Congenital heart block (CHB) is a model of passively acquired autoimmunity that is strongly associated with the transplacental passage of maternal IgG against Ro/SSA and La/SSB ribonucleoproteins (1). CHB is a cardiac manifestation of the so-called neonatal lupus syndromes, a constellation of clinical manifestations found in babies born to mothers with anti-Ro/anti-La antibodies, which may include a cutaneous rash or, very rarely, blood or liver abnormalities that improve over time (2, 3).
CHB is the most serious manifestation of neonatal lupus and can cause permanent and often life-threatening damage to the fetal heart. The incidence of CHB in the offspring of mothers with the pathologic autoantibodies ranges between 1% and 2% (4, 5), but the recurrence rate in subsequent pregnancies following the birth of a child with neonatal lupus is ∼18% (6–8). It appears that the presence of maternal antibodies is a necessary, but not sufficient, condition for the development of CHB. The uterine environment, fetal factors (e.g., genes, viral infections), and maternal factors (e.g., specific autoantibody profile) (9) or the absence of an active antiidiotypic network targeting pathogenic autoantibodies (10) have been proposed as possible cofactors in the development of CHB. In a previous study, in an attempt to define the role of this network in the development of neonatal lupus, we evaluated the idiotypic–antiidiotypic (Id–anti-Id) network of antibodies targeting the dominant epitopes of La/SSB (10); we concluded that the presence of anti-Id antibodies to autoantibodies against La/SSB may protect the fetus by blocking pathogenic maternal autoantibodies.
In the present study of mothers who were enrolled in the Preventive IVIG Therapy for Congenital Heart Block (PITCH) study, we evaluated the effects of intravenous immunoglobulin (IVIG) therapy on the Id–anti-Id network of anti-La/SSB antibodies. We found that, although administration of intravenous immunoglobulin (IVIG) at doses consistent with replacement does not prevent the recurrence of CHB overall (11), it significantly alters the Id–anti-Id network of pathogenic autoantibodies, increasing the anti-Id response against them. We also showed that in the pregnancies resulting in the birth of a healthy child, the anti-La/SSB antibody Id:anti-Id ratio in maternal serum was significantly lower compared to the ratio observed in pregnancies resulting in the birth of a child with neonatal lupus.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
CHB is closely associated with the presence of anti-Ro and anti-La antibodies (2). It has been hypothesized that maternal autoantibodies trigger an inflammatory cascade that leads to irreversible fibrotic replacement of the fetal AV node (3, 14). Therefore, once (third-degree) CHB develops in the fetus, it cannot be reversed with any available treatment. The condition is fatal in ∼20% of cases. The incidence of CHB is 10-fold higher in the offspring of mothers who are positive for anti-Ro and anti-La antibodies and have had a child with neonatal lupus (8), comprising the group of high-risk pregnancies. In these cases, prophylactic therapy would be particularly beneficial.
Proposed mechanisms to explain how IVIG prevents tissue damage have included the following: 1) it increases elimination of maternal anti-Ro and anti-La; 2) it reduces transplacental transport of antibodies; and 3) it modulates the inhibitory signals on macrophages, with consequent reduction of the inflammatory response and fibrosis in the fetal heart (15–17). In the PITCH study, IVIG was given to prevent development of CHB in the fetuses of 20 high-risk pregnant women (11). In parallel, a study using an identical treatment protocol in 15 high-risk pregnant women was conducted in Europe (18). Both studies were discontinued after reaching the stopping rule, i.e., detection of 3 cases of CHB among the first patients who were enrolled (11, 18). Despite the finding that administration of IVIG at low doses does not prevent the recurrence of CHB, it was of particular interest to study how IVIG alters the Id–anti-Id network of pathogenic autoantibodies in the sera of mothers enrolled in the PITCH study.
Our previous work suggested that the presence of anti-Id antibodies to autoantibodies against La/SSB may protect the fetus by blocking the pathogenic maternal autoantibodies (10). This finding prompted us to investigate in detail the Id–anti-Id network of anti-La/SSB antibodies among mothers enrolled in the PITCH study. Sixteen of 20 mothers who completed the PITCH study were anti-La/SSB positive, and 3 gave birth to a child with neonatal lupus (2 with CHB and 1 with rash). The sera of the mothers of all 3 infants with neonatal lupus had a significantly higher Id:anti-Id antibody ratio than those of the mothers who gave birth to a healthy child, suggesting an inadequate antiidiotypic response after IVIG administration.
Therefore, unbalanced idiotypic antibodies, targeting the major epitope of La/SSB by antiidiotypic antibodies, may be an important determinant of neonatal lupus occurrence in the offspring of mothers treated with IVIG. However, it was unclear how each dose of IVIG alters the anti-Id network and influences Id and anti-Id antibody levels. To address this question, we carried out detailed analyses of Id and anti-Id reactivity in maternal sera at each time point of IVIG administration. Repeated IVIG administration led to a gradual increase of anti-Id activity and a decrease of Id activity in the majority of cases, bringing the levels of anti-Id antibodies close to the levels of Id antibodies. However, this was not the case for the mothers of the affected children with neonatal lupus; in sera from these mothers, anti-Id antibody titers did not approach those of the Id antibodies.
Inadequate anti-Id response has been recently correlated with the appearance of other autoimmune diseases, such as type 1 diabetes mellitus (DM) (19). Type 1 DM is characterized by the presence of autoantibodies to glutamate decarboxylase 65 (GAD65) (20). GAD65 antibodies often herald the onset of type 1 DM by months or years and, together with other autoantibodies to islet cells, are used to predict disease (21). Recently, Oak et al demonstrated that masked GAD65 antibodies are present in the healthy population and the absence of a particular anti-Id antibody, rather than GAD65 antibodies per se, is a characteristic of type 1 DM (19). Consistent with this concept are our results, which suggest the existence of masked Id antibodies (by anti-Id antibodies) in the mothers who gave birth to a healthy child. More specifically, we found that elimination of anti-Id antibodies from the sera of mothers treated with IVIG substantially enhanced the autoantibody (Id) activity in 5 of 7 series of sera tested (but not in the sera of 1 mother who gave birth to a child with neonatal lupus).
The increase of anti-Id activity and decrease of Id activity observed in many of the maternal sera following IVIG administration could be attributed to direct addition of anti-Id antibody to the mothers' sera. This hypothesis was confirmed by examining the Id and anti-Id antibody content of the different IVIG preparations administered to 13 of the 16 anti-La/SSB–positive mothers enrolled in the PITCH study. These experiments showed that all but 2 of the mothers had received IVIG preparations with higher anti-Id activity than Id activity. The 2 mothers in whom this was not the case each gave birth to a child with CHB. Thus, it is plausible that the reason IVIG did not prevent CHB in 2 mothers was that they received IVIG with a high Id:anti-Id activity ratio, and thus anti-Id antibodies were not effectively increased.
Previous studies have indicated that IVIG exerts reactivity against different autoantigens, as well as antiidiotypic antibodies (22–25). This reactivity is likely due to the presence of polyreactive IgG antibodies and anti-Id antibodies in the pooled plasma used for IVIG fractionation (26, 27). However, it can be augmented by various chemical treatments to which the IgG molecules have been subjected during the fractionation process (22, 28). More specifically, the production process of commercial IVIG preparations (which differs among brands) involves fractionation and virus inactivation steps that include, in some cases, treatment at extreme physical conditions. These procedures significantly affect the autoreactivity of IVIG (22). Therefore, it is not surprising that different brands and batches of commercial IVIG (as used in the PITCH study) do not have identical Id and anti-Id antibody content.
Exogenous addition, to individual women's serum, of the IVIG preparation that they had received at each study time point during pregnancy led to a large increase in anti-Id activity and a small decrease in Id activity in the majority of sera tested. In this respect, higher doses of IVIG (up to 2 mg/kg), which are presumed to be safe in pregnancy (29, 30), might be more efficient for the prevention of neonatal lupus. Moreover, given that the IVIG preparations administered to mothers who gave birth to a child with CHB had an Id:anti-Id activity ratio of ≥1, pretesting of IVIG and selecting preparations with low Id–anti-Id potential for administration to high-risk women might be more efficacious in preventing neonatal lupus and CHB.
In conclusion, our study demonstrates that IVIG preparations contain anti-Id antibodies, and their administration can enhance the antiidiotypic antibody response in pregnant women who are positive for anti-La/SSB. The success of IVIG therapy may be a function of the anti-Id potential of each IVIG preparation as well as the Id:anti-Id ratio achieved in maternal sera after administration of IVIG.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Tzioufas had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Routsias, Moutsopoulos, Buyon, Tzioufas.
Acquisition of data. Routsias, Kyriakidis, Friedman, Llanos, Clancy, Moutsopoulos, Buyon, Tzioufas.
Analysis and interpretation of data. Routsias, Kyriakidis, Moutsopoulos, Buyon, Tzioufas.