Neurodevelopmental Disorders

Epidemiologic data suggest that children born to women with SLE may have an increased risk of neurodevelopmental disorders compared with children born to healthy women. Several retrospective studies suggest that children, particularly sons, of mothers with SLE are at increased risk (up to 25%–45%) for learning disabilities. In a small retrospective study using parental report, the prevalence of learning problems in offspring of mothers with SLE was more than twice that reported for controls. A prospective study assessed the neurodevelopment of 57 children born to mothers with SLE and 49 controls using standardized tests. Offspring of mothers with SLE had more than a threefold increase in anomalies related to learning and memory, as well as behavior. In a retrospective cohort study of 60 SLE offspring, in utero exposure to azathioprine conferred more than a sixfold increased risk of having special educational needs (used as a proxy for developmental delays), when adjusting for disease severity and obstetric complications.

Although these previous studies support the hypothesis of an increased risk of neurodevelopmental disorders in offspring of mothers with SLE, the studies were marked by important methodological limitations: all had limited sample size; only one controlled for obstetric complications and medication exposures; and most used parental report, did not include a control group, and/or were retrospective in nature.

In 2015, investigators reported data from the Offspring of SLE Mothers Registry (OSLER), a large population-based cohort using Quebec’s health care databases and including 719 children born to mothers with SLE, and a matched control group of 8493 children born to unaffected mothers. SLE offspring were more frequently found to have a diagnosis of autism spectrum disorders (ASD) compared with unexposed children (frequency of recorded ASDs 1.4% [95% confidence interval (95% CI) 0.8–2.5] vs 0.6% [95% CI 0.5–0.8]), a difference of 0.8% (95% CI 0.1–1.9). The mean age at ASD diagnosis was younger in offspring of mothers with SLE (mean 3.8 years, 95% CI 1.8–5.8) compared with offspring of controls (mean 5.7 years, 95% CI 4.9–6.5). In multivariate analysis accounting for maternal characteristics and obstetric complications, SLE offspring had a substantially increased risk of ASD compared with controls (odds ratio [OR] 2.19, 95% CI 1.09–4.39). The younger age at ASD diagnosis could suggest either more severe cases or increased surveillance within the SLE population.

In addition to cohort evidence of an increased risk of neurodevelopmental disorders in offspring of mothers with SLE, numerous case-control studies have suggested an increased prevalence of SLE and other autoimmune diseases in mothers of children affected with neurodevelopmental disorders. In a case-control study of 61 children with ASD and 46 healthy controls, affected children had more than an eightfold increase in the odds of having a mother with an autoimmune disorder (by self-report) than unaffected children. SLE was observed in 13% of children with ASD, versus 4% of healthy controls. Another large population-based study showed similar results. In this study, children with ASD were more likely than unaffected children to have a mother diagnosed with an autoimmune rheumatic disease (relative risk 1.56, 95% CI 1.08, 2.17), whereas the likelihood of having a father with these diseases did not differ. This suggests that the association between SLE and neurodevelopmental disorders might be influenced by prenatal exposure to maternal antibodies and/or fetal environment during gestation.

Limited evidence suggests a potentially increased risk of neurodevelopmental disorders in RA offspring, as shown by a recent systematic literature review that identified only 2 observational studies assessing this issue. In one study, 46% of ASD offspring had a first-degree relative with RA compared with 26% of controls ( P = .04). Moreover, in a population-based study including 3325 children with ASD and more than 650,000 controls, investigators observed an increased risk of ASD in children with a maternal RA compared with children born to unaffected mothers (OR 1.70, 95% CI 1.07–2.54). These observational studies are marked by important methodological limitations; study effect estimates were relatively imprecise, it is unclear to what extent the subjects represented the population base, neither controlled for medication exposures, and only one accounted for obstetric complications and considered the timing of RA diagnosis in relation to the pregnancy.

In utero exposure to maternal immunoglobulin G (IgG) antibodies is increasingly recognized as an important environmental risk factor for neurodevelopmental disorders. Maternal IgG antibodies begin to cross the placenta during the second trimester of pregnancy, reaching circulating levels in the newborn that exceed maternal levels due to active transport across the placenta. In the presence of maternal autoimmunity, autoantibodies also cross the placenta and can interfere with fetal development. Although offending maternal autoantibodies are cleared from the child’s circulation within the first 6 months of life, autoantibody-mediated injury in utero can result in long-term damage to organs (eg, congenital heart block in neonatal lupus).

The blood-brain barrier blocks IgG entry into the adult central nervous system (CNS), but in the fetus, the immature blood-brain barrier allows IgG access to the developing brain. Genetic predisposition may increase susceptibility to neurodevelopmental disorders in children exposed in utero to offending maternal IgG. Antibodies directed against fetal brain proteins (yet to be identified) have been observed in 10% to 12% of mothers of children with ASD. This antibody reactivity has been shown to be absent in mothers of normally developing children. When human maternal fetal brain-reactive antibodies from mothers of children with ASD were administered to pregnant mice, behavioral alterations in the offspring were noted. In this mouse model, an increased number of microglial cells were observed in the brain of exposed offspring, suggesting that these brain-reactive antibodies may mediate their effects through inflammatory changes.

Diamond and colleagues recently showed that 53% of mothers of a child with ASD with fetal brain-reactive antibodies also exhibited anti-nuclear autoantibodies compared with 13% of mothers with ASD without fetal brain-reactive antibodies and 15% of control women. They also observed an increased prevalence of autoimmune diseases in mothers with ASD with fetal brain-reactive antibodies. Mothers with ASD with fetal brain-reactive antibodies were 3 times more likely to have SLE and RA compared with mothers of a child with ASD without these antibodies and control women of childbearing age, suggesting that a subset of ASD may be related to in utero maternal antibody exposure.

New experimental data further support a potential link between in utero exposure to SLE and neurodevelopmental disorders. A subset of anti-double-stranded DNA antibodies, anti–N-methyl-D-aspartate receptor (NMDAR) antibodies, are present in up to 60% of women with SLE. In a murine model, these antibodies have been shown to cross the placenta, induce fetal brain neuronal apoptosis, and cause cognitive impairments in offspring, preferentially in male individuals. Affected offspring displayed smaller-sized neocortical neurons and neuronal migration defects, findings observed in histologic studies of humans affected with learning disabilities.

It is noteworthy that pregnant mice exposed to anti-NMDAR antibodies had a marked preferential loss of female fetuses, resulting in an increased male-to-female ratio in their offspring. Interestingly, investigators have recently demonstrated that mothers with SLE had substantially increased odds of having male offspring compared with mothers without SLE (OR 1.18, 95% CI 1.01–1.38). This finding mirrors experimental data and parallels the male predominance seen in neurodevelopmental disorders.

Other autoantibodies found in SLE may also potentially alter fetal brain development. Antiphospholipid antibodies (aPL), present in 30% of women with SLE (and 15% of subjects with RA), have been found at high levels in the serum of exposed neonates. These antibodies can bind CNS cells and, in murine models, prolonged exposure to aPL induces hyperactive behavior and neurologic dysfunction. In theory, aPL might be implicated in inducing neurodevelopmental disorders in children born to women with SLE.

Maternal cytokines may reach the fetal circulation, and the maternal cytokine milieu might constitute an another important environmental risk factor. Interleukin-6 (IL-6) is known for its primordial role in brain development : administration in pregnant mice caused substantial behavioral and social deficits in the offspring, whereas coadministration with an anti–IL-6 antibody prevented these deficits. IL-6 is involved in autoantibody production in RA and SLE, and affected patients have markedly elevated IL-6 blood levels. IL-6 could have a direct effect on the fetal brain or enhance the production of maternal fetal brain-reactive antibodies, which could cross-react with the fetal brain, leading to neurodevelopmental disorders.

Recently, maternal IL-17a has been identified as a potentially critical cytokine in the development of ASD in offspring. Rodents subjected to maternal immune activation (MIA) develop ASD phenotypes. In a mouse model, Choi and colleagues showed that T-helper 17 (TH17) cells and the effector cytokine IL-17a are required in mothers for MIA-induced behavioral abnormalities in offspring. MIA induced abnormal cortical development in the fetal brain, which was dependent on maternal IL-17a; treatment with anti–IL-17a antibodies in the pregnant mothers improved both behavioral abnormalities and abnormal cortical development in offspring.

Genes implicated in autoimmune disorders, including SLE and RA are significantly more prevalent in subjects with ASD. One such gene is the C4B null allele, strongly associated with SLE. Of particular interest, the C4B null allele is 4 times more common in individuals with ASD compared with controls. As presence of the C4B null allele leads to partial C4B deficiency, and because the complement system is involved in brain tissue remodeling and repair, alterations in C4B levels might alter the fetal immune response to in utero immunologic insults, resulting in pathologic changes.

Subjects with both RA and ASD share a common genetic predisposition to HLA-DRB1*04 alleles. The association with the HLA-DRB1*04 alleles represents the firmest link between a genetic susceptibility factor and RA, conferring up to an 11-fold increase in RA risk. Previous case-control studies have consistently shown increased frequency of HLA-DRB1*04 in ASD offspring and their mothers, but not their fathers. In a recent study assessing transmission disequilibrium of the HLA-DRB1*04 alleles in 31 families of ASD offspring, investigators observed significant transmission disequilibrium for HLADRB1*04 (OR 4.67, 95% CI 1.34–16.24) from maternal grandparents to mothers of ASD cases, whereas they did not observe HLA-DRB1*04 transmission from mothers or fathers to offspring with ASD. These findings support a role for HLA-DRB1*04 as an ASD risk factor acting in mothers during pregnancy (ie, not due to genetic transmission from parents, but potentially through maternal action in utero), raising the possibility of a maternal immune component to ASD pathogenesis.

Finally, SLE and RA pregnancies are at increased risk of adverse obstetric outcomes, such as prematurity and SGA, which are potential risk factors for neurodevelopmental disorders. Observational studies report a 1.5 to 3.0-fold increase in neurodevelopmental disorders in children born preterm or SGA versus controls. Thus, obstetric complications in women with SLE and RA may also increase neurodevelopmental disorders in offspring.

Congenital Heart Defects

CHDs are the most frequent type of birth defects, accounting for approximately a third of all congenital anomalies ; they are associated with substantial childhood morbidity. In utero exposures, such as maternal illness and medications, are thought to play an important role in the yet to be fully elucidated etiology of CHD. In particular, a recent study suggests a threefold increased risk of CHD in children born to mothers with various systemic autoimmune rheumatic diseases ; however, the investigators did not specifically assess the SLE and RA effect estimates for risk and did not control for medication exposures.

Until recently, very few uncontrolled observational studies had assessed CHD in offspring of mothers with SLE. Notably, in a study of fetal echocardiography in a small number of SLE pregnancies, 7.5% of fetuses had a CHD, which is more than fivefold greater than that observed among live births from the general population (0.6%–1.3%). After excluding cases with CHD that could have caused congenital heart block, investigators observed CHD in 16% to 42% of children with congenital heart block born to mothers with anti-Ro/SSA antibodies. The most frequently observed CHDs were atrial septal defects, ventricular septal defects (VSD), and valve anomalies.

In 2015, Vinet and colleagues assessed the risk of CHD in SLE offspring within the large population-based OSLER cohort. In comparison with unexposed children, SLE offspring experienced more CHD (5.2% [95% CI 3.7–7.1] vs 1.9% [95% CI 1.6–2.2], difference 3.3% [95% CI 1.9–5.2]). In multivariable analyses, children born to women with SLE had a substantially increased risk of CHD (OR 2.62, 95% CI 1.77–3.88) compared with unexposed children. Subgroup analyses accounting for medication exposures were similar. In addition, offspring of mothers with SLE had a substantially increased risk of having a CHD repair procedure (OR 5.82, 95% CI 1.77–19.09). These data suggest that SLE offspring are at increased risk of CHD, and at least part of this risk might be independent of in utero medication exposures.

Data are more limited regarding the risk of CHD in RA offspring. In a large cohort study using Quebec’s administrative data (n = 8810), investigators observed a substantially increased risk of CHD in children born to women diagnosed with RA during childhood compared with children from the general population (OR 2.09, 95% CI 1.23–3.55). A CHD diagnosis was documented in 1.2% of RA offspring (95% CI 0.69–1.71) as opposed to 0.6% of control children (95% CI 0.42–0.78). Although analyses were limited by the lack of drug information, these findings suggest that RA offspring might also be at increased risk of CHD.

Maternal SLE/RA-related mechanisms that could be implicated in the physiopathology of CHD in offspring include autoantibody-mediated damage and cytokine imbalance. Anti-Ro/SSA and anti-La/SSB antibodies, found in, respectively, 40% and 20% of women with SLE and RA, cross the placenta and are associated with development of neonatal lupus, with congenital heart block being the most characteristic cardiac manifestation. Investigators have demonstrated that maternal anti-Ro/SSA and anti-La/SSB antibodies bind apoptotic fetal cardiocytes, resulting in the release of proinflammatory and profibrosing cytokines, and, ultimately, scarring. This process likely extends beyond the conduction tissue, involving the myocardium, endocardium, and valves. In a recent retrospective analysis of autopsies from 18 cardiac neonatal lupus cases, cardiac histologic damage outside of the conduction system was frequently observed. Six (40%) of 15 of deaths due to congenital heart block had pathology findings, such as fibrosis and calcification of the valves and/or valve apparatus.

Cardiac septation occurs early in embryogenesis and is complete by 6 weeks’ gestation. Because transplacental passage of maternal autoantibodies occurs later (beginning at approximately 16 weeks), it is unlikely that maternal autoantibodies directly interfere. However, muscular VSDs, which account for 75% of all VSDs, are thought to arise from foci of cellular death during active cardiac remodeling within an already formed ventricular septum. In addition, maternal autoantibodies might prevent closure of cardiac septal defects that might have closed naturally, possibly explaining the excess risk of cardiac septal defects in offspring of mothers with SLE.

aPL antibodies also cross the placenta: in one recent study, 40% of neonates born to women with antiphospholipid syndrome had positive aPL in cord blood. In aPL-positive adult patients with and without SLE, aPLs are strongly associated with valvular disease, and valvular deposits of aPL are thought to play an important pathogenic role. Although prior studies have reported perinatal thrombotic events occurring in children born to aPL-positive mothers, there are no data on the prevalence of congenital valve anomalies or other types of CHD in these children ; in theory, however, transplacental aPL could play a role in valve anomalies in exposed fetuses.

Cytokines, such as transforming growth factor beta (TGF-beta), play an important role in cardiac embryogenesis. In particular, adequate endocardial cushion formation, which is a critical step in cardiac septation, requires appropriate expression of TGF-beta. Defective levels of TGF-beta have been associated with CHD in animal models, whereas high levels have been linked to CHD in human studies. Serum levels of TGF-beta-1 are substantially lower in subjects with SLE (inversely correlating with disease activity), whereas they are higher in subjects with RA (correlating with disease activity). Because transplacental transfer of circulating TGF-beta can occur from mother to fetus, abnormal levels of maternal TGF-beta might alter normal fetal heart development, potentially leading to an increased risk of CHD.

Hematological Malignancies

Hematologic malignancies, such as leukemia and lymphoma, account for approximately 40% of new cancer diagnoses in children, with an incidence of up to 12 cases per 100,000 annually. Diffuse large B-cell lymphoma (DLBCL), a non-Hodgkin lymphoma, is one of the most common lymphomas among children and adolescents, representing 20% and 40% of new lymphoma cases in children and adolescents, respectively.

Patients with SLE and RA have an increased risk of hematological malignancies, particularly non-Hodgkin lymphoma, compared with the general population. Large population-based studies have consistently shown more than a twofold increase in the risk of leukemia and lymphoma in SLE and RA, thought to be due to chronic immune stimulation. The most common type of hematological malignancy in SLE and RA is DLBCL.

Data on hematological malignancies in RA offspring are very limited, and such data are nonexistent in SLE. In one cohort study, mothers with RA had an increased risk of having a child with lymphoma or leukemia compared with general population rates (relative risk 1.7, 95% CI 0.9–2.8). In another cohort study of offspring born to RA parents, there was a substantially increased risk of Hodgkin’s lymphoma (standardized incidence ratio 3.2, 95% CI 1.0–7.4) ; however, the investigators did not provide a specific estimate for offspring born to mothers with RA, as opposed to fathers with RA. In addition, in both studies, no attention was paid to the timing of RA diagnosis in relation to the pregnancy.

As childhood hematological cancers are thought to arise from an aberrant immune response, genetic studies have investigated the potential role of genes involved in the immune system. Notably, several investigators have reported an association between HLA-DRB1*04 alleles and a twofold increase in the risk of acute lymphoblastic leukemia in children. Thus, offspring of subjects with RA might be at increased risk of hematological malignancies through inheritance of risk alleles.

Rheumatic and Nonrheumatic Autoimmune Diseases

Several studies have suggested a familial aggregation of SLE and RA, as well as several autoimmune diseases, such as type 1 diabetes. However, only 2 studies have specifically assessed the risk of autoimmune diseases in offspring born to mothers with SLE or RA. Preliminary findings from the OSLER cohort showed a potentially twofold increase in the risk of nonrheumatic autoimmune diseases (including type 1 diabetes and inflammatory bowel diseases) in children born to women with SLE compared with children from the general population. However, the effect estimate for the risk of rheumatic autoimmune diseases was inconclusive.

In a large Danish population-based study of 13,566 RA offspring, Rom and colleagues studied children whose mothers had RA and compared them with children whose mothers did not have RA. In RA offspring, there was higher morbidity for 8/11 International Classification of Diseases groups. Similar results were observed in 6330 children whose fathers had RA. The investigators reported a substantial increase in the risk of specific rheumatic and nonrheumatic autoimmune diseases with up to a threefold increase in the risk of juvenile idiopathic arthritis (hazard ratio [HR] for maternal RA 3.30 [95% CI 2.71–4.03] and for paternal RA 2.97 [95% CI 2.20–4.01]), an increased risk of up to 40% of type 1 diabetes (HR for maternal RA 1.37 [95% CI 1.12–1.66] and for paternal RA 1.44 [95% CI 1.09–1.90]), and up to a 30% increased risk of asthma (HR for maternal RA 1.28 [95% CI 1.20–1.36] and for paternal RA 1.15 [95% CI 1.04–1.26]). As an increased risk of autoimmune diseases was found both in children exposed to maternal RA and children exposed to paternal RA, genetic factors are likely to play an important role, although this was not specifically studied by the investigators.

Interestingly, in the previously described study, increased risk of infectious diseases as well as mental and behavioral disorders were seen only in offspring exposed to maternal RA (the investigators did not provide a specific risk estimate for neurodevelopmental disorders). In addition, stronger associations were observed in children born to mothers with RA compared with children born to fathers with RA. Thus, fetal programming, through in utero exposures to maternal autoantibodies, cytokines, drugs, and obstetric complications, might all have an etiologic role in certain adverse health outcomes in offspring.

Neurodevelopmental Disorders

Epidemiologic data suggest that children born to women with SLE may have an increased risk of neurodevelopmental disorders compared with children born to healthy women. Several retrospective studies suggest that children, particularly sons, of mothers with SLE are at increased risk (up to 25%–45%) for learning disabilities. In a small retrospective study using parental report, the prevalence of learning problems in offspring of mothers with SLE was more than twice that reported for controls. A prospective study assessed the neurodevelopment of 57 children born to mothers with SLE and 49 controls using standardized tests. Offspring of mothers with SLE had more than a threefold increase in anomalies related to learning and memory, as well as behavior. In a retrospective cohort study of 60 SLE offspring, in utero exposure to azathioprine conferred more than a sixfold increased risk of having special educational needs (used as a proxy for developmental delays), when adjusting for disease severity and obstetric complications.

Although these previous studies support the hypothesis of an increased risk of neurodevelopmental disorders in offspring of mothers with SLE, the studies were marked by important methodological limitations: all had limited sample size; only one controlled for obstetric complications and medication exposures; and most used parental report, did not include a control group, and/or were retrospective in nature.

In 2015, investigators reported data from the Offspring of SLE Mothers Registry (OSLER), a large population-based cohort using Quebec’s health care databases and including 719 children born to mothers with SLE, and a matched control group of 8493 children born to unaffected mothers. SLE offspring were more frequently found to have a diagnosis of autism spectrum disorders (ASD) compared with unexposed children (frequency of recorded ASDs 1.4% [95% confidence interval (95% CI) 0.8–2.5] vs 0.6% [95% CI 0.5–0.8]), a difference of 0.8% (95% CI 0.1–1.9). The mean age at ASD diagnosis was younger in offspring of mothers with SLE (mean 3.8 years, 95% CI 1.8–5.8) compared with offspring of controls (mean 5.7 years, 95% CI 4.9–6.5). In multivariate analysis accounting for maternal characteristics and obstetric complications, SLE offspring had a substantially increased risk of ASD compared with controls (odds ratio [OR] 2.19, 95% CI 1.09–4.39). The younger age at ASD diagnosis could suggest either more severe cases or increased surveillance within the SLE population.

In addition to cohort evidence of an increased risk of neurodevelopmental disorders in offspring of mothers with SLE, numerous case-control studies have suggested an increased prevalence of SLE and other autoimmune diseases in mothers of children affected with neurodevelopmental disorders. In a case-control study of 61 children with ASD and 46 healthy controls, affected children had more than an eightfold increase in the odds of having a mother with an autoimmune disorder (by self-report) than unaffected children. SLE was observed in 13% of children with ASD, versus 4% of healthy controls. Another large population-based study showed similar results. In this study, children with ASD were more likely than unaffected children to have a mother diagnosed with an autoimmune rheumatic disease (relative risk 1.56, 95% CI 1.08, 2.17), whereas the likelihood of having a father with these diseases did not differ. This suggests that the association between SLE and neurodevelopmental disorders might be influenced by prenatal exposure to maternal antibodies and/or fetal environment during gestation.

Limited evidence suggests a potentially increased risk of neurodevelopmental disorders in RA offspring, as shown by a recent systematic literature review that identified only 2 observational studies assessing this issue. In one study, 46% of ASD offspring had a first-degree relative with RA compared with 26% of controls ( P = .04). Moreover, in a population-based study including 3325 children with ASD and more than 650,000 controls, investigators observed an increased risk of ASD in children with a maternal RA compared with children born to unaffected mothers (OR 1.70, 95% CI 1.07–2.54). These observational studies are marked by important methodological limitations; study effect estimates were relatively imprecise, it is unclear to what extent the subjects represented the population base, neither controlled for medication exposures, and only one accounted for obstetric complications and considered the timing of RA diagnosis in relation to the pregnancy.

In utero exposure to maternal immunoglobulin G (IgG) antibodies is increasingly recognized as an important environmental risk factor for neurodevelopmental disorders. Maternal IgG antibodies begin to cross the placenta during the second trimester of pregnancy, reaching circulating levels in the newborn that exceed maternal levels due to active transport across the placenta. In the presence of maternal autoimmunity, autoantibodies also cross the placenta and can interfere with fetal development. Although offending maternal autoantibodies are cleared from the child’s circulation within the first 6 months of life, autoantibody-mediated injury in utero can result in long-term damage to organs (eg, congenital heart block in neonatal lupus).

The blood-brain barrier blocks IgG entry into the adult central nervous system (CNS), but in the fetus, the immature blood-brain barrier allows IgG access to the developing brain. Genetic predisposition may increase susceptibility to neurodevelopmental disorders in children exposed in utero to offending maternal IgG. Antibodies directed against fetal brain proteins (yet to be identified) have been observed in 10% to 12% of mothers of children with ASD. This antibody reactivity has been shown to be absent in mothers of normally developing children. When human maternal fetal brain-reactive antibodies from mothers of children with ASD were administered to pregnant mice, behavioral alterations in the offspring were noted. In this mouse model, an increased number of microglial cells were observed in the brain of exposed offspring, suggesting that these brain-reactive antibodies may mediate their effects through inflammatory changes.

Diamond and colleagues recently showed that 53% of mothers of a child with ASD with fetal brain-reactive antibodies also exhibited anti-nuclear autoantibodies compared with 13% of mothers with ASD without fetal brain-reactive antibodies and 15% of control women. They also observed an increased prevalence of autoimmune diseases in mothers with ASD with fetal brain-reactive antibodies. Mothers with ASD with fetal brain-reactive antibodies were 3 times more likely to have SLE and RA compared with mothers of a child with ASD without these antibodies and control women of childbearing age, suggesting that a subset of ASD may be related to in utero maternal antibody exposure.

New experimental data further support a potential link between in utero exposure to SLE and neurodevelopmental disorders. A subset of anti-double-stranded DNA antibodies, anti–N-methyl-D-aspartate receptor (NMDAR) antibodies, are present in up to 60% of women with SLE. In a murine model, these antibodies have been shown to cross the placenta, induce fetal brain neuronal apoptosis, and cause cognitive impairments in offspring, preferentially in male individuals. Affected offspring displayed smaller-sized neocortical neurons and neuronal migration defects, findings observed in histologic studies of humans affected with learning disabilities.

It is noteworthy that pregnant mice exposed to anti-NMDAR antibodies had a marked preferential loss of female fetuses, resulting in an increased male-to-female ratio in their offspring. Interestingly, investigators have recently demonstrated that mothers with SLE had substantially increased odds of having male offspring compared with mothers without SLE (OR 1.18, 95% CI 1.01–1.38). This finding mirrors experimental data and parallels the male predominance seen in neurodevelopmental disorders.

Other autoantibodies found in SLE may also potentially alter fetal brain development. Antiphospholipid antibodies (aPL), present in 30% of women with SLE (and 15% of subjects with RA), have been found at high levels in the serum of exposed neonates. These antibodies can bind CNS cells and, in murine models, prolonged exposure to aPL induces hyperactive behavior and neurologic dysfunction. In theory, aPL might be implicated in inducing neurodevelopmental disorders in children born to women with SLE.

Maternal cytokines may reach the fetal circulation, and the maternal cytokine milieu might constitute an another important environmental risk factor. Interleukin-6 (IL-6) is known for its primordial role in brain development : administration in pregnant mice caused substantial behavioral and social deficits in the offspring, whereas coadministration with an anti–IL-6 antibody prevented these deficits. IL-6 is involved in autoantibody production in RA and SLE, and affected patients have markedly elevated IL-6 blood levels. IL-6 could have a direct effect on the fetal brain or enhance the production of maternal fetal brain-reactive antibodies, which could cross-react with the fetal brain, leading to neurodevelopmental disorders.

Recently, maternal IL-17a has been identified as a potentially critical cytokine in the development of ASD in offspring. Rodents subjected to maternal immune activation (MIA) develop ASD phenotypes. In a mouse model, Choi and colleagues showed that T-helper 17 (TH17) cells and the effector cytokine IL-17a are required in mothers for MIA-induced behavioral abnormalities in offspring. MIA induced abnormal cortical development in the fetal brain, which was dependent on maternal IL-17a; treatment with anti–IL-17a antibodies in the pregnant mothers improved both behavioral abnormalities and abnormal cortical development in offspring.

Genes implicated in autoimmune disorders, including SLE and RA are significantly more prevalent in subjects with ASD. One such gene is the C4B null allele, strongly associated with SLE. Of particular interest, the C4B null allele is 4 times more common in individuals with ASD compared with controls. As presence of the C4B null allele leads to partial C4B deficiency, and because the complement system is involved in brain tissue remodeling and repair, alterations in C4B levels might alter the fetal immune response to in utero immunologic insults, resulting in pathologic changes.

Subjects with both RA and ASD share a common genetic predisposition to HLA-DRB1*04 alleles. The association with the HLA-DRB1*04 alleles represents the firmest link between a genetic susceptibility factor and RA, conferring up to an 11-fold increase in RA risk. Previous case-control studies have consistently shown increased frequency of HLA-DRB1*04 in ASD offspring and their mothers, but not their fathers. In a recent study assessing transmission disequilibrium of the HLA-DRB1*04 alleles in 31 families of ASD offspring, investigators observed significant transmission disequilibrium for HLADRB1*04 (OR 4.67, 95% CI 1.34–16.24) from maternal grandparents to mothers of ASD cases, whereas they did not observe HLA-DRB1*04 transmission from mothers or fathers to offspring with ASD. These findings support a role for HLA-DRB1*04 as an ASD risk factor acting in mothers during pregnancy (ie, not due to genetic transmission from parents, but potentially through maternal action in utero), raising the possibility of a maternal immune component to ASD pathogenesis.

Finally, SLE and RA pregnancies are at increased risk of adverse obstetric outcomes, such as prematurity and SGA, which are potential risk factors for neurodevelopmental disorders. Observational studies report a 1.5 to 3.0-fold increase in neurodevelopmental disorders in children born preterm or SGA versus controls. Thus, obstetric complications in women with SLE and RA may also increase neurodevelopmental disorders in offspring.

Congenital Heart Defects

CHDs are the most frequent type of birth defects, accounting for approximately a third of all congenital anomalies ; they are associated with substantial childhood morbidity. In utero exposures, such as maternal illness and medications, are thought to play an important role in the yet to be fully elucidated etiology of CHD. In particular, a recent study suggests a threefold increased risk of CHD in children born to mothers with various systemic autoimmune rheumatic diseases ; however, the investigators did not specifically assess the SLE and RA effect estimates for risk and did not control for medication exposures.

Until recently, very few uncontrolled observational studies had assessed CHD in offspring of mothers with SLE. Notably, in a study of fetal echocardiography in a small number of SLE pregnancies, 7.5% of fetuses had a CHD, which is more than fivefold greater than that observed among live births from the general population (0.6%–1.3%). After excluding cases with CHD that could have caused congenital heart block, investigators observed CHD in 16% to 42% of children with congenital heart block born to mothers with anti-Ro/SSA antibodies. The most frequently observed CHDs were atrial septal defects, ventricular septal defects (VSD), and valve anomalies.

In 2015, Vinet and colleagues assessed the risk of CHD in SLE offspring within the large population-based OSLER cohort. In comparison with unexposed children, SLE offspring experienced more CHD (5.2% [95% CI 3.7–7.1] vs 1.9% [95% CI 1.6–2.2], difference 3.3% [95% CI 1.9–5.2]). In multivariable analyses, children born to women with SLE had a substantially increased risk of CHD (OR 2.62, 95% CI 1.77–3.88) compared with unexposed children. Subgroup analyses accounting for medication exposures were similar. In addition, offspring of mothers with SLE had a substantially increased risk of having a CHD repair procedure (OR 5.82, 95% CI 1.77–19.09). These data suggest that SLE offspring are at increased risk of CHD, and at least part of this risk might be independent of in utero medication exposures.

Data are more limited regarding the risk of CHD in RA offspring. In a large cohort study using Quebec’s administrative data (n = 8810), investigators observed a substantially increased risk of CHD in children born to women diagnosed with RA during childhood compared with children from the general population (OR 2.09, 95% CI 1.23–3.55). A CHD diagnosis was documented in 1.2% of RA offspring (95% CI 0.69–1.71) as opposed to 0.6% of control children (95% CI 0.42–0.78). Although analyses were limited by the lack of drug information, these findings suggest that RA offspring might also be at increased risk of CHD.

Maternal SLE/RA-related mechanisms that could be implicated in the physiopathology of CHD in offspring include autoantibody-mediated damage and cytokine imbalance. Anti-Ro/SSA and anti-La/SSB antibodies, found in, respectively, 40% and 20% of women with SLE and RA, cross the placenta and are associated with development of neonatal lupus, with congenital heart block being the most characteristic cardiac manifestation. Investigators have demonstrated that maternal anti-Ro/SSA and anti-La/SSB antibodies bind apoptotic fetal cardiocytes, resulting in the release of proinflammatory and profibrosing cytokines, and, ultimately, scarring. This process likely extends beyond the conduction tissue, involving the myocardium, endocardium, and valves. In a recent retrospective analysis of autopsies from 18 cardiac neonatal lupus cases, cardiac histologic damage outside of the conduction system was frequently observed. Six (40%) of 15 of deaths due to congenital heart block had pathology findings, such as fibrosis and calcification of the valves and/or valve apparatus.

Cardiac septation occurs early in embryogenesis and is complete by 6 weeks’ gestation. Because transplacental passage of maternal autoantibodies occurs later (beginning at approximately 16 weeks), it is unlikely that maternal autoantibodies directly interfere. However, muscular VSDs, which account for 75% of all VSDs, are thought to arise from foci of cellular death during active cardiac remodeling within an already formed ventricular septum. In addition, maternal autoantibodies might prevent closure of cardiac septal defects that might have closed naturally, possibly explaining the excess risk of cardiac septal defects in offspring of mothers with SLE.

aPL antibodies also cross the placenta: in one recent study, 40% of neonates born to women with antiphospholipid syndrome had positive aPL in cord blood. In aPL-positive adult patients with and without SLE, aPLs are strongly associated with valvular disease, and valvular deposits of aPL are thought to play an important pathogenic role. Although prior studies have reported perinatal thrombotic events occurring in children born to aPL-positive mothers, there are no data on the prevalence of congenital valve anomalies or other types of CHD in these children ; in theory, however, transplacental aPL could play a role in valve anomalies in exposed fetuses.

Cytokines, such as transforming growth factor beta (TGF-beta), play an important role in cardiac embryogenesis. In particular, adequate endocardial cushion formation, which is a critical step in cardiac septation, requires appropriate expression of TGF-beta. Defective levels of TGF-beta have been associated with CHD in animal models, whereas high levels have been linked to CHD in human studies. Serum levels of TGF-beta-1 are substantially lower in subjects with SLE (inversely correlating with disease activity), whereas they are higher in subjects with RA (correlating with disease activity). Because transplacental transfer of circulating TGF-beta can occur from mother to fetus, abnormal levels of maternal TGF-beta might alter normal fetal heart development, potentially leading to an increased risk of CHD.

Hematological Malignancies

Hematologic malignancies, such as leukemia and lymphoma, account for approximately 40% of new cancer diagnoses in children, with an incidence of up to 12 cases per 100,000 annually. Diffuse large B-cell lymphoma (DLBCL), a non-Hodgkin lymphoma, is one of the most common lymphomas among children and adolescents, representing 20% and 40% of new lymphoma cases in children and adolescents, respectively.

Patients with SLE and RA have an increased risk of hematological malignancies, particularly non-Hodgkin lymphoma, compared with the general population. Large population-based studies have consistently shown more than a twofold increase in the risk of leukemia and lymphoma in SLE and RA, thought to be due to chronic immune stimulation. The most common type of hematological malignancy in SLE and RA is DLBCL.

Data on hematological malignancies in RA offspring are very limited, and such data are nonexistent in SLE. In one cohort study, mothers with RA had an increased risk of having a child with lymphoma or leukemia compared with general population rates (relative risk 1.7, 95% CI 0.9–2.8). In another cohort study of offspring born to RA parents, there was a substantially increased risk of Hodgkin’s lymphoma (standardized incidence ratio 3.2, 95% CI 1.0–7.4) ; however, the investigators did not provide a specific estimate for offspring born to mothers with RA, as opposed to fathers with RA. In addition, in both studies, no attention was paid to the timing of RA diagnosis in relation to the pregnancy.

As childhood hematological cancers are thought to arise from an aberrant immune response, genetic studies have investigated the potential role of genes involved in the immune system. Notably, several investigators have reported an association between HLA-DRB1*04 alleles and a twofold increase in the risk of acute lymphoblastic leukemia in children. Thus, offspring of subjects with RA might be at increased risk of hematological malignancies through inheritance of risk alleles.

Rheumatic and Nonrheumatic Autoimmune Diseases

Several studies have suggested a familial aggregation of SLE and RA, as well as several autoimmune diseases, such as type 1 diabetes. However, only 2 studies have specifically assessed the risk of autoimmune diseases in offspring born to mothers with SLE or RA. Preliminary findings from the OSLER cohort showed a potentially twofold increase in the risk of nonrheumatic autoimmune diseases (including type 1 diabetes and inflammatory bowel diseases) in children born to women with SLE compared with children from the general population. However, the effect estimate for the risk of rheumatic autoimmune diseases was inconclusive.

In a large Danish population-based study of 13,566 RA offspring, Rom and colleagues studied children whose mothers had RA and compared them with children whose mothers did not have RA. In RA offspring, there was higher morbidity for 8/11 International Classification of Diseases groups. Similar results were observed in 6330 children whose fathers had RA. The investigators reported a substantial increase in the risk of specific rheumatic and nonrheumatic autoimmune diseases with up to a threefold increase in the risk of juvenile idiopathic arthritis (hazard ratio [HR] for maternal RA 3.30 [95% CI 2.71–4.03] and for paternal RA 2.97 [95% CI 2.20–4.01]), an increased risk of up to 40% of type 1 diabetes (HR for maternal RA 1.37 [95% CI 1.12–1.66] and for paternal RA 1.44 [95% CI 1.09–1.90]), and up to a 30% increased risk of asthma (HR for maternal RA 1.28 [95% CI 1.20–1.36] and for paternal RA 1.15 [95% CI 1.04–1.26]). As an increased risk of autoimmune diseases was found both in children exposed to maternal RA and children exposed to paternal RA, genetic factors are likely to play an important role, although this was not specifically studied by the investigators.

Interestingly, in the previously described study, increased risk of infectious diseases as well as mental and behavioral disorders were seen only in offspring exposed to maternal RA (the investigators did not provide a specific risk estimate for neurodevelopmental disorders). In addition, stronger associations were observed in children born to mothers with RA compared with children born to fathers with RA. Thus, fetal programming, through in utero exposures to maternal autoantibodies, cytokines, drugs, and obstetric complications, might all have an etiologic role in certain adverse health outcomes in offspring.