Publications

OBJECTIVE: Eighty percent of patients with a diagnosis of tetralogy of Fallot (TOF) do not have a known genetic etiology or syndrome. We sought to identify key molecular pathways and biological processes that are enriched in non-syndromic TOF, the most common form of cyanotic congenital heart disease, rather than single driver genes to elucidate the pathogenesis of this disease.

METHODS: We undertook exome sequencing of 362 probands with non-syndromic TOF and their parents within the Pediatric Cardiac Genomics Consortium (PCGC). We identified rare (minor allele frequency <1 × 10-4), de novo variants to ascertain pathways and processes affected in this population to better understand TOF pathogenesis. Pathways and biological processes enriched in the PCGC TOF cohort were compared to 317 controls without heart defects (and their parents) from the Simons Foundation Autism Research Initiative (SFARI).

RESULTS: A total of 120 variants in 117 genes were identified as most likely to be deleterious, with CHD7, CLUH, UNC13C, and WASHC5 identified in two probands each. Gene ontology analyses of these variants using multiple bioinformatic tools demonstrated significant enrichment in processes including cell cycle progression, chromatin remodeling, myocyte contraction and calcium transport, and development of the ventricular septum and ventricle. There was also a significant enrichment of target genes of SOX9, which is critical in second heart field development and whose loss results in membranous ventricular septal defects related to disruption of the proximal outlet septum. None of these processes was significantly enriched in the SFARI control cohort.

CONCLUSION: Innate molecular defects in cardiac progenitor cells and genes related to their viability and contractile function appear central to non-syndromic TOF pathogenesis. Future research utilizing our results is likely to have significant implications in stratification of TOF patients and delivery of personalized clinical care.

BACKGROUND: Congenital heart disease (CHD) is the most common major congenital anomaly and causes significant morbidity and mortality. Epidemiologic evidence supports a role of genetics in the development of CHD. Genetic diagnoses can inform prognosis and clinical management. However, genetic testing is not standardized among individuals with CHD. We sought to develop a list of validated CHD genes using established methods and to evaluate the process of returning genetic results to research participants in a large genomic study.

METHODS: Two-hundred ninety-five candidate CHD genes were evaluated using a ClinGen framework. Sequence and copy number variants involving genes in the CHD gene list were analyzed in Pediatric Cardiac Genomics Consortium participants. Pathogenic/likely pathogenic results were confirmed on a new sample in a clinical laboratory improvement amendments-certified laboratory and disclosed to eligible participants. Adult probands and parents of probands who received results were asked to complete a post-disclosure survey.

RESULTS: A total of 99 genes had a strong or definitive clinical validity classification. Diagnostic yields for copy number variants and exome sequencing were 1.8% and 3.8%, respectively. Thirty-one probands completed clinical laboratory improvement amendments-confirmation and received results. Participants who completed postdisclosure surveys reported high personal utility and no decision regret after receiving genetic results.

CONCLUSIONS: The application of ClinGen criteria to CHD candidate genes yielded a list that can be used to interpret clinical genetic testing for CHD. Applying this gene list to one of the largest research cohorts of CHD participants provides a lower bound for the yield of genetic testing in CHD.

BACKGROUND: Paenibacillus thiaminolyticus is a cause of postinfectious hydrocephalus among Ugandan infants. To determine whether Paenibacillus spp is a pathogen in neonatal sepsis, meningitis, and postinfectious hydrocephalus, we aimed to complete three separate studies of Ugandan infants. The first study was on peripartum prevalence of Paenibacillus in mother-newborn pairs. The second study assessed Paenibacillus in blood and cerebrospinal fluid (CSF) from neonates with sepsis. The third study assessed Paenibacillus in CSF from infants with hydrocephalus.

METHODS: In this observational study, we recruited mother-newborn pairs with and without maternal fever (mother-newborn cohort), neonates (aged ≤28 days) with sepsis (sepsis cohort), and infants (aged ≤90 days) with hydrocephalus with and without a history of neonatal sepsis and meningitis (hydrocephalus cohort) from three hospitals in Uganda between Jan 13, 2016 and Oct 2, 2019. We collected maternal blood, vaginal swabs, and placental samples and the cord from the mother-newborn pairs, and blood and CSF from neonates and infants. Bacterial content of infant CSF was characterised by 16S rDNA sequencing. We analysed all samples using quantitative PCR (qPCR) targeting either the Paenibacillus genus or Paenibacillus thiaminolyticus spp. We collected cranial ultrasound and computed tomography images in the subset of participants represented in more than one cohort.

FINDINGS: No Paenibacillus spp were detected in vaginal, maternal blood, placental, or cord blood specimens from the mother-newborn cohort by qPCR. Paenibacillus spp was detected in 6% (37 of 631 neonates) in the sepsis cohort and, of these, 14% (5 of 37 neonates) developed postinfectious hydrocephalus. Paenibacillus was the most enriched bacterial genera in postinfectious hydrocephalus CSF (91 [44%] of 209 patients) from the hydrocephalus cohort, with 16S showing 94% accuracy when validated by qPCR. Imaging showed progression from Paenibacillus spp-related meningitis to postinfectious hydrocephalus over 1-3 months. Patients with postinfectious hydrocephalus with Paenibacillus spp infections were geographically clustered.

INTERPRETATION: Paenibacillus spp causes neonatal sepsis and meningitis in Uganda and is the dominant cause of subsequent postinfectious hydrocephalus. There was no evidence of transplacental transmission, and geographical evidence was consistent with an environmental source of neonatal infection. Further work is needed to identify routes of infection and optimise treatment of neonatal Paenibacillus spp infection to lessen the burden of morbidity and mortality.

FUNDING: National Institutes of Health and Boston Children's Hospital Office of Faculty Development.

Neonatal infections due to Paenibacillus species have increasingly been reported over the last few years. We performed a structured literature review of human Paenibacillus infections in infants and adults to compare the epidemiology of infections between these distinct patient populations. Thirty-nine reports describing 176 infections met our inclusion criteria and were included. There were 37 Paenibacillus infections occurring in adults caused by 23 species. The clinical presentations of infections were quite variable. In contrast, infections in infants were caused by only 3 species: P. thiaminolyticus (112/139, 80%), P. alvei (2/139, 1%) and P. dendritiformis (2/139, 1%). All of the infants with Paenibacillus infection presented with a sepsis syndrome or meningitis, often complicated by extensive cerebral destruction and hydrocephalus. Outcomes were commonly poor with 17% (24/139) mortality. Cystic encephalomalacia due to brain destruction was common in both Ugandan and American cases and 92/139 (66%) required surgical management of hydrocephalus following their infection. Paenibacillus infections are likely underappreciated in infants and effective treatments are urgently needed.

BACKGROUND: Therapeutic hypothermia (TH) is standard of care for moderate to severe neonatal hypoxic ischemic encephalopathy (HIE) but many survivors still suffer lifelong disabilities and benefits of TH for mild HIE are under active debate. Development of objective diagnostics, with sensitivity to mild HIE, are needed to select, guide, and assess response to treatment. The objective of this study was to determine if cerebral oxygen metabolism (CMRO2) in the days after TH is associated with 18-month neurodevelopmental outcomes as the first step in evaluating CMRO2's potential as a diagnostic for HIE. Secondary objectives were to compare associations with clinical exams and characterise the relationship between CMRO2 and temperature during TH.

METHODS: This was a prospective, multicentre, observational, cohort study of neonates clinically diagnosed with HIE and treated with TH recruited from the tertiary neonatal intensive care units (NICUs) of Boston Children's Hospital, Brigham and Women's Hospital, and Beth Israel Deaconess Medical Center between December 2015 and October 2019 with follow-up to 18 months. In total, 329 neonates ≥34 weeks gestational age admitted with perinatal asphyxia and suspected HIE were identified. 179 were approached, 103 enrolled, 73 received TH, and 64 were included. CMRO2 was measured at the NICU bedside by frequency-domain near-infrared and diffuse correlation spectroscopies (FDNIRS-DCS) during the late phases of hypothermia (C), rewarming (RW) and after return to normothermia (NT). Additional variables were body temperature and clinical neonatal encephalopathy (NE) scores, as well as findings from magnetic resonance imaging (MRI) and spectroscopy (MRS). Primary outcome was the Bayley Scales of Infant and Toddler Development, Third Edition (BSID-III) at 18 months, normed (SD) to 100 (15).

FINDINGS: Data quality for 58 neonates was sufficient for analysis. CMRO2 changed by 14.4% per °C (95% CI, 14.2-14.6) relative to its baseline at NT while cerebral tissue oxygen extraction fraction (cFTOE) changed by only 2.2% per °C (95% CI, 2.1-2.4) for net changes from C to NT of 91% and 8%, respectively. Follow-up data for 2 were incomplete, 33 declined and 1 died, leaving 22 participants (mean [SD] postnatal age, 19.1 [1.2] month; 11 female) with mild to moderate HIE (median [IQR] NE score, 4 [3-6]) and 21 (95%) with BSID-III scores >85 at 18 months. CMRO2 at NT was positively associated with cognitive and motor composite scores (β (SE) = 4.49 (1.55) and 2.77 (1.00) BSID-III points per 10-10 moL/dl × mm2/s, P = 0.009 and P = 0.01 respectively; linear regression); none of the other measures were associated with the neurodevelopmental outcomes.

INTERPRETATION: Point of care measures of CMRO2 in the NICU during C and RW showed dramatic changes and potential to assess individual response to TH. CMRO2 following TH outperformed conventional clinical evaluations (NE score, cFTOE, and MRI/MRS) at predicting cognitive and motor outcomes at 18 months for mild to moderate HIE, providing a promising objective, physiologically-based diagnostic for HIE.

FUNDING: This clinical study was funded by an NIH grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, United States (R01HD076258).

IMPORTANCE: Neurodevelopmental disabilities are commonly associated with congenital heart disease (CHD), but medical and sociodemographic factors explain only one-third of the variance in outcomes.

OBJECTIVE: To examine whether potentially damaging de novo variants (dDNVs) in genes not previously linked to neurodevelopmental disability are associated with neurologic outcomes in CHD and, post hoc, whether some dDNVs or rare putative loss-of-function variants (pLOFs) in specific gene categories are associated with outcomes.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study was conducted from September 2017 to June 2020 in 8 US centers. Inclusion criteria were CHD, age 8 years or older, and available exome sequencing data. Individuals with pathogenic gene variants in known CHD- or neurodevelopment-related genes were excluded. Cases and controls were frequency-matched for CHD class, age group, and sex.

EXPOSURES: Heterozygous for (cases) or lacking (controls) dDNVs in genes not previously associated with neurodevelopmental disability. Participants were separately stratified as heterozygous or not heterozygous for dDNVs and/or pLOFs in 4 gene categories: chromatin modifying, constrained, high level of brain expression, and neurodevelopmental risk.

MAIN OUTCOMES AND MEASURES: Main outcomes were neurodevelopmental assessments of academic achievement, intelligence, fine motor skills, executive function, attention, memory, social cognition, language, adaptive functioning, and anxiety and depression, as well as 7 structural, diffusion, and functional brain magnetic resonance imaging metrics.

RESULTS: The study cohort included 221 participants in the post hoc analysis and 219 in the case-control analysis (109 cases [49.8%] and 110 controls [50.2%]). Of those 219 participants (median age, 15.0 years [IQR, 10.0-21.2 years]), 120 (54.8%) were male. Cases and controls had similar primary outcomes (reading composite, spelling, and math computation on the Wide Range Achievement Test, Fourth Edition) and secondary outcomes. dDNVs and/or pLOFs in chromatin-modifying genes were associated with lower mean (SD) verbal comprehension index scores (91.4 [20.4] vs 103.4 [17.8]; P = .01), Social Responsiveness Scale, Second Edition, scores (57.3 [17.2] vs 49.4 [11.2]; P = .03), and Wechsler Adult Intelligence Scale, Fourth Edition, working memory scores (73.8 [16.4] vs 97.2 [15.7]; P = .03), as well as higher likelihood of autism spectrum disorder (28.6% vs 5.2%; P = .01). dDNVs and/or pLOFs in constrained genes were associated with lower mean (SD) scores on the Wide Range Assessment of Memory and Learning, Second Edition (immediate story memory: 9.7 [3.7] vs 10.7 [3.0]; P = .03; immediate picture memory: 7.8 [3.1] vs 9.0 [2.9]; P = .008). Adults with dDNVs and/or pLOFs in genes with a high level of brain expression had greater Conners adult attention-deficit hyperactivity disorder rating scale scores (mean [SD], 55.5 [15.4] vs 46.6 [12.3]; P = .007).

CONCLUSIONS AND RELEVANCE: The study findings suggest neurodevelopmental outcomes are not associated with dDNVs as a group but may be worse in individuals with dDNVs and/or pLOFs in some gene sets, such as chromatin-modifying genes. Future studies should confirm the importance of specific gene variants to brain function and structure.

Congenital heart disease (CHD) and prematurity are leading causes of infant mortality in the United States. Infants with CHD born prematurely are often described as facing "double jeopardy" with vulnerability from their underlying heart disease and from organ immaturity. They endure additional complications of developing in the extrauterine environment while healing from interventions for heart disease. While morbidity and mortality for neonates with CHD have declined over the past decade, preterm neonates with CHD remain at higher risk for adverse outcomes. Less is known about their neurodevelopmental and functional outcomes. In this perspective paper, we review the prevalence of preterm birth among infants with CHD, highlight the medical complexity of these infants, and emphasize the importance of exploring outcomes beyond survival. We focus on current knowledge regarding overlaps in the mechanisms of neurodevelopmental impairment associated with CHD and prematurity and discuss future directions for improving neurodevelopmental outcomes.

BACKGROUND: Known genetic causes of congenital heart disease (CHD) explain <40% of CHD cases, and interpreting the clinical significance of variants with uncertain functional impact remains challenging. We aim to improve diagnostic classification of variants in patients with CHD by assessing the impact of noncanonical splice region variants on RNA splicing.

METHODS: We tested de novo variants from trio studies of 2649 CHD probands and their parents, as well as rare (allele frequency, <2×10-6) variants from 4472 CHD probands in the Pediatric Cardiac Genetics Consortium through a combined computational and in vitro approach.

RESULTS: We identified 53 de novo and 74 rare variants in CHD cases that alter splicing and thus are loss of function. Of these, 77 variants are in known dominant, recessive, and candidate CHD genes, including KMT2D and RBFOX2. In 1 case, we confirmed the variant's predicted impact on RNA splicing in RNA transcripts from the proband's cardiac tissue. Two probands were found to have 2 loss-of-function variants for recessive CHD genes HECTD1 and DYNC2H1. In addition, SpliceAI-a predictive algorithm for altered RNA splicing-has a positive predictive value of ≈93% in our cohort.

CONCLUSIONS: Through assessment of RNA splicing, we identified a new loss-of-function variant within a CHD gene in 78 probands, of whom 69 (1.5%; n=4472) did not have a previously established genetic explanation for CHD. Identification of splice-altering variants improves diagnostic classification and genetic diagnoses for CHD.

REGISTRATION: URL: https://clinicaltrials.gov; Unique identifier: NCT01196182.

Advances in prenatal/neonatal genetic screening practices and next generation sequencing (NGS) technologies have made the detection of molecular causes of pediatric diseases increasingly more affordable, accessible, and rapid in return of results. In the past, families searching for answers often required diagnostic journeys leading to delays in targeted care and missed diagnoses. Non-invasive prenatal NGS is now used routinely in pregnancy, significantly altering the obstetric approach to early screening and evaluation of fetal anomalies. Similarly, exome sequencing (ES) and genome sequencing (GS) were once only available for research but are now used in patient care, impacting neonatal care and the field of neonatology as a whole. In this review we will summarize the growing body of literature on the role of ES/GS in prenatal/neonatal care, specifically in neonatal intensive care units (NICU), and the molecular diagnostic yield. Furthermore, we will discuss the impact of advances in genetic testing in prenatal/neonatal care and discuss challenges faced by clinicians and families. Clinical application of NGS has come with many challenges in counseling families on interpretation of diagnostic results and incidental findings, as well as re-interpretation of prior genetic test results. How genetic results may influence medical decision-making is highly nuanced and needs further study. The ethics of parental consent and disclosure of genetic conditions with limited therapeutic options continue to be debated in the medical genetics community. While these questions remain unanswered, the benefits of a standardized approach to genetic testing in the NICU will be highlighted by two case vignettes.

Congenital heart disease (CHD) is the most common birth anomaly, affecting almost 1% of infants. Neurodevelopmental delay is the most common extracardiac feature in people with CHD. Many factors may contribute to neurodevelopmental risk, including genetic factors, CHD physiology, and the prenatal/postnatal environment. Damaging variants are most highly enriched among individuals with extracardiac anomalies or neurodevelopmental delay in addition to CHD, indicating that genetic factors have an impact beyond cardiac tissues in people with CHD. Potential sources of genetic risk include large deletions or duplications that affect multiple genes, such as 22q11 deletion syndrome, single genes that alter both heart and brain development, such as CHD7, and common variants that affect neurodevelopmental resiliency, such as APOE. Increased use of genome-sequencing technologies in studies of neurodevelopmental outcomes in people with CHD will improve our ability to detect relevant genes and variants. Ultimately, such knowledge can lead to improved and more timely intervention of learning support for affected children.