The complex challenge of antenatal steroid therapy nonresponsiveness

A review of the potential off-target effects of antenatal steroid exposures on fetal development

Antenatal steroids (ANS) are one of the most widely prescribed medications in pregnancy, being administered to women at risk of preterm delivery. In the setting of preterm delivery at or below 35 weeks' gestation, systematic review data show ANS reduce perinatal morbidity and mortality, primarily by promoting fetal lung maturation. However, with the expanding use of this intervention has come a growing appreciation for the potential off-target, adverse effects of ANS therapy on wider fetal development. We undertook a narrative literature review of the animal and clinical literature to assess current evidence for adverse effects of ANS exposure and fetal development. This review presents a summary of the evidence relating to the potential for wide-ranging, off-target, adverse effects of ANS therapy on fetal development and programming. We highlight an urgent need for further animal and clinical studies investigating the effects of ANS on the fetal immune, cardiovascular, renal and hepatic systems given a current sparsity of evidence. We also strongly suggest an emphasis on open disclosure, discussion and education of clinicians and patients with regard to the potential benefits and risks of ANS therapy, particularly in late preterm and term gestations where infants derive relatively few benefits from these drugs. We also propose further studies on the optimisation of ANS therapy through improved patient selection and improved dosing regimens based on a pharmacokinetic-pharmacodynamic informed understanding of ANS action on the fetal lung.

Single-nucleotide polymorphisms in dizygotic twin ovine fetuses are associated with discordant responses to antenatal steroid therapy

BACKGROUND

Antenatal steroid (ANS) therapy is given to women at risk of preterm delivery to accelerate fetal lung maturation. However, the benefit of ANS therapy is variable and how maternal and fetal factors contribute to this observed variability is unknown. We aimed to test the degree of concordance in preterm lung function, and correlate this with genomic, transcriptomic, and pharmacokinetic variables in preterm dizygotic twin ovine fetuses.

METHODS

Thirty-one date-mated ewes carrying twin fetuses at 123 ± 1 days' gestation received maternal intramuscular injections of either (i) 1 × 0.25 mg/kg betamethasone phosphate and acetate (CS1, n = 11 twin pairs) or (ii) 2 × 0.25 mg/kg betamethasone phosphate and acetate, 24 h apart (CS2, n = 10 twin pairs) or (iii) 2 × saline, 24 h apart (negative control, n = 10 twin pairs). Fetuses were surgically delivered 24 h after their final treatment and ventilated for 30 min.

RESULTS

ANS-exposed female fetuses had lower arterial partial pressure of carbon dioxide (PaCO2) values than male fetuses (76.5 ± 38.0 vs. 97.2 ± 42.5 mmHg), although the observed difference was not statistically significant (p = 0.1). Only 52% of ANS-treated twins were concordant for lung maturation responses. There was no difference in fetal lung tissue or plasma steroid concentrations within or between twin pairs. Genomic analysis identified 13 single-nucleotide polymorphisms (SNPs) statistically associated with ANS-responsiveness, including in the proto-oncogene MET and the transcription activator STAT1.

CONCLUSIONS

Twin fetal responses and ANS tissue levels were comparable with those from singleton fetuses in earlier studies. Twin ovine fetuses thus benefit from ANS in a similar manner to singleton fetuses, and a larger dose of betamethasone is not required. Assuming no difference in input from the placental or maternal compartments, fetal lung responses to ANS therapy in dizygotic twin preterm lambs are dependent on the fetus itself. These data suggest a potential heritable role in determining ANS responsiveness.

Optimization of the betamethasone and dexamethasone dosing regimen during pregnancy: a combined placenta perfusion and pregnancy physiologically based pharmacokinetic modeling approach

BACKGROUND

Antenatal betamethasone and dexamethasone are prescribed to women who are at high risk of premature birth to prevent neonatal respiratory distress syndrome (RDS). The current treatment regimens, effective to prevent neonatal RDS, may be suboptimal. Recently, concerns have been raised regarding possible adverse long-term neurological outcomes due to high fetal drug exposures. Data from nonhuman primates and sheep suggest maintaining a fetal plasma concentration above 1 ng/mL for 48 hours to retain efficacy, while avoiding undesirable high fetal plasma levels.

OBJECTIVE

We aimed to re-evaluate the current betamethasone and dexamethasone dosing strategies to assess estimated fetal exposure and provide new dosing proposals that meet the efficacy target but avoid excessive peak exposures.

STUDY DESIGN

A pregnancy physiologically based pharmacokinetic (PBPK) model was used to predict fetal drug exposures. To allow prediction of the extent of betamethasone and dexamethasone exposure in the fetus, placenta perfusion experiments were conducted to determine placental transfer. Placental transfer rates were integrated in the PBPK model to predict fetal exposure and model performance was verified using published maternal and fetal pharmacokinetic data. The verified pregnancy PBPK models were then used to simulate alternative dosing regimens to establish a model-informed dose.

RESULTS

Ex vivo data showed that both drugs extensively cross the placenta. For betamethasone 15.7±1.7% and for dexamethasone 14.4±1.5%, the initial maternal perfusate concentration reached the fetal circulations at the end of the 3-hour perfusion period. Pregnancy PBPK models that include these ex vivo-derived placental transfer rates accurately predicted maternal and fetal exposures resulting from current dosing regimens. The dose simulations suggest that for betamethasone intramuscular, a dose reduction from 2 dosages 11.4 mg, 24 hours apart, to 4 dosages 1.425 mg, 12 hours apart would avoid excessive peak exposures and still meet the fetal response threshold. For dexamethasone, the dose may be reduced from 4 times 6 mg every 12 hours to 8 times 1.5 mg every 6 hours.

CONCLUSION

A combined placenta perfusion and pregnancy PBPK modeling approach adequately predicted both maternal and fetal drug exposures of 2 antenatal corticosteroids (ACSs). Strikingly, our PBPK simulations suggest that drug doses might be reduced drastically to still meet earlier proposed efficacy targets and minimize peak exposures. We propose the provided model-informed dosing regimens are used to support further discussion on an updated ACS scheme and design of clinical trials to confirm the effectiveness and safety of lower doses.

Antenatal steroids elicited neurodegenerative-associated transcriptional changes in the hippocampus of preterm fetal sheep independent of lung maturation

BACKGROUND

Antenatal steroid therapy for fetal lung maturation is routinely administered to women at risk of preterm delivery. There is strong evidence to demonstrate benefit from antenatal steroids in terms of survival and respiratory disease, notably in infants delivered at or below 32 weeks' gestation. However, dosing remains unoptimized and lung benefits are highly variable. Current treatment regimens generate high-concentration, pulsatile fetal steroid exposures now associated with increased risk of childhood neurodevelopmental diseases. We hypothesized that damage-associated changes in the fetal hippocampal transcriptome would be independent of preterm lung function.

METHODS

Date-mated ewes carrying a single fetus at 122 ± 2dGA (term = 150dGA) were randomized into 4 groups: (i) Saline Control Group, 4×2ml maternal saline intramuscular(IM) injections at 12hr intervals (n = 11); or (ii) Dex High Group, 2×12mg maternal IM dexamethasone phosphate injections at 12hr intervals followed by 2×2ml IM saline injections at 12hr intervals (n = 12; representing a clinical regimen used in Singapore); or (iii) Dex Low Group, 4×1.5mg maternal IM dexamethasone phosphate injections 12hr intervals (n = 12); or (iv) Beta-Acetate Group, 1×0.125mg/kg maternal IM betamethasone acetate injection followed by 3×2ml IM sterile normal saline injections 12hr intervals (n = 8). Lambs were surgically delivered 48hr after first maternal injection at 122-125dGA, ventilated for 30min to establish lung function, and euthanised for necropsy and tissue collection.

RESULTS

Preterm lambs from the Dex Low and Beta-Acetate Groups had statistically and biologically significant lung function improvements (measured by gas exchange, lung compliance). Compared to the Saline Control Group, hippocampal transcriptomic data identified 879 differentially significant expressed genes (at least 1.5-fold change and FDR < 5%) in the steroid-treated groups. Pulsatile dexamethasone-only exposed groups (Dex High and Dex Low) had three common positively enriched differentially expressed pathways related in part to neurodegeneration ("Prion Disease", "Alzheimer's Disease", "Arachidonic Acid metabolism"). Adverse changes were independent of respiratory function during ventilation.

CONCLUSIONS

Our data suggests that exposure to antenatal steroid therapy is an independent cause of damage- associated transcriptomic changes in the brain of preterm, fetal sheep. These data highlight an urgent need for careful reconsideration and balancing of how antenatal steroids are used, both for patient selection and dosing regimens.

Antenatal corticosteroids: an updated assessment of anticipated benefits and potential risks

Antenatal steroid therapy is increasingly central to the obstetrical management of women at imminent risk of preterm birth. For women likely to deliver between 24 and 34 weeks' gestation, antenatal steroid therapy is the standard of care, conferring sizable benefits and few risks in high-resource environments when appropriately targeted. Recent studies have focused on antenatal steroid use in periviable and late preterm populations, and in term cesarean deliveries. As a result, antenatal steroid therapy has now been applied from 22 to 39+6 weeks of estimated gestational age. There is also an increased appreciation that the vast majority of randomized control data informing the use of antenatal steroids are derived from predominantly high-resource, White populations. Accordingly, a sizable amount of work has recently been undertaken to test how to safely use antenatal steroids in low- and middle-resource environments, wherein the often high rates of preterm birth make these low-cost, easily administered interventions an attractive proposition. It is likely underappreciated by the obstetrical and neonatal communities that the overall efficacy of antenatal steroid therapy is highly variable (including when preterm risk is accurately assessed), the treatment regimens used are largely arbitrary, dosing is suprapharmacologic for effect, and the benefit-risk balance is significantly and differentially modified by gestation. It is also very likely that the patients consenting to receive these treatments are similarly unaware of the complex balance of potential benefits and harms. Although a small number of follow-up studies present a generally benign picture of long-term antenatal steroid risk, several large, population-based retrospective studies have identified associations between antenatal steroid use, childhood mental disease, and newborn infections that warrant urgent attention. Of particular contemporary importance are emergent efforts to optimize antenatal steroid regimens on the basis of the pharmacokinetics and pharmacodynamics of the agents themselves, the need for better targeting of these potent drugs, and clear articulation of the potential benefits and harms of antenatal steroid use at differing stages of pregnancy and in different delivery contexts.

Single nucleotide polymorphisms in surfactant protein A1 are not associated with a lack of responsiveness to antenatal steroid therapy in a pregnant sheep model

Treatment with antenatal steroids (ANS) is standard practice for reducing the risk of respiratory distress in the preterm infant. Despite clear overall benefits when appropriately administered, many fetuses fail to derive benefit from ANS therapies. In standardized experiments using a pregnant sheep model, we have demonstrated that around 40% of ANS-exposed lambs did not have functional lung maturation significantly different from that of saline-treated controls. Surfactant protein A is known to play an important role in lung function. In this genotyping study, we investigated the potential correlation between polymorphisms in SFTPA1, messenger RNA and protein levels, and ventilation outcomes in animals treated with ANS. 45 preterm lambs were delivered 48 h after initial ANS therapy and 44 lambs were delivered 8 days after initial ANS therapy. The lambs were ventilated for 30 min after delivery. SFTPA1 mRNA expression in lung tissue was not correlated with arterial blood PaCO2 values at 30 min of ventilation in lambs delivered 48 h after treatment. SFTPA1 protein in lung tissue was significantly correlated with PaCO2 at 30 min of ventilation in lambs ventilated both 48 h and 8 days after ANS treatment. Six different single nucleotide polymorphisms (SNPs) in the Ovis aries SFTPA1 sequence were detected by Sanger Sequencing. No individual SNPs or SNP haplotypes correlated with alterations in PaCO2 at 30 min of ventilation or SFTPA1 protein levels in the lung. For the subset of animals analyzed in the present study, variable lung maturation responses to ANS therapy were not associated with mutations in SFTPA1.