Antenatal diagnosis and maternal sirolimus treatment of polyhydramnios, megalencephaly, and symptomatic epilepsy (PMSE) syndrome

Abstract

Background

Polyhydramnios, megalencephaly, and symptomatic epilepsy (PMSE) syndrome is a rare autosomal recessive mTORopathy caused by biallelic STE20‐related kinase adaptor alpha (STRADA) loss‐of‐function variants. Animal models demonstrate that in utero mechanistic target of rapamycin (mTOR) inhibition can prevent cortical dyslamination, suggesting a prenatal therapeutic window.

Methods

A gravida 3 para 2 patient presented at 32+2 weeks with medication‐refractory preterm labor and severe polyhydramnios. Fetal magnetic resonance imaging (MRI) demonstrated macrocephaly and an enlarged cavum septum pellucidum (CSP). Amniocentesis with microarray revealed homozygosity across the STRADA locus; sequencing confirmed the founder STRADA exons 9–13 deletion. After counseling, maternal sirolimus therapy was initiated at 35+6 weeks using a 20 mg loading regimen followed by 6–8 mg/day maintenance dosing. Serial ultrasound monitored fetal biometry and CSP size. Delivery occurred at 39+1 weeks. Pediatric follow‐up was extended nearly 5 years.

Results

Maternal sirolimus was well tolerated. Amniotic fluid index (AFI) declined from 36.1 to 26.2 cm, head circumference normalized from the 89th to 35th percentile, and CSP diameter decreased from 14.0 to 12.3 mm. A term vaginal delivery produced a 3554 g infant with normocephaly. Cord blood sirolimus level (3.9 mcg/L) approximated maternal level (3.4 mcg/L). Postnatally, sirolimus was continued from day of life 7. The first seizure occurred at 4.5 months, coincident with subtherapeutic trough levels; seizure control was achieved thereafter. At nearly 5 years, the child shows adequate growth but severe global developmental delay typical of PMSE.

This case demonstrates the feasibility and pharmacokinetic effectiveness of maternal sirolimus during late gestation, with no identified maternal or fetal adverse consequences in this first case of third‐trimester prenatal mTOR inhibition in confirmed PMSE.

1. INTRODUCTION

Polyhydramnios, megalencephaly, and symptomatic epilepsy (PMSE) syndrome is caused by biallelic STE20‐related kinase adaptor alpha (STRADA) loss, resulting in constitutive mTORC1 hyperactivation and profound neurodevelopmental disability (Figure 1) [1]. The syndrome is strongly associated with a founder variant in the Old Order Mennonite community (and their descendants) and is characterized by severe third‐trimester polyhydramnios, macrocephaly, cortical dyslamination, intractable epilepsy, and high early childhood mortality (Figure 2), with approximately 40% dying by age 6 [2]. Fewer than 100 individuals with PMSE have been identified worldwide, predominantly within founder populations, making the accumulation of comparative data extremely challenging.

FIGURE 1.

mTOR signaling pathway showing STRADA/PMSE defect and sirolimus mechanism of action. STRADA normally activates LKB1, which activates AMPK, which inhibits mTORC1 through TSC1/TSC2. Loss of STRADA results in unopposed mTORC1 activity. Sirolimus directly inhibits mTORC1, bypassing the upstream defect. AKT, protein kinase B; BDNF, brain‐derived neurotrophic factor; GTP, guanosine triphosphate; LKB1, liver kinase B1; mTOR, mechanistic target of rapamycin; NGF, nerve growth factor; PI3K, phosphatidylinositol 3‐kinase; PMSE, polyhydramnios, megalencephaly, and symptomatic epilepsy; STRADA, STE20‐related kinase adaptor alpha; TGF, transforming growth factor; TSC, tuberous sclerosis complex.

FIGURE 2.

mTOR pathway legend and key concepts. This figure illustrates the relationship between STRADA function, PMSE pathology, sirolimus mechanism, and the critical importance of intervention timing during fetal cortical development. AMPK, adenosine monophosphate‐activated protein kinase; LKB1, liver kinase B1; mTOR, mechanistic target of rapamycin; PMSE, polyhydramnios, megalencephaly, and symptomatic epilepsy; STRADA, STE20‐related kinase adaptor alpha; TSC, tuberous sclerosis complex.

Preclinical studies demonstrate that rapamycin administered during murine cortical neurogenesis prevents cortical dyslamination and rescues neurological phenotypes [1]. These findings suggest a human therapeutic window during late second to early third trimester. mechanistic target of rapamycin (mTOR) inhibitors are established treatments for other mTORopathies, particularly tuberous sclerosis complex (TSC), where everolimus improves seizure control and tumor burden [3]. Limited prenatal experience in TSC‐associated cardiac rhabdomyomas demonstrates that maternal therapy can achieve fetal exposure and reduce tumor mass [4, 5]. Pregnancy outcome data from transplant recipients treated with sirolimus show acceptable safety profiles [6]. Unlike TSC, where prenatal mTOR inhibition targets discrete tumors, PMSE presents a diffuse cortical migration disorder requiring intervention during a narrow developmental window.

We report the first antenatal diagnosis and maternal sirolimus treatment of PMSE, with long‐term follow‐up. Although therapy began late in gestation, the case provides key feasibility and pharmacokinetic insights supporting earlier prenatal intervention in future cases.

2. MATERIALS AND METHODS

2.1. Clinical presentation

A 28‐year‐old gravida 3 para 2 Lancaster Mennonite woman presented at 32+2 weeks gestation with severe polyhydramnios (amniotic fluid index [AFI] > 40 cm) and medication‐refractory preterm labor. She received nifedipine (40 mg loading, then 20 mg q8h), magnesium sulfate (4 g bolus) for neuroprotection, and betamethasone 12 mg IM for fetal lung maturity [7], but continued contracting every 2–4 min with progressive cervical dilation. Maternal‐fetal medicine consultation was requested at this admission.

Ultrasound demonstrated macrocephaly (head circumference 316.2 mm, 91st percentile; all other biometry was < 50th percentile) and enlarged cavum septum pellucidum (CSP) (14 mm). Amniocentesis was performed [8], removing 1200 mL of fluid, providing significant relief of preterm labor symptoms.

2.2. Genetic diagnosis

Fluorescence in situ hybridization (FISH) and karyotype were normal (46, XY). Microarray revealed homozygosity at 17q22–17q24.1 encompassing STRADA. Targeted sequencing at the Clinic for Special Children confirmed the founder deletion of exons 9–13. The family subsequently discovered cases of PMSE in both paternal and maternal lineages.

2.3. Counseling and treatment decision

Counseling reviewed disease prognosis, preclinical evidence for prenatal mTOR inhibition, human safety data, timing limitations, and uncertainty of neurological benefit. Alternative options included postnatal therapy alone or pregnancy termination. After meeting with affected families and extensive deliberation, the patient elected therapy under institutional ethics approval through a “right to try” framework [9] and written informed consent was obtained prior to treatment.

2.4. Sirolimus treatment protocol

A 72‐h treatment hold was implemented for infection risk mitigation when the family developed upper respiratory symptoms (Days 15–17), as sirolimus immunosuppression could potentially exacerbate viral illness; treatment resumed 72 h prior to induction.

Serial ultrasounds assessed fetal response at 1–2 week intervals. Maternal laboratory monitoring (complete blood count [CBC], comprehensive metabolic panel [CMP], triglycerides, urinalysis) was performed weekly and showed no toxicity. Sirolimus trough levels were measured on Days 9, 14, and 21, with values of 9.0, 7.2, and 3.4 mcg/L, respectively. The loading regimen was based on published pregnancy pharmacokinetic data from transplant recipients, targeting the lower therapeutic range (3–18 mcg/L). Dose escalation criteria included trough levels below 5 mcg/L with ongoing polyhydramnios. Neonatal monitoring included CBC, hepatic function, and sirolimus levels at delivery and Day 7.

2.5. Delivery and follow‐up

The patient underwent uncomplicated induction at 39+1 weeks (treatment day 23). Cord and maternal blood samples were obtained for sirolimus measurement. Postnatal sirolimus was initiated on Day 7, with pediatric neurology follow‐up through 5 years.

3. RESULTS

3.1. Maternal response

Dramatic improvements in polyhydramnios and fetal biometry occurred over 3 weeks (Table S1; Figures 3 and 4)

1. AFI: 36.1 cm (measured at treatment initiation, post‐amnioreduction) → 26.2 cm (27% reduction)

2. Head circumference: 89th → 35th percentile

3. CSP diameter: 14.0 mm → 12.3 mm (12% reduction, Figure 5); normal CSP width at term is typically 3–6 mm.

FIGURE 3.

Fetal head circumference trajectory during maternal sirolimus therapy. Head circumference decreased from 89th to 35th percentile over 3 weeks of treatment, from initiation at 35+6 weeks through delivery at 39+1 weeks.

FIGURE 4.

Amniotic fluid index (AFI) response during therapy. AFI decreased from 36.1 to 26.2 cm, representing a 27% reduction that facilitated vaginal delivery and eliminated the need for repeated amnioreductions.

FIGURE 5.

Change in cavum septum pellucidum (CSP) diameter. CSP diameter decreased from 14.0 to 12.3 mm (12% reduction) during the 3‐week treatment period, suggesting potential effects on brain morphology even at this late gestational age.

These changes conferred significant maternal benefit by facilitating uncomplicated vaginal delivery, avoiding the 60%–70% cesarean rate typical for PMSE due to polyhydramnios complications and cephalopelvic disproportion from fetal macrocephaly.

3.2. Delivery and neonatal period

A healthy 3554 g male infant was delivered vaginally with Apgar scores 9/9. Head circumference was 36.2 cm (57th percentile), demonstrating normocephaly. Physical examination revealed minor stigmata of PMSE: joint laxity and slight ribcage widening.

Cord sirolimus level (3.9 mcg/L) approximated maternal level (3.4 mcg/L), demonstrating cord: maternal ratio of 1.15 and confirming efficient placental transfer. This validates maternal dosing strategies for achieving therapeutic fetal levels.

Initial feeding difficulties required gavage support until Day 40, typical of PMSE and likely reflecting underlying neurological dysfunction. The infant was discharged on Day 3 with coordinated follow‐up.

3.3. Postnatal course

Postnatal sirolimus therapy was initiated on Day 7 at 0.75 mg daily, targeting trough levels of 10 mcg/L per TSC protocols. By 8 months, 2 mg daily was required to maintain therapeutic levels.

During the first 5 months, the infant demonstrated early developmental milestones with supportive therapy, showing persistent hypotonia of the neck and shoulder girdle. At 4.5 months, the infant experienced his first clinical seizure coincident with subtherapeutic sirolimus levels. Oxcarbazepine was initiated, and following restoration of therapeutic sirolimus levels, seizure freedom was achieved at 8 months.

3.4. Long‐term outcomes

At 15 months, the child could sit independently with arm support, produce vowel sounds, and roll front‐to‐back. However, developmental progress subsequently plateaued. Sirolimus was discontinued after 2 years; oxcarbazepine monotherapy continued.

Breakthrough seizures emerged at age 4, prompting valproate addition. At nearly 5 years, the child demonstrates adequate somatic growth but significant developmental impairment requiring 24‐h skilled nursing care. He tolerates pureed foods, remains non ambulatory and nonverbal, and attends school 3 days per week with developmental support. Overall, the postnatal course resembles that of other individuals with PMSE.

Additional clinical and pharmacokinetic data are provided in Table S2.

4. DISCUSSION

4.1. Clinical significance

This case establishes that maternal sirolimus therapy in late gestation is feasible, well‐tolerated, and capable of producing measurable fetal improvements in polyhydramnios and macrocephaly. Efficient placental transfer was demonstrated by concordant maternal and cord blood levels, consistent with prior human prenatal mTOR inhibitor experiences [4, 5].

4.2. Maternal benefits

Tangible maternal benefits included polyhydramnios reduction (27%) and fetal head circumference normalization, enabling successful vaginal delivery. Given that cesarean rates in PMSE approach 60%–70% due to polyhydramnios complications and cephalopelvic disproportion, these maternal benefits alone justify considering in utero sirolimus therapy independent of fetal neurological benefit.

4.3. Neurodevelopmental outcomes

Neurological outcomes were unchanged from the natural history of PMSE [2], aligning with mechanistic data showing that intervention after cortical layer formation is unlikely to provide neurodevelopmental benefit [1]. In contrast, early intervention in related mTORopathies such as TSC has demonstrated improved outcomes when initiated during infancy [10].

The late timing of intervention (35+6 weeks) occurred after critical cortical development (16–24 weeks in humans), explaining limited long‐term neurological benefit despite dramatic effects on polyhydramnios and head circumference. The STRADA mouse studies demonstrate that prenatal rapamycin during cortical neurogenesis (embryonic Days 15–19, equivalent to human 20–32 weeks) nearly completely normalized cortical architecture and rescued behavioral phenotypes, while postnatal treatment provided minimal benefit. Importantly, the observed CSP reduction (14.0–12.3 mm) falls within measurement variability of 2–3 mm typical for near‐term CSP assessment, and CSP diameter can fluctuate independent of underlying pathology. Similarly, while AFI reduction was substantial, polyhydramnios in PMSE may partially resolve spontaneously in late gestation. These biomarker changes, though temporally associated with treatment, cannot establish causality in a single uncontrolled case [1].

4.4. Rationale for earlier intervention

Current prenatal diagnostic capabilities enable earlier intervention. Cell‐free fetal DNA screening identifies at‐risk pregnancies in founder populations. CVS (11–13 weeks) or amniocentesis (15–16 weeks) provides a definitive diagnosis. Fetal magnetic resonance imaging (MRI) (18–20 weeks) documents baseline brain morphology. Together, these enable diagnosis by late second trimester—the optimal window identified in animal studies.

Based on mouse data and human cortical timelines, intervention should ideally begin at 20–24 weeks and continue through 32–34 weeks, covering active cortical migration. Target maternal trough levels of 8–12 mcg/L would balance efficacy with safety, with close fetal surveillance identifying concerning growth restriction.

4.5. Broader applications

This case joins an expanding literature on prenatal molecular‐targeted interventions, including enzyme replacement for lysosomal storage disorders, in utero hematopoietic stem cell transplantation for severe combined immunodeficiency, and prenatal asfotase alfa for hypophosphatasia. Regarding placental transfer, sirolimus demonstrates superior transplacental passage compared to everolimus, supporting its selection for fetal therapy. These principles apply to other mTORopathies, particularly TSC. Given epileptogenesis in tuberous sclerosis complex (EPISTOP) trial evidence that early postnatal mTOR inhibition improves TSC outcomes [10], and proof‐of‐concept for treating fetal cardiac rhabdomyomas [4, 5], prenatal intervention for severe TSC cases is increasingly rational. Phosphatase and tensin homolog (PTEN) hamartoma syndrome and hemimegaloencephaly might also benefit.

An open‐label Phase I/II trial in PMSE is ethically justifiable given rarity and poor prognosis. Key elements would include: confirmed biallelic STRADA mutations diagnosed by 24 weeks; maternal sirolimus initiated at diagnosis; primary outcomes of safety, drug levels, and successful delivery; secondary outcomes including fetal MRI changes, seizure onset, and developmental milestones through age 5–10.

4.6. Ethical considerations

The regulatory pathway was justified by the absence of U.S. Food and Drug Administration (FDA)‐approved treatments for PMSE, the severe prognosis, completion of safety studies in related conditions, and explicit informed consent.

Fundamental to any fetal therapy is respect for maternal autonomy. The mother must consent to treatment, bear any risks, and make decisions on behalf of the fetus. The family's right to try an experimental therapy with scientific support deserves consideration, given the uniform severity and high mortality.

4.7. Temporal precision medicine

This case exemplifies temporal precision medicine: timing matters as much as molecular targeting. The 2015 Precision Medicine Initiative emphasized “the right drug at the right dose to the right patient” [11]. Our case extends this to include temporal precision: at the right time.

Human cortical neurogenesis and migration peak between 16 and 22 weeks of gestation. The STRADA mouse studies illustrate windows of vulnerability and opportunity: the developing cortex is vulnerable to mTOR hyperactivation during migration, but this window represents an intervention opportunity [1]. After layer formation, neither persists. These concepts align with a broader understanding of developmental timing from initiatives like DARPA's BIOCHRONICITY program [12] and the 4D Nucleome project [13].

4.8. Limitations

Important limitations include: (1) single case without controls; (2) late timing beyond critical cortical development; (3) postnatal treatment confounding attribution of outcomes; (4) outcome assessment at only 5 years in a variable‐course disease; (5) limited generalizability to other mTORopathies; (6) potential expectation bias in subjective developmental assessments by clinicians aware of treatment status; (7) lack of standardized neurodevelopmental testing instruments; and (8) the developmental plateau at approximately 2 years coincided with sirolimus discontinuation, though causality cannot be established.

5. CONCLUSIONS

The intervention was feasible, well‐tolerated without identified adverse effects, and associated with meaningful maternal benefits

Long‐term neurodevelopmental outcomes at nearly 5 years provide important observational data, though attribution specifically to prenatal versus postnatal intervention remains uncertain. Regardless of definitive efficacy, this case establishes proof‐of‐concept for maternal‐fetal mTOR inhibitor pharmacotherapy and provides critical dosing, monitoring, and safety data.

Most importantly, this case builds the scientific and clinical rationale for prospective trials of second‐trimester intervention in antenatally diagnosed mTORopathies. The convergence of improved prenatal diagnostic capability, compelling preclinical evidence for timing‐dependent efficacy, expanding clinical mTOR inhibitor experience including in pregnancy, and devastating natural history creates an imperative to pursue earlier intervention studies.

We propose that sirolimus warrants formal investigation for prenatal therapy. An open‐label Phase I/II trial enrolling pregnancies with antenatally diagnosed PMSE or TSC, with maternal sirolimus initiated at 20–24 weeks and continued through critical cortical development, represents a feasible and ethically justified next step. Such a trial would provide definitive data on safety, pharmacokinetics, target engagement, and preliminary efficacy signals to guide future development.

The broader implication extends beyond mTORopathies to encompass a new paradigm of temporally precise, pathway‐targeted prenatal medicine. As understanding of developmental biology deepens and diagnostic and therapeutic tools advance, the fetus increasingly becomes accessible as a patient who may benefit from precisely timed interventions during critical developmental windows. Success requires integration of genomics, developmental biology, maternal‐fetal medicine, pharmacology, and ethics, exemplified by this case.

AUTHOR CONTRIBUTIONS

Christian Macedonia: Conceptualization; writing—original draft; methodology; validation; formal analysis; writing—review and editing; investigation. Vincent J. Carson: Conceptualization; writing—review and editing; investigation. D. Holmes Morton: Conceptualization; writing—review and editing. Karin J. Blakemore: Writing—review and editing. Mark I. Evans: Writing—review and editing. Peter B. Crino: Writing—review and editing.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

FUNDING INFORMATION

The authors received no specific funding for this work.

ETHICS STATEMENT

This case was managed with institutional ethics committee chair case review and IRB exemption under right to try doctrine. Written informed consent was obtained from the patient for treatment.

Supporting information

Supporting Information

DATA AVAILABILITY STATEMENT

Additional clinical data are available upon request with appropriate privacy protections and family consent.