Maternal Thyroid Peroxidase Antibodies and Neurodevelopment of the Offspring: Innocent Bystander or Guilty Accomplice

Seropositivity for thyroid peroxidase antibody (TPOAb) occurs in 5–14% of pregnant women and reflects a diagnosis of thyroid autoimmunity. ¹ The presence of TPOAbs could be used as a surrogate marker for reduced thyroid functional capacity. TPOAb positivity may also serve as a nonspecific marker for a broader autoimmune imbalance. ² Compared with pregnant women who are negative for TPOAbs, TPOAb-positive women tend to have higher thyrotropin (TSH) and lower free T4 (fT4) levels and are at higher risk of developing thyroid hypofunction during gestation. ³ TPOAb positivity in and of itself is associated with adverse pregnancy outcomes including miscarriage and premature delivery, which has been supported by convincing evidence. ^1,2

The past two decades have witnessed our attempts to understand whether thyroid autoimmunity in pregnant women might be linked to impaired neurodevelopment in their offspring. Previous publications have shown mixed results. For example, Derakhshan et al. ⁴ reported that maternal TPOAb positivity in early pregnancy was associated with lower mean child intelligence quotient (IQ) in a Dutch, iodine-sufficient population-based cohort (Generation R) but not in a mildly iodine-deficient cohort from the United Kingdom (the Avon Longitudinal Study of Parents and Children, ALSPAC). The findings were unaffected by adjusting for maternal TSH or fT4 or urinary iodine/creatinine ratio. ⁴ Other studies that assessed offspring’s cognitive and motor scales and emotional and behavioral problems have yielded inconsistent results. ^1,5 Since available data are still sparse and highly heterogeneous in study design, further research is warranted.

Teng et al. ⁶ now report in this issue of Thyroid their results as part of the Ma’anshan Birth Cohort (MABC) prospective study, addressing the sex-specific association between maternal TPOAb positivity and cognitive development of their offspring during the preschool years. A total of 1849 Chinese mother–child pairs in a non-iodine-deficiency region were included. The mothers had no preconception thyroid dysfunction or family history of thyroid diseases. In pregnancy, the mothers were not taking levothyroxine and had no adverse gestational outcomes. Their children underwent cognitive assessments at the mean age of 57.2 (SD, 5.7) months, using the Wechsler Preschool and Primary Scale of Intelligence-Fourth Edition. Maternal thyroid function and TPOAbs were retrospectively measured at early, middle, and late gestation stages using the serum stored in the sample bank during childhood visits. The authors found that maternal TPOAb positivity at early gestation was associated with an increased risk of below-average processing speed index in girls and below-average full-scale intelligence quotient in boys, while maternal TPOAb positivity at late gestation was associated with below-average working memory index (WMI) in girls. In addition, girls who were born to women with a high TPOAb trajectory throughout pregnancy were more likely to develop below-average WMI. ⁶

In a prior study, ⁷ the same research team included 2455 mother–child pairs from the MABC. They utilized the Child Behavior Checklist 1.5–5 to evaluate offspring’s emotional and behavioral development at preschool age. The authors reported that in euthyroid mothers, exposure to persistent TPOAb positivity across gestation was associated with autism spectrum problems in their boys, whereas the presence of TPOAbs at the third trimester of pregnancy was associated with depressive problems in the girls. ⁷

Taken together, Teng et al. ^6,7 have presented comprehensive data suggesting the link between positive maternal TPOAbs and poorer neurodevelopmental outcomes of offspring, encompassing aspects of cognition, emotion, and behavior. Among several strengths of their work, two are particularly worth mentioning. First, the analyses were based on a large-sample birth cohort study, neurodevelopmental outcomes, and potential confounders that were evaluated in detail prospectively, minimizing selection and recall bias. Second, trimester- and sex-specific associations between positive maternal TPOAbs exposure and offsprings’ neurodevelopmental performance were investigated. This newly acquired evidence suggests that the relationship of maternal TPOAb positivity with neurodevelopment in offspring may be more intricate than previously assumed and suggests a potential explanation for the inconsistent findings of former studies. In addition, it extends the dialog on potential sex-related disparities and critical time windows for the effects of maternal thyroid autoimmunity exposure on neurodevelopmental issues.

While an association between maternal TPOAbs and impaired neurodevelopment in offspring appears evident, the question remains whether TPOAbs is an innocent bystander that is a marker of subtle thyroid hypofunction and/or generalized autoimmune dysfunctions, or a guilty accomplice that has direct impact on neurodevelopment.

From the data available to date, the TPOAb-associated neurodevelopmental problems are unlikely to be mediated by insufficient thyroid hormone availability because of the following arguments: (i) adverse outcomes were observed in children of euthyroid mothers exhibiting TPOAbs antibodies during the late stage of gestation, when the fetal neurodevelopment barely relies on maternal thyroid hormones ^{6 –8} ; (ii) after adjusting for maternal TSH or fT4 (Generation R) ⁴ or sensitivity analyses according to maternal TSH (MABC), ⁶ the association between TPOAbs and impaired neurodevelopmental outcomes remains robust; and (iii) randomized clinical trials ^9,10 have failed to demonstrate benefits in child IQ when treating mild maternal hypothyroidism with levothyroxine. Thus, other underlying mechanisms need to be further explored. One hypothesis suggests that the thyroid autoantibodies passing through the placental barrier could potentially exert direct effects on offspring’s brain growth and neurodevelopment despite unaffected fetal thyroid function. ^1,11 However, in an experimental autoimmune thyroiditis mice model, isolated positive maternal thyroid autoantibodies did not alter learning ability and memory or nerve cell migration despite a transient increase in thyroglobulin antibody (TgAb) in neonatal mice. ¹² Alternatively, an underlying maternal immune dysfunction reflected by TPOAb positivity might interfere with neurodevelopment through maternal antibodies that recognize proteins in the developing fetal brain. ¹³ Moreover, inspired by the current study, whether sex-related factors might be involved in possible mechanisms suggests additional areas of research to be explored.

Until more large-scale prospective data from different populations and well-powered interventional trials become available, it will remain unclear whether TPOAbs is truly a guilty accomplice or just an innocent bystander. At present, evidence is lacking to support the benefit of levothyroxine or immunosuppressant treatment in euthyroid TPOAb-positive women for the purpose of improving offspring’s neurodevelopment. Hence, the implications of the current study should not be overstated. It is important to communicate these associations with caution and highlight that causality has not been established. The maternal psychological stress and anxiety that may result from the stated “transgenerational harms of TPOAbs” have the potential to outweigh the risks related to the underlying laboratory abnormality.