Accumulated data suggest that women with anti-SSA/Ro antibodies identified by commercial enzyme-linked immunosorbent assay (ELISA) face a 2% risk of having a child with neonatal lupus–congenital heart block (CHB), most often third-degree (1). While recent reports suggest that this risk may increase to 5% in the presence of anti-SSB/La antibodies (2), identification of more specific antibody reactivity would be a major advance in the translational approach to CHB. There is a clear need to identify an early marker of CHB, given that established third-degree block has not been reversed to date, despite treatment with maternal steroids known to be effective in the fetal circulation (3). Presently, weekly echocardiographic evaluation of all fetuses exposed to anti-SSA/Ro antibodies has been recommended to detect potentially reversible incomplete blocks (4). The discovery of an antibody specificity that induces injury to the developing human conduction system could lead to treatments directed toward removal of the putative autoantibody from the maternal circulation before placental transfer becomes effective.
Antibodies to the 52-kd SSA/Ro protein (Ro 52) are found in >80% of mothers whose children have CHB (2, 5–7). Initial epitope mapping of this response revealed an immunodominant region spanning aa169-291 (which contains the leucine zipper) that was recognized by the majority of the CHB sera, frequently in the context of HLA–DRB1*0301, DQA1*0501, and DQB1*0201 (5). The finer specificity of the anti–Ro 52 response has been confirmed and extended, with the current focus on aa200-239 (p200) (7). In a limited study of 9 mothers of children with CHB and 26 anti-SSA/Ro–positive mothers of healthy children, antibodies to p200 predicted CHB with greater certainty than currently available testing for either 60-kd or 52-kd SSA/Ro (7). Recent studies integrating an in vivo rodent model and an in vitro culturing system suggest that anti-p200 antibodies bind neonatal rodent cardiocytes and alter calcium homeostasis (8).
To address both the clinical necessity and the sufficiency of this newly identified p200 reactivity in the development of CHB, as well as the reduced risk of CHB reportedly associated with aa176-196 (p176) and aa197-232 (p197) (7), maternal sera from the Research Registry for Neonatal Lupus (RRNL) (9) and the prospective, multicenter study, PR Interval and Dexamethasone Evaluation (PRIDE) in CHB (4) were evaluated. In addition, the PRIDE study provided the opportunity to address whether the level of anti-p200 antibodies correlated positively with the length of the Doppler mechanical PR interval.
- Top of page
- PATIENTS AND METHODS
The identification of a fetus at risk for developing CHB would be invaluable, given the substantial morbidity (∼65% require lifelong pacing) and mortality (∼20%) associated with this disease (5, 9, 18–20). Moreover, current echocardiographic technologies are being evaluated for early in utero detection at a time when injury might still be reversible (12, 13, 21–23). Among putative factors likely to reveal the fetus at highest risk is a maternal autoantibody specificity more precise than reactivity with full-length components of the SSA/Ro–SSB/La system that would segregate with affected but not unaffected pregnancies.
In this study, neither the mean level of p200 reactivity nor frequency was significantly different among anti–Ro 52–positive mothers who had children with CHB, mothers who had children with isolated neonatal lupus rash but not CHB, or mothers who had only healthy children with no neonatal lupus. Approximately 25% of the sera from the mothers of children with CHB did not display reactivity to p200. This contrasts with findings of a more limited study in which 9 of 9 anti–Ro 52–positive mothers of children with CHB demonstrated reactivity with p200, compared with lower titer and frequency in 26 mothers (not all anti–Ro 52–positive) of healthy children (7). Since our data show that antibody reactivity against p200 is not observed in the absence of reactivity to full-length Ro 52, evaluation of reactivity to p200 should be restricted to individuals positive for anti–Ro 52 to ascertain whether this finer specificity truly exaggerates the risk of CHB in a fetus. It is readily acknowledged that many of the sera tested were not obtained at the time of a given pregnancy. However, if titers of anti-p200 antibodies parallel levels of reactivity with full-length Ro 52, then steady-state levels of anti-p200 would be expected, since anti–Ro 52 antibody levels only rarely fluctuate over time (24). This would further validate the findings reported herein. In fact, many of the sera assigned to the group of mothers of children with CHB were actually obtained during the time of a healthy pregnancy, which suggests that pathogenicity of anti-p200 antibodies requires additional factors.
One difficulty in identifying a pathogenic effect of an autoantibody is accounting for the heterogeneity of that effect. CHB is a paradigmatic example in that not only is the injury seemingly rare, but the degree of injury varies along a spectrum from clinically inconsequential first-degree block through third-degree (complete) block and even, in some cases, an associated cardiomyopathy that is often fatal (25, 26). Identification of a necessary or essential factor is only part of the challenge in defining the pathology of CHB, since recurrence rates from one pregnancy to the next are 18%, not 100%, and identical twins are with rare exception discordant for disease (27).
Even if anti-p200 or an even finer epitope specificity were truly pathogenic and cross-reactive with a structure on the cardiocyte surface, subsequent events are required to convert risk of disease to full expression. Many healthy fetuses were exposed to maternal anti-p200 antibodies, as demonstrated by results observed in sera obtained during pregnancies followed prospectively in the PRIDE study. Salomonsson and colleagues (8) reported that 19% of rat pups born to mothers actively immunized with p200 had first-degree block. These same investigators concluded that an epitope within the predicted leucine zipper structure is recognized by similar antibodies from children with CHB and from rat pups with first-degree block, and by selected human anti-p200 monoclonal antibodies, although there was not complete uniformity among the groups when a series of peptides was evaluated (8). While the antibody testing herein did not evaluate the finer specificity of reactivity within the α-helical p200 region, the fact that the majority of anti–p200-exposed rat pups had normal AV conduction suggests that finer mapping is not likely to change the conclusions presented herein. Thus, data both from rodents and from humans support the notion of a partnership between maternal and fetal contributions in the pathogenesis of CHB. Other candidate factors include fetal genes relating to the pathologic cascade, which might amplify inflammatory and fibrosing responses (28, 29), and environmental influences.
This study did not confirm results recently reported by Sonesson and colleagues (21), in which prolongation of the AV time interval and levels of anti-p200 antibodies were positively correlated in 8 of 24 anti–Ro 52–positive mothers (21). The fetal Doppler mechanical PR interval measurement is a putative surrogate marker for the earliest sign of CHB, i.e., first-degree AV block. It was initially reported (12) as a tool for surveillance of mothers at risk for the development of fetal CHB and has been validated in a multicenter study of a normal population (13) with proof of concept in an affected fetus (30). Differences between the findings in the present study and those of Sonesson and coworkers (21) may have arisen from ethnic differences between the 2 cohorts, differing gestational age range and a smaller number of anti–Ro 52–positive patients in the Swedish study (21), and interindividual variation of measurement, particularly in the onset of the mitral A wave. However, it should be acknowledged that, at least postnatally, PR intervals are highly variable and change over time, with alterations in heart rate and sympathetic and parasympathetic tone. For example, the length of the PR interval does not even correlate with level of digoxin, or with the degree of carditis in acute rheumatic fever (31).
In conclusion, fetuses can develop CHB in the absence of detectable maternal anti-p200 antibodies, and exposure to antibodies of this fine specificity does not invariably result in CHB. Accordingly, this reactivity is neither necessary nor sufficient to account for all cases of isolated CHB detected in utero. Injury to the fetal conduction system and surrounding myocardium is rare and likely the consequence of several factors, both maternal and fetal. However, continued search for an essential antibody specificity, perhaps related to an epitope within the region of p200, remains important. Removal of this putative single maternal factor could prevent initiation of injury/inflammation and the subsequent scarring amplified by the fetus as well as the in utero environment.