Targets of autoantibodies in acquired hemophilia A are not restricted to factor VIII: data from the GTH-AH 01/2010 study

Abstract

The root cause of autoantibody formation against factor VIII (FVIII) in acquired hemophilia A (AHA) remains unclear. We aimed to assess whether AHA is exclusively associated with autoantibodies toward FVIII or whether patients also produce increased levels of autoantibodies against other targets. A case-control study was performed enrolling patients with AHA and age-matched controls. Human epithelial cell (HEp-2) immunofluorescence was applied to screen for antinuclear (ANA) and anticytoplasmic autoantibodies. Screening for autoantibodies against extractable nuclear antigens was performed by enzyme immunoassay detecting SS-A/Ro, SS-B/La, U1RNP, Scl-70, Jo-1, centromere B, Sm, double-stranded DNA, and α-fodrin (AF). Patients with AHA were more often positive for ANA than control patients (64% vs 30%; odds ratio [OR] 4.02, 1.98-8.18) and had higher ANA titers detected than controls. Cytoplasmic autoantibodies and anti-AF immunoglobulin A autoantibodies were also more frequent in patients with AHA compared with controls. Autoantibodies against any target other than FVIII were found in 78% of patients with AHA compared with 46% of controls (OR 4.16, 1.98-8.39). Results were similar preforming sensitivity analyses (excluding either subjects with autoimmune disorders, cancer, pregnancy, or immunosuppressive medication at baseline) and in multivariable binary logistic regression. To exclude that autoantibody staining was merely a result of cross-reactivity of anti-FVIII autoantibodies, we tested a mix of 7 well-characterized monoclonal anti-FVIII antibodies. These antibodies did not stain HEp-2 cells used for ANA detection. In conclusion, a diverse pattern of autoantibodies is associated with AHA, suggesting that a more general breakdown of immune tolerance might be involved in its pathology.

Graphical abstract

Figure 1.

Representative HEp-2 immunofluorescent patterns of patients with AHA. (A) ANA (1:320) of nuclear speckled pattern with few nuclear dots in an 85-year-old male patient with FVIII:C of 2.2 IU/dL and an inhibitor of 4 BU/mL. (B) ANA (1:640) of nuclear homogeneous pattern in an 86-year-old male patient with FVIII:C of 2.4 IU/dL and an inhibitor of 9 BU/mL. (C) ANA (1:640) of nucleolar pattern in an 81-year-old female patient with FVIII:C of 2.5 IU/dL and an inhibitor of 104 BU/mL. (D) Cytoplasmic reticular, “mitochondrial” pattern (without ANA) in a 52-year-old female patient with FVIII:C of 4 IU/dL and an inhibitor of 71 BU/mL. (E) Cytoplasmic discrete dots/GW body-like pattern in a 73-year-old male patient with FVIII:C of <1 IU/dL and an inhibitor of 1.5 BU/mL. (F) Combined nuclear and cytoplasmic pattern (1:160) in a 73-year-old female patient with FVIII:C of 6 IU/dL and an inhibitor of 60 BU/mL. FVIII:C, factor VIII activity.

Figure 2.

ANA titers in patients with AHA and controls. Histogramof HEp-2 immunofluorescent ANA titers of patients with AHA (n = 69) compared with matched control patients (n = 69). Only nuclear patterns were considered.

Figure 3.

HEp-2 immunofluorescent staining of control antibodies. Anti-FVIII monoclonal mAb, isotype controls or human recombinant FVIII alone do not bind to HEp-2 cells, whereas immune complexes of anti-FVIII mAb plus FVIII produce a cytoplasmic reticular staining pattern like that observed in 6 patients with AHA (bottom row, compare Figure 1D).