In HPA 1a-immunized women the decrease in anti-HPA 1a antibody level during pregnancy is not associated with anti-idiotypic antibodies

A large screening and intervention study, aimed at reducing morbidity and mortality associated with severe anti-HPA 1a antibody induced neonatal alloimmune thrombocytopenia (NAIT), has recently been carried out in Norway.1 Recently, it was shown that the anti-HPA 1a levels surprisingly decreased in 92 of 147 women who had been pregnant previously and who carried an HPA 1a positive fetus (P_{92 or more of 147}=0.003).2

Figure 1.The frequency distribution of patient and control samples. The anti-idiotypic reactivities (OD at 405 nm) against F(ab’)2 fragments from the two HPA 1a-immunized women (P1 and P2) and the healthy control (C) are shown.

The reason for decline in anti-HPA 1a antibodies during pregnancy in multigravida is not known. One mechanism by which the antibody repertoire can be regulated is via idiotypic networks. According to this concept an HPA 1a-immunized woman may develop anti-idiotypic antibodies (usually designated Ab₂) which are antibodies against antigenic determinants (idiotopes) on the variable region of the anti-HPA 1a antibodies. These anti-idiotypic antibodies may play an important immunoregulatory role as they can blunt the initial immune response (Ab₁).3,4 We therefore examined whether the observed decline in anti-HPA 1a antibody level in immunized pregnant women was associated with a concurrent increase in anti-idiotypic antibodies. The study was approved by the Regional Committee for Medical Research Ethics, North Norway (approval n. P-REK V 13/1995).

A total number of 829 samples of EDTA plasma were collected from 157 HPA 1a-incompatible pregnancies included in the screening and intervention study.1 As controls we used 18 samples collected during pregnancy in 4 non-HPA 1a-immunized pregnant women, and 28 samples from normal blood donors (11 males and 17 females). The labeling system for the control samples was similar to the system used for the patients’ samples. The coding (patients versus controls) was concealed until all analyses were completed.

Anti-idiotypic activity was assessed as previously described.5 Briefly, IgG was purified from 2 HPA 1a immunized women (P₁ and P₂) and from one non-immunized healthy control (C). F(ab’)₂ fragments (from P₁, P₂ and C) prepared by pepsin digestion were used as coating proteins in an enzyme-linked immunosorbent assay (ELISA) for detection of anti-idiotypic antibodies (Ab₂) in plasma from patients and controls.

On each ELISA plate four different dilutions of immunoglobulin (100, 50, 25 and 12.5 μg/mL; Gamunex, Talecris Biotherapeutics, Mississauga, ON, Canada) as well as plasma samples from 2 of 4 healthy individuals were included as controls. The results from these healthy individuals were not analyzed in a blinded fashion, and hence they were not included in the statistical analysis. All samples from individual pregnant women were analyzed on one ELISA plate.

There was no significant difference in anti-idiotypic reactivity between samples from HPA 1a-immunized women and controls. There was no significant difference in the dispersion of anti-idiotypic reactivity between the study objects and the controls and no obvious difference in the frequency distribution pattern of anti-idiotypic reactivity between study objects and controls (Figure 1), indicating that the observed reactivity was not directed against the anti-HPA 1a specific F(ab’)₂ fragments. When the analysis was restricted to those women in whom there was a decrease in anti-HPA 1a level during pregnancy, we again could not find a concurrent increase in anti-idiotypic reactivity.

Our results contrast with a previous report suggesting that anti-idiotypic networks play a pivotal role in regulation of anti-HLA antibody levels. Atlas et al. showed that 55 of 82 multitransfused HLA immunized patients with decreasing anti-HLA antibody levels over time, had concurrently increasing levels of anti-idiotypic antibodies in their sera.6 Anti-idiotypic antibodies could not be found in sera from patients with persistently high anti-HLA antibody levels.6 In addition, more than one third of the anti-idiotypic antibodies inhibited the binding of the anti-HLA antibodies to platelets, indicating that they were specific for the paratopes of the anti-HLA andibodies.6

One possible explanation for the discrepancy between our results and those reported by Atlas et al. is that alloimmunization in NAIT is caused by a point mutation where a single nucleotide substitution results in one amino acid replacement at position 33 in GPIIIa (from proline in HPA 1b to leucine in HPA 1a), whereas in HLA-alloimmunization the antigenic diversity between different HLA molecules is much larger. Consequently the antibody repertoire of anti-HLA antibodies is considerably larger than that of anti-HPA 1a antibodies and perhaps the latter antibodies (Ab1) cannot effectively generate the production of anti-idiotypic antibodies (Ab2).

In conclusion, it is unlikely that idiotypic regulation of anti-HPA 1a antibodies occurs during pregnancy in HPA 1a-immunized women.