Mapping T-Cell Peptide Epitopes on Activated Recombinant Factor VII (rFVIIa) Contributing to Inhibitor Antibody Response

Bypassing agents such as activated recombinant factor VII (rFVIIa) are used to treat acute bleeding episodes in patients with hemophilia and inhibitors to coagulation factors VIII (FVIII) or IX. BAY 86-6150 is a modified rFVIIa protein with 6 amino acid substitutions in the rFVII molecule that prolong half-life and improve potency compared with the currently available rFVIIa.

In a dose-escalation clinical study, 1 patient out of a cohort of 10 treated with 6.5 μg/kg BAY 86-6150 developed low-titer neutralizing antibodies that were detected after the third exposure. The patient's anti–BAY 86-6150 antibodies also cross-reacted with and neutralized wild-type FVIIa (WT-FVIIa) in vitro. T-cell epitope mapping was performed to identify BAY 86-6150 sequence(s) that may have contributed to the immunogenic response in the patient by measuring CD4+ T-cell response to individual 15-mer peptides spanning BAY 86-6150.

The epitope mapping study did not identify any of the 14 peptides unique to BAY 86-6150 as epitopes recognized by the patient's T cells. However, strong responses were detected against 2 WT-FVIIa peptides, WT p6-20 (EELRPGSLERECKEE) and WT p156-170 (GKVCPKGECPWQVLL), indicating that CD4+ T helper cells recognizing these WT peptides may have contributed to the immune response that resulted in the production of anti–BAY 86-6150 antibodies during treatment. It should be noted that although the patient had no detectable anti-FVIIa antibodies before the start of the study, he had been treated with factor eight inhibitor bypassing activity (FEIBA), which contains active FVII, on 3 consecutive days 3 months before entry into the study. Hence it is possible that the patient had been primed for a response against FVIIa that was triggered by the subsequent exposure to BAY 86-6150.   

The fact that the patient's T cells only responded to WT-FVIIa peptides might be explained if the WT-FVIIa peptides were seen as foreign/non-self peptides by his T-cell repertoire. Unfortunately, the patient's FVII gene sequence was not obtainable, and this possibility remains unanswered.

The Universal Protein Resource (UniProt) database reports 2 natural FVII gene variants that encompass the WT p6-20 sequence (Millar, et al. Hum Genet. 2000;107:327-42; Herrmann, et al. Haemophilia. 2009;15:267-280) and one for the WT p156-170 sequence (Wulff and Herrmann. Hum Mutat. 2000;15:489-496) that resulted from single amino acid substitutions. Hence the possibility exists that the patient's T-cell repertoire sees WT-FVIIa sequences as foreign and immunogenic. T cells from 40 healthy donors were also tested for reactivity against the peptide panel to assess relative immunogenicity of BAY 86-6150 vs WT-FVIIa. Seven BAY 86-6150 neoepitopes were stimulatory for 8 unrelated healthy donor T cells, but their mean stimulation indices were not statistically higher than those observed against WT FVIIa peptides. Mean % stimulation values of the positive responses against the 7 BAY 86-6150 neoepitopes versus 39 WT FVIIa peptides were 1.09 (N=13 responses) vs 1.01 (N=70 responses; P=0.5059). Statistical analysis of the in vitro T-cell response indicates that specific mutations to BAY 86-6150 do not result in BAY 86-6150 being more immunogenic than WT FVIIa. This would suggest that BAY 86-6150 will not elicit stronger or higher frequency of anti-FVIIa antibody response than WT FVIIa in patients with hemophilia.