Background: Fetal autoimmune congenital heart block is a rare but life-threatening condition that is difficult to predict. This study sought to identify a serological biomarker predictive of autoimmune congenital heart block in pregnancies at risk due to maternal systemic lupus erythematosus, Sjögren's disease, undifferentiated connective tissue disease, or a history of congenital heart block offspring.
Methods: This case-control study analysed maternal blood samples from pregnancies affected and unaffected by autoimmune congenital heart block from two centres (The Hospital for Sick Children, Canada, and The University of Padua, Italy). Serum samples collected across varying gestational ages were used for biomarker discovery, verification, and validation. The key inclusion criterion was a positive clinical test for maternal anti-Sjögren's syndrome-related antigen A/Ro autoantibodies, and the key exclusion criterion was structural congenital heart disease associated with heart block. Cases were pregnancies that resulted in fetal or neonatal congenital heart block and controls were pregnancies that resulted in offspring with a normal heart rhythm. Two-dimensional western blotting was used to identify maternal autoantibodies targeting fetal cardiac proteins, using fetal heart tissue and stem-cell-derived cardiomyocytes. Findings were validated using commercial proteins. Sensitivity and specificity for predicting fetal heart block outcomes were assessed using receiver-operating characteristic curves. The primary study outcomes were the presence and specificity of anti-cardiac autoantibodies in autoimmune congenital heart block cases versus controls, the association between specific autoantibodies and congenital heart block development, and the predictive accuracy of identified autoantibodies. This study did not involve individuals with lived experience in the study design or implementation.
Findings: Serum samples were collected between Jan 1, 2010, and Dec 31, 2020, in the discovery cohort (The Hospital for Sick Children), between Aug 1, 1999, and Dec 31, 2019 in the verification cohort (University Hospital of Padua), and between June 1, 2018, and Aug 31, 2023, in the validation cohort (The Hospital for Sick Children). Mean maternal age was 34·2 years (SD 5·2) in the discovery cohort, 33·5 years (4·6) in the verification cohort, and 32·8 years (4·3) in the validation cohort. Maternal serum samples from pregnancies affected by fetal autoimmune congenital heart block (cases; n=46) showed an expanded repertoire of anti-cardiac autoantibodies compared with unaffected pregnancies (controls; n=65). Mass spectrometry identified 11 potential cardiac protein targets for maternal autoantibodies and the presence of seven autoantibodies were confirmed in serum samples from affected pregnancies. Four targets were identified before the onset of congenital heart block (ANXA1, BIP, MYPC3, and AT1A1). Maternal autoantibodies against the Na+ and K+ ATPase α1 isoform (AT1A1) were detected as early as 7 weeks gestation and were present in all affected pregnancies of mothers with and without previous fetal heart block history (six [100%] of six in the discovery cohort, 22 [100%] of 22 in the verification cohort, and 16 [100%] of 16 in the validation cohort). Anti-AT1A1 autoantibodies were absent in all unaffected pregnancies in pregnant women with no fetal heart block history (54/65 [83%]).
Interpretation: This study identifies a novel maternal autoantibody targeting the fetal AT1A1 cardiac protein as a potential biomarker for the early and accurate detection of fetal autoimmune congenital heart block in pregnant women who had not previously had an affected pregnancy. This advancement might improve the maternal-fetal management of at-risk pregnancies through enhanced surveillance and timely and informed therapeutic interventions.
Funding: Heart and Stroke Foundation of Canada and Canadian Institutes of Health Research.
Copyright © 2024 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar technologies.