transplantation 2006; 37: 627-628. Disruption of the mucosal barrier after myeloablative conditioning represents an increased risk for bleeding, infection and decreased wound healing. In order to minimize this risk we maintained FVIII-activity levels above 25%. 2 of the previously reported patients were treated with a similar strategy and did not encounter relevant bleeding (Buchbinder D et al., Pediatric transplantation 2013; 17: E20-E24; Caselli D et al., Hemophilia 2012; 18: e48-e49). Prior to HSCT we had likewise evaluated the possibility to prophylactically treat the patient with 1-Desamino-8-D-Arginin-Vasopressin (DDAVP) to prevent bleeding episodes during HSCT. A part from the fact, that the increase of FVIII after DDAVP-stimulation was not as efficient as expected the increase in VWF-antigen and FVIII-activity is known to decrease after repeated DDAVP-administration. With regard to possible side effects of recurrent DDAVP administration such as changes in electrolyte parameters and the risk for seizures and given the fact, that DDAVP is generally used for short interventions but not for a long-term bleeding prophylaxis we abolished this strategy and opted for an intensified F VIII substitution regimen. At the last follow-up (2 years post-HSCT) FVIII-activity levels in our patient were within the same range as before HSCT. This is consistent with the previous reports, which do not provide any evidence that donor derived hematopoietic cells increase FVIII-synthesis after HSCT (Uprichard J et al., Hemophilia 2010; 16: 143–147, Caselli D et al., Hemophilia 2012; 18: e48-e49).

Table 1 Summarised Results of Patients with Haemophilia and after HSCT.

Patients	Patient 1 (Turner M et al., BMT 1992: 9: 387–389)	Patient 2 (Ostronoff M et al., BMT 2006: 37: 627–628)	Patient 3 (Uprichard J et al., Hemophilia 2010: 16: 143–147)	Patient 4 (Buchbinder D et al.,. Ped. Transplant. 2013: 17:E20–E24)	Patient 5 (Caselli D et al., Hemophilia 2012: 18: e48–e49)	Patient 6 (Patient in this study)
Diagnosis	HIV, sec. Burkitt type ALL	Aplastic anemia	FVIII-inhibitors → HSCT for Immune tolerance induction	Osteopetrosis	ALCL	MDS-RAEB
Severity of haemophilia (FVIII-activity)	Severe (< 1%.)	Moderate (3%)	Severe (< 1%.)	Mild (16%)	Severe (0,6-1,2%l)	Mild (19%)
Age at HSCT in years	26	7	22	0.7	14	4
Donor	Sibling	Sibling	Unrelated	Unrelated	Unrelated	Unrelated
HLA-Match	10/10	10/10	10/10	n.i.	4/6	9/10
Stem Cell Source	BM	BM	BM	CB	CB	BM
Management for invasive procedures prior to HSCT	n.i. **	Platelet transfusion (< 50.000 G/l) rFVIII 60 U/kg before,2×30 U/kg for 3 days	rFVIIa	FVIII 50–75 IU/kg once daily	n.i.	F VIII 50 U/kg before, followed by 4×30 U/kg /d for 1 day, 3×30 U/kg /d for 4 days
Management Haemophilia Post HSCT	n.i.	Platelet transfusion every 2 days (< 30.000 G/l) rFVIII-substitution if needed	FVIII, rFVIIa FEIBA, tranexamic acid	Platelets<25.000 G/l FVIII 50-75IU/kg every 48 h	Platelet transfusion (< 30.000 G/l) rFVIII 2×1000 U/d	Platelet transfusion (< 30.000 G/l) FVIII 2×30 U/kg /d
Bleeding complications	n.i.	Joint bleeding, bleeding at catheter implantation site	Soft tissue bleeds	None	None	None
FVIII-activity after HSCT	n.i.	Unchanged	n.i.	Unchanged	Unchanged, prophylactic regimen	Short term increase, long term unchanged
Status at last follow up	Death after CNS relapse	Alive and well (4mo post HSCT)	Death of sepsis (d+46)	Impairment due to underlying disease	Alive and well (3 y post HSCT)	Alive and well (1,5y post HSCT)

FEIBA: (Factor-VIII-Inhibitor-bypass-activity), n.i.: not investigated

Surprisingly, a temporary increase of FVIII-activity occurred a few weeks after transplantation that can be explained by the elevated VWF activity and VWF-antigen (e. g. secondary to an elevated inflammatory state after HSCT or infection) entailing reduced FVIII cleavage. Consistent with this, the decline of FVIII-activity coincided with the normalization of VWF-activity and VWF-antigen. One might likewise argue that VWF and FVIII represent acute phase proteins and that the temporary VWF-Ag- and FVIII-increase represent an acute phase reaction (O’Donnell J et al., Thromb Haemost 1997; 77: 825-828).

On the other hand, one might hypothesize that donor derived cells had intermittently engrafted and contributed to FVIII-synthesis in the patient’s liver before exhausting (due to lack of self-renewal capacity). Indeed, there are some indications for the contribution of hematopoietic progenitor cells to FVIII-production: (I) cord blood (CB) and BM-derived CD34⁺ cells as well as platelets express FVIII-mRNA. (II) Preclinical studies have shown that murine donor-BM derived mononuclear or stromal cells can engraft in the recipient’s liver and contribute to FVIII-production. However, additional chemical induction of hepatic and endothelial damage in the recipient animals was necessary to observe relevant contribution to FVIII-production (Follenzi A, et al., Blood 2012; 119: 5532–5542). Most recently, Zanolini et al. demonstrated that human CB-derived CD14⁺ cells express FVIII and improve FVIII-activity when transferred into humanized hemophilic mice (Zanolini D et al., Haematologica 2015; 100: 881–892).

Although the experience is very limited, current clinical studies do not provide evidence for a relevant contribution of donor BM-derived progenitors to FVIII-synthesis.