Inhibitory antibodies (inhibitors) to exogenous factor VIII (FVIII) are the major complication of treatment with recombinant FVIII for patients with hemophilia A. Inflammatory “danger signals”, such as would be produced by surgical injury, have been hypothesized as important acquired determinants of inhibitor risk. Surgery has been associated with an increased inhibitor risk in observational studies of humans. However, it is not possible to experimentally separate surgical injury from intensive FVIII exposure in humans, and this relationship has not previously been explored in animal models of hemophilia A. We investigated the relationship between surgical injury and FVIII inhibitor development in two distinct mouse models of hemophilia A.
The hemophilia A models used were F8 exon 16 knockout (E16KO) mice and F8 exon 17 knockout mice with the humanized major histocompatibility complex II allele HLA-DRB1*1501 (E17KO/hMHC). The model surgical procedure was an open-and-close laparotomy under isofluorane anaesthesia at 6 to 12 weeks of age. Recombinant human FVIII, with or without lipopolysaccharide (LPS), was administered intravenously through the tail vein. Preoperatively, FVIII was given in doses of 2 units (approximately 0.1 mcg/unit) weekly for 4 weeks; postoperatively FVIII was given in doses of 2 to 6 units daily for 3 to 5 days. Blood was obtained via retro-orbital sampling or cardiac puncture, and plasma was separated by centrifugation. Plasma was assayed for IL-1 and IL-6 concentration by ELISA, for anti-FVIII IgG by ELISA, and for inhibition of FVIII coagulant activity (FVIII:C) by Bethesda assay. Single cell suspensions of splenocytes were analyzed by flow cytometry. Data was analyzed using Student t tests, Mann-Whitney U tests, and chi-square tests. All animal experiments were approved by the Animal Care Committee at Queen's University.
In E16KO mice, surgery produced greater than 5-fold increases in IL-1 levels (p=0.0005) and 150-fold increases in IL-6 levels (p=0.000023), compared to FVIII controls in the 24 hrs after the surgical injury. Surgery also resulted in significantly increased expression by splenic dendritic cells of CD80 (p=0.0004), but not of CD40 or CD86. All E16KO mice developed high-titre antibodies (by both ELISA and Bethesda assay) and there was no difference in Bethesda titres between surgery and control groups, either for mice FVIII-naïve at time of surgery (p=0.27), or for mice with FVIII exposure prior to surgery (p=0.66). In contrast to E16KO hemophilic mice, FVIII immune responses were seen in only some E17KO/hMHC animals. Proportions of FVIII-naïve E17KO/hMHC mice that did develop detectable antibodies were similar between surgery and control groups (47% vs. 53% for ELISA, p=0.72; 7% vs. 22% for Bethesda, p=0.19); among mice who developed antibodies, ELISA and Bethesda titres were not significantly different between surgery and control groups (p>0.05). E17KO/hMHC mice were exposed to FVIII, and those that were tolerant to FVIII (ie. did not have antibodies detectable by ELISA) then had either surgery with post-operative FVIII exposure, FVIII without surgery, or FVIII and LPS. No FVIII-tolerant E17KO/hMHC developed antibodies detectable by ELISA after surgery or after FVIII exposure alone, but 100% of FVIII-LPS-exposed mice developed antibodies (p=0.001).
Although laparotomy produces acute increases in the inflammatory cytokines IL-1 and IL-6 and upregulates expression of the costimulatory molecule CD80 on antigen presenting cells, E16KO mice that underwent laparotomy did not have greater immunologic responses to FVIII than those who did not. FVIII-naïve E17KO/hMHC mice who underwent surgery were no more likely to develop detectable antibodies than those who did not, and surgery did not result in higher-titre immune responses. In E17KO/hMHC mice who were immunologically tolerant to FVIII, surgery did not break this tolerance, although coadministration of LPS and FVIII did; tolerance could be broken in these mice, but was not broken by surgical injury. Given the limitations of existing clinical research in this area, our results indicate that understanding of specific inflammatory stimuli involved in FVIII inhibitor development in humans should be obtained before proceeding with the assumption that surgical injury alone is a significant risk factor for inhibitors in patients with hemophilia A.
Moorehead:Baxter Biosciences: Honoraria, Travel funding for educational meetings Other; Bayer: Honoraria, Travel funding for educational meetings, Travel funding for educational meetings Other. Steinitz:Baxter BioScience: Employment. Reipert:Baxter Biosciences: Employment.
Asterisk with author names denotes non-ASH members.