Coagulation Functions with Clinical Phenotype and Inhibitory Mechanisms in Acquired Factor V Inhibitors.

Abstract 2232

Development of acquired factor V (aFV) inhibitors rarely occurs, but its clinical phenotype varies from asymptomatic to life-threatening bleeding. A recent systematic report describes that little bleeding symptom is present in 20% at the diagnosis for patients with acquired FV (aFV) inhibitors. However, the coagulation function and its mechanism(s) on the different clinical phenotype are poorly understood. In this study, we examined the coagulation function on aFV inhibitors by using comprehensive coagulation assays, thrombin generation test (TGT) and clot waveform analysis (CWA). TGT was performed using tissue factor (0.5 pM), phospholipid (PL; 4 μM) and ellagic acid (0.3 μM). CWA, that evaluates the parameters of min1 as maximum coagulation velocity and min2 as maximum coagulation acceleration, was performed on MDA-II^® system. We tested 7 cases with aFV inhibitors. Four cases were asymptomatic (FV:C; 3.6±3.4 IU/dl, inhibitor 5.8±3.3 BU/ml: non-B group), and 3 cases had severe bleeding tendency (2.9 ± 4.5 IU/dl, 66 ±51 BU/ml; B group). In TGT, all cases in both groups little showed the thrombin generation within 60 min, independently of FV:C level and clinical phenotype, showing little informative in functional evaluation for aFV inhibitors. However, in a PT-based CWA, the clotting time observed in non-B group was markedly shorted compared to that in B group (62.2±17.0/112±15 sec; p=0.006). In addition, both parameters in non-B group were significantly greater than those in B group ( min1 ; 2.89±1.10/0.98±0.29 dT/dt; p=0.014) and ( min2 ; 0.75±0.40/0.15±0.07 d²T/dt²; p=0.028), suggesting that CWA was useful for the prediction and monitoring of hemorrhagic symptoms in patients with aFV inhibitors. To confirm the distinct mechanism(s) on both groups, the IgGs from aFV plasmas were immune-purified using protein G-Sepharose. In the reactant mixtures with normal plasma and aFV IgGs, all parameters obtained in CWA were similar to those obtained in patients' plasmas. SDS-PAGE and western blotting revealed that 2 cases in B group reacted the light chain of FV(a). However, 2 cases in non-B group reacted the heavy chain, and other 2 cases were reacted with both chains (heavy>light), indicative of the distinct epitopes of IgGs in both groups. Since the light chain contains the PL-binding site(s), the effects of aFV IgGs were examined on the FV-PL binding in an ELISA. All IgGs in B group inhibited this binding (by 40–90%) dose-dependently, whilst little affected in non-B group. Since FV acts as a cofactor of activated protein C (APC) on inactivation of FVIIIa, the effects of aFV IgGs on the ability of APC on FVIIIa inactivation were examined using intrinsic FXa generation assay. The APC sensitivity ratio (APCsr) was expressed as ratio of amounts of generated FXa in the absence of APC relative to its presence. A low level of APCsr indicates the reduction in FVIIIa inactivation, consequently APC resistance. APCsr values in B group were >2.0 within normal range, whilst those in non-B group were decreased to 1.5, supportive of APCR in non-B group. Based on these findings, we propose that severe bleeding tendency in B group would be appeared through negligible prothrombinase activity, since aFV IgGs blocked the FV(a)-PL binding. While, clinical phenotype in non-B group would be asymptomatic, since aFV IgGs unaffect the FV(a)-PL binding and further cause the APC resistance. In addition, various clinical phenotypes in aFV inhibitors appear to be dependent on the recognizing epitope of these IgGs.