In this study, we describe a flow cytometry (FC) system for detecting antibodies to factor VIII (FVIII) and compare its results with those of enzyme-linked immunosorbent assay (ELISA) that detects both inhibitory (I-Ab) and non-inhibitory (NI-Ab) antibodies and the Nijmegen modification of the Bethesda method, detecting I-Ab. FC was set up in our laboratory. Recombinant FVIII (rFVIII) was coupled to microspheres (FVIII-m) and reacted with different plasma dilutions. Microspheres without rFVIII were used as control (control-m). Captured anti-FVIII antibodies were detected using anti-human IgG. Plasma samples from the following patients with severe haemophilia A (SHA) patients were evaluated: 17 P (patients without I-Ab, <0.5 BU mL(-1)); 13 PI (patients with I-Ab, 1.1-8200 BU mL(-1)). Of these 13, two PI were referred during immune tolerance induction (ITI), and plasmas from 12 healthy donors (HD) were evaluated. Semiquantitative results were given as an index (the highest mean fluorescence intensity ratio between FVIII-m and control-m multiplied by the inverse of the corresponding plasma dilution). Both plasma and serum were suitable for the test. FC agreed with the Bethesda method (r = 0.8; P = 0.0001). FC and ELISA had 80% of coincidence. Four of 17 patients (23.5%) had NI-Ab by FC, and two of them developed high levels of I-Ab later on. This test provides a useful alternative for measuring FVIII antibodies supplementing Bethesda assay. FC is fast and easy to perform. No more than 200 μL of plasma or serum is required especially making it useful for paediatric patients.

The development of inhibitors is a complication of replacement treatment in Haemophilia. Loss of factor VIII-specific memory B cells in the spleen is associated with down regulation of antibodies in mice treated with high doses of FVIII, but changes in B cell memory have not been described in haemophilic patients. The aim of this study was to evaluate the phenotype of circulating lymphocytes in severe haemophilia A. Twenty patients with inhibitors (PI), 22 without inhibitors (P), nine patients during immune tolerance induction (ITI) treatment and 20 healthy donors (HD) were included. Peripheral blood lymphocytes were examined using flow cytometry. Anti-FVIII antibodies were measured using Bethesda and flow cytometry. Percentages of T subsets and B lymphocytes were similar in all groups. In contrast, memory B cells (CD27+) were decreased in PI and P compared with HD, but the level of significance was higher in PI (P = 0.001) than P (P = 0.01). PI with high level of anti-FVIII antibodies presented the lowest B memory values. CD70 expression was also lowest in PI. Non-switched CD27+ subpopulation (IgD+) was prevalent in PI, but did not show statistical significance. When ITI failed, the percentages of CD27+ B cells after 12 months of ITI were lowest. In a longitudinal study performed in four patients, an increased percentage of CD27+ and CD70+ B cells during ITI was found. This work suggests that different peripheral lymphocyte markers, such as CD27 and CD70 on B cells, may be helpful to evaluate anti-FVIII response and to monitor the success of ITI.

We suggest the use of the FC method to supplement functional traditional assays and to help to improve the management of patients with SHA.