Rapid Targeted Genomics in Critically Ill Newborns

Cleo C. van Diemen, PhD,a Wilhelmina S. Kerstjens-Frederikse, MD, PhD,a Klasien A. Bergman, MD, PhD,b Tom J. Rapidde Koning, MD, PhD, Targeteda, b Birgit Sikkema-Raddatz, PhD, Genomicsa Joeri K. van der Velde, MSc, ina Kristin M. Abbott, PhD,a Johanna C. Herkert, MD,a Katharina Löhner, MD,a Patrick Rump, MD, PhD,a Martine T. Meems-Veldhuis, BSc,a Pieter B.T. Neerincx, PhD,a Jan D.H. Jongbloed, PhD,a Conny M. van Ravenswaaij-Arts, MD, PhD,a Morris CriticallyA. Swertz, PhD,a Richard J. Sinke, Ill PhD, Newbornsa Irene M. van Langen, MD, PhD,a Cisca Wijmenga, PhDa

BACKGROUND: abstract

Rapid diagnostic whole-genome sequencing has been explored in critically ill newborns, hoping to improve their clinical care and replace time-consuming and/or invasive diagnostic testing. A previous retrospective study in a research setting showed promising results with diagnoses in 57%, but patients were highly selected for known and likely Mendelian disorders. The aim of our prospective study was to assess the speed and METHODS: yield of rapid targeted genomic diagnostics for clinical application. We included 23 critically ill children younger than 12 months in ICUs over a period of 2 years. A quick diagnosis could not be made after routine clinical evaluation and diagnostics. Targeted analysis of 3426 known disease genes was performed by using whole- genome sequencing data. We measured diagnostic yield, turnaround times, and clinical RESULTS: consequences.

A genetic EPG5diagnosis was obtained in 7 patientsRMND1 (30%), with a median turnaround time of 12 days (ranging from 5 to 23 days).EIF2B5 We identified compound heterozygous mutations in the gene (ViciKLHL41 syndrome), the gene (combinedGFER oxidative phosphorylation deficiency-11), and the geneGLB1 (vanishing white matter), and homozygous mutations in the gene (nemaline myopathy), the gene (progressive mitochondrial myopathy), and the gene (GM1-gangliosidosis). In CONCLUSIONS: addition, a 1p36.33p36.32 microdeletion was detected in a child with cardiomyopathy. Rapid targeted genomics combined with copy number variant detection adds important value in the neonatal and pediatric intensive care setting. It led to a fast diagnosis in 30% of critically ill children for whom the routine clinical workup was unsuccessful.

aDepartment of Genetics, University of Groningen; and bBeatrix Children’s Hospital, University Medical Center What’s Known on This Subject: Clinical decision- Groningen, Groningen, Netherlands making in critically ill newborns is challenging. Dr van Diemen conceptualized and designed the study, coordinated the study overall, analyzed Whole-genome sequencing offers the possibility to and interpreted data, codrafted the initial manuscript, and revised the manuscript; Dr Kerstjens- simultaneously test all known disease genes to aid in Frederikse developed the logistics for data collection, collected and interpreted data, codrafted clinical decision-making but has not been tested in a the initial manuscript, and revised the manuscript; Drs de Koning and Bergman designed the clinical prospective study. study, collected and interpreted data, and critically reviewed the manuscript; Profs Swertz, Sinke, van Ravenswaaij-Arts, and van Langen conceptualized and designed the study, interpreted What This Study Adds: This prospective study data, and critically reviewed the manuscript; Prof Wijmenga initiated, conceptualized, and shows that rapid targeted genomics combined designed the study, interpreted data, and critically reviewed the manuscript; Drs Sikkema- with copy number variant detection increases Raddatz and Jongbloed developed laboratory and administrative logistics, interpreted data, and the diagnostic yield in the neonatal and pediatric critically reviewed the manuscript; Mrs Meems-Veldhuis developed laboratory, administrative, intensive care setting and has a great impact on and analytical logistics, performed laboratory work, and analyzed and interpreted data; Drs clinical decision-making. Herkert, Rump, and Löhner acquired clinical data, interpreted data, and critically reviewed the To cite: van Diemen CC, Kerstjens-Frederikse WS, Bergman KA, et al. Rapid Targeted Genomics in Critically Ill Newborns. Pediatrics. 2017;140(4):e20162854

Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017:e20162854 Article

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF Diagnosing a genetic disease on of the NICU, but these were3,4 all based muscular atrophy, cystic fibrosis, the basis of clinical presentation on a 4retrospective design. ‍ Willig etc) or structural variations (such as in critically ill newborns and small etn al reported a method in which trisomy 21 or microdeletion 22q11). infants can be extremely challenging almost all Mendelian disease genes The decision to include a patient was because the symptoms and features ( = 4300) were tested by rapid made by a multidisciplinary working of known genetic syndromes may WGS (STATseq) within 50 hours. group comprising pediatricians, not be present at birth, may change They were able to diagnose 20 of 35 clinical geneticists, technicians, rapidly, or be difficult to observe infants (57%) with a median time to bioinformaticians, and laboratory in a small child on life support. provisional diagnosis of 23 days. specialists. Patients did not have Moreover, the standard genetic Here, we present the results of a exome sequencing or any positive diagnostic workup of sequential prospective pilot study in which we result from genetic testing before testing of disease genes considered aimed to implement rapid genomic inclusion, but all regular genetic in the preliminary diagnosis is diagnostics by performing WGS and other investigations were time consuming. Because diseases combined with filtering on a gene performed in parallel, with results can progress rapidly, warranting panel of 3426 genes in a clinical later than those of the rapid genetic clinical intervention, it is of the setting for critically ill newborns diagnostics. See Supplemental Table utmost importance to diagnose these and infants. We included patients 3 for all genetic tests performed in children as soon as possible to enable suspected of having a genetic disease regular diagnostics. For all patients, timely interventions that reduce but excluded those with a clear we followed the procedure outlined morbidity, suffering, and mortality clinical diagnosis for which a single in Fig 2. Rapid targeted genomics and avoid pointless and expensive targeted test or gene panel was was performed in all patients in intensive care. available. We aimed to provide a accordance with the regulations genetic diagnosis within 2 weeks. and ethical guidelines of the UMCG To date, there are >4400 genetic Methods (UMCG Medical Ethics Committee diseases with known causes that, Primaryapproval E numbernd Points 2014092). collectively, explain the majority Selection of Patients

of infant mortality,1 particularly in NICUs and PICUs. Of these 4400 We measured diagnostic yield, genes, mutations in some 3300 have We studied 23 critically ill children turnaround times, and clinical clinical consequences as reported admitted to the NICU and/or PICU consequences of a rapid genetic- by the Clinical Genomic Database in the University Medical Center 2 diagnostic approach using WGS. (CGD). In the Netherlands, regular Groningen (UMCG) (Groningen, The turnaround time included the genetic diagnostics for these children Netherlands) over a 2-year period moment of inclusion of the patient include molecular and cytogenetic and performed rapid targeted in the study, DNA isolation, data testing to identify larger, structural genomics aimed at reaching a generation, data analysis, and data chromosomal variations, such as diagnosis (see Fig 1). Critically ill interpretation until provisional trisomies and microdeletions, as well was defined as cardiorespiratory genetic diagnosis. as single-gene and gene-panel testing insufficiency needing ventilator Counseling and Consent in the case of suspected monogenic support (16 of 23) or organ diseases. Turnaround times for these dysfunction (the brain, heart, diagnostic procedures range from 1 lungs, liver, or kidneys), which Parents of patients were counseled week to more than 1 year, especially was predicted to be a high risk for by the clinical geneticist before if multiple, consecutive tests are cardiorespiratory insufficiency in and after the rapid targeted- needed. However, there is an urgent the near future (7 of 23). Criteria genomics test. Informed consent need to speed up this process for inclusion were age <1 year at covered reporting on the diagnostic

in critically ill children. Whole- presentation and the presence of 1 results for some 3300 known 2 genome sequencing (WGS) offers or more congenital anomalies and/ disease genes based on the CGD this possibility by simultaneously or severe neurologic symptoms, such and included the option to use testing for the presence of mutations as intractable seizures, suggestive full genome sequencing data in all known disease genes and of a genetic cause of the disease. for analysis after the window of for small, numerical chromosomal Exclusion criteria were clear rapid targeted diagnostics. The variants. Proof-of-concept studies indications for a specific syndrome consent also stated the possibility have already shown the usefulness that could be tested by targeted of detecting incidental findings of WGS in diagnosing suspected analysis of known genes (such that would be communicated to genetic diseases in the acute setting as epidermolysis bullosa, spinal the family, although we minimized Downloaded from www.aappublications.org/news by guest on September 28, 2021 2 van Diemen et al

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF DNA Isolation and Sequencing

Blood samples from the patient and both parents were collected for DNA isolation. At the start of the study, we chose WGS instead of whole-exome sequencing to avoid time-consuming capturing steps. DNA from the patient was prepared for WGS according to the procedure described in the Supplemental Information. Because a ’ WGS trio analysis was too expensive, the parents DNA samples were only used for confirmatory Sanger sequencing of candidate variants and segregation analysis. For quality assurance, 80% of the CGD-based FIGURE 1 Overview of inclusion criteria for patients, diagnostic yield, and time to diagnosis. Patients without gene panel should be covered at an obvious diagnosis often need sequential, time-consuming genetic tests. Critically ill children were least 20 times, otherwise, additional included in a pilot study to see if rapid WGS led to a diagnosis. Dsequencingata Analysis data were produced.

Raw WGS data from the patients were processed according to –

standardized protocols as described5 8 in the Supplemental Information. ‍ ‍ Sequence variants were filtered by – using Cartagenia Next-Generation Sequencing Bench Laboratory software (Agilent, Santa Clara, CA) by using an automated filtering tree. We generated a virtual gene panel of monogenic diseases based on 3426 genes from the CGD and removed

the genes2, associated9 with late-onset diseases. ‍ We further supplemented the gene panel with genes that were on a standard, clinical exome-capturing panel (SureSelect Inherited Diseases; Agilent, Santa FIGURE 2 Clara, CA) that was used in our Schematic overview of the study set-up and time required for each part of the procedure. genome diagnostics laboratory but not included in the CGD. A full gene ’ list can be found in Supplemental Table 4. We refer to this gene panel the detection of such findings by not attributing to the patient s as the CGD-based gene panel in the excluding late-onset disease genes. current phenotype with preventive remaining text. Only genes included An independent review board options for the health of the in the CGD-based gene panel were was set up comprising a patient patient and/or family (carrier assessed. organization representative, status for autosomal recessive a health care lawyer, and a diseases was not reported). When We analyzed variants in the CGD- medical ethics specialist to discuss potential actionability was not based gene panel using Human incidental findings, which were obvious, the review board was Phenotype Ontology (HPO) terms predefined as being classified as consulted to assist weighing and minor allele frequencies from likely pathogenic or pathogenic the arguments for and against 5 databases (1000 Genomes, mutations in known disease genes disclosure. Genome of the Netherlands, Exome Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017 3

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF Sequencing Project 6500, Exome were in the NICU or PICU). From 125 overview of the results). Six out Aggregation Consortium, and complex patients, we considered 30 of 7 diagnoses were made within the database of single nucleotide infants for inclusion in the study. We the CGD-based gene panel; we polymorphisms) for filtering and excluded the following 6 children identified compound heterozygous performed subsequent annotation after discussion: 3 because they mutationsEPG5 in the ectopic P-granules with Online Mendelian Inheritance were not critically ill (2 children autophagy protein 5 homolog in Man terms, Combined Annotation with multiple congenital anomalies gene ( ) (Vici syndrome, Dependent Depletion– scores, and and 1 with neonatal cholestasis) presenting with microcephaly,

reported modes of10 inheritance19 and 3 more because we did not seizures, and developmental using MOLGENIS. ‍ The variants have a strong suspicion they had delay),RMND1 the required for meiotic remaining after these filtering a monogenic disease (1 child with nuclear division 1 homolog gene ’ steps were manually evaluated hydrops, 1 with a congenital heart ( ) (combined oxidative for matching with the patients defect and unexplained respiratory phosphorylation deficiency-11, – phenotypes in a multidisciplinary failure, and 1 with an omphalocele). presenting with microcephaly, meeting comprising at least the Over a period of 2 years (May 2013 seizures, and deafness), the ε EIF2B5 operating technician, a clinical May 2015), 24 children met our eukaryotic translation initiation geneticist, and a laboratory specialist. criteria, but in 1 case, the parents factor 2B subunit gene ( ) In this way, an average of 40 genes did not give consent for the rapid (vanishing white matter, presenting per patient were evaluated. The targeted-genomics test. This led with acute respiratory insufficiency remaining variants and genes to the inclusion of 23 children in and leukoencephalopathy), the were then classified according to the study (see Fig 1). For 22 of the homozygousKLHL41 mutations in the standardized 20guidelines based on patients, we also obtained DNA from kelch-like family member 41 gene Richards et al by using Alamut both parents; for 1 patient, only the ( ) (nemaline myopathy, software, taking into account mother was available. The median presenting with severe neonatal – GFER (among others) the effect of the age at inclusion was 28 days (range contractures), the growth factor candidate variants on the protein as of 1 day 11 months). Four children ERV1-like gene ( ) (progressive predicted by scale invariant feature presented with cardiomyopathy, mitochondrial myopathy, presenting transform (SIFT), Polymorphism 5 with severe seizure disorders, 6 with neonatal respiratory distress β GLB1 Phenotyping (PolyPhen), Grantham with an abnormal muscle tone, 2 and lactic acidosis), and the score, MutationTaster, Align GVGD, with microcephaly without seizures, galactosidase gene– ( ) (GM1

and PhyloP. The resulting candidate 3 with liver failure, 1 with coma gangliosidosis presenting22 29 with genes were further evaluated in a because of leukoencephalopathy, 1 cardiomyopathy). ‍ ‍ Lastly, a larger multidisciplinary working with multiple congenital anomalies, 1p36.33p36.32 microdeletion group comprising pediatricians, and 1 with interstitial pulmonary was detected in a child with clinical geneticists, technicians, disease. Table 1 shows the clinical cardiomyopathy by using copy

bioinformaticians, and laboratory presentations and standardized number30,31 variant calling on the WGS specialists. All candidate causal phenotypes using the HPO terms data. ‍ The 7 cases are discussed

variants and any potential unsolicited of the 2123 patients included in this in detail in the Supplemental findings were validated by using study. Information.Effects of Rapid Genome Diagnostics Sanger sequencing in the patient and on Patient Management For 2 patients, we did not reach the parents. A detailed description of the target coverage criteria of 80% of variant filtering can be found in the the CGD-based gene panel covered Supplemental Information. Rapid targeted genomics had a major at least 20 times, and we therefore impact on the decision-making Results needed to generate additional in our clinical services, NICU and sequencing data, which delayed the PICU, and it led to the withdrawal of turnaround time by 2 days (1 extra unsuccessful intensive care treatment run on the sequencer). Our median We included children who were in 5 of the 7 children diagnosed in turnaround time was 12 days, with a all referred to the UMCG in the this prospective group. In addition, minimum of 5 days and a maximum Netherlands. Over the time span of concrete diagnoses and appropriate of 23. this study, the NICU admitted 932 genetic counseling had an impact on new patients, the PICU admitted 322 In summary, we identified a causal the choices parents made for future new patients <1 year old, and clinical mutation in 7 of 23 patients, and offspring. Two sets of parents who geneticists were consulted for 497 we had 1 case of an incidental had decided not to have any more children <1 year old (155 of which finding (see Table 2 for an children changed their minds after Downloaded from www.aappublications.org/news by guest on September 28, 2021 4 van Diemen et al

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF TABLE 1 Overview of 23 Patients’ Characteristics in a Prospective Study of Critically Ill Children ID M or F Age at Specific Critical Illness Clinical Features HPO Terms HPO Codes Presentation Criteria 1401 F 2 mo Cardiorespiratory Lack of spontaneous Abnormality of the nervous HP:0000707, HP:0100022 insufficiency, ventilator movements, tremors, no system, abnormality of dependent for 7 d swallowing movement 1402 F 4 mo Cardiorespiratory Dilated cardiomyopathy Dilated cardiomyopathy HP:0001644 insufficiency, ventilator dependent for 14 d 1405 F 1 d Cardiorespiratory Intractable seizures, perinatal Epileptic encephalopathy, HP:0200134, HP:0001336, HP: insufficiency, ventilator asphyxia myoclonus, hypotonia 0001290 dependent for 11 d 1406 M 10 d Cardiorespiratory Intractable seizures, Status epilepticus, myoclonus HP:0002133, HP:0001336 insufficiency, ventilator intracerebral hemorrhage dependent for 12 d 1407 M 11 mo West syndrome, inability to Microcephaly, vermis Abnormality of the nervous HP:0000707, HP:0000252 swallow hypoplasia, West syndrome, system, microcephaly developmental delay, VSD 1408 M 7 mo Cardiorespiratory Prematurity (24+6 w), IVF Acute liver failure HP:0006554 insufficiency, ventilator pregnancy, cholestasis dependent for 7 d 1501 F 5 wk Hypoventilation, high flow Hydrocephaly, intraventricular Hydrocephaly, hypotonia, HP:0000238, HP:0001290, HP: nasal cannula for 8 d hemorrhage macrocephaly 0000256 1502 F 2 wk Cardiorespiratory IUGR, seizures, Cardiomyopathy HP:0001638 insufficiency, ventilator cardiomyopathy, dependent for 10 d atrioventricular block, PDA, VSD 1504 F 9 d Multifocal seizures, Intractable seizures, dystonia Abnormality of the nervous HP:0000707, HP:0100022 encephalopathic system, abnormality of EEG, imminent movement cardiorespiratory insufficiency 1505 M 10 d Cardiorespiratory Prematurity (31+2 w), Malformation of the heart and HP:0030680, HP:0002817 insufficiency, ventilator coarctation of the great vessels, abnormality dependent for 8 d aorta, VSD, sagittal of the upper limb craniosynostosis, bilateral postaxial polydactyly, cleft palate, eventration of diaphragm, hypospadias 1506 M 1 d Cardiorespiratory Bilateral hip dislocation, Flexion contracture, myopathy HP:0001371, HP:0003198 insufficiency, ventilator flexion contractures, dependent for 7 d supernumerary nipple, cryptorchidism, consanguineous parents 1508 F 5 mo Severe hyponatremia (116 Microcephaly, dystonia, Microcephaly, dystonia, HP:0000252, HP:0001332, HP: mmol/L), imminent developmental delay, developmental 0001263, HP:0012715, HP: cardiorespiratory deafness delay, profound 0003429 insufficiency hearing impairment, hypomyelination 1509 F 4 mo Imminent respiratory Pulmonary interstitial Interstitial pulmonary disease HP:0006530 insufficiency, continuous glycogenosis, positive airway pressure consanguineous parents for 3 d 1511 F 4 wk Cardiorespiratory Prematurity (30 w), Myopathy, fatigable weakness HP:0003198, HP:0003473 insufficiency, ventilator polyhydramnion, hypotonia, dependent for >4 wk mother muscle weakness 1513 F 2 wk Severe hypoglycemia, Hypotonia, cholestasis and Hyperinsulinimia, cholestasis HP:0000842, HP:0001396 imminent (transient) hyperinsulinism cardiorespiratory insufficiency; low flow for 3 d

Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017 5

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF TABLE 1 Continued ID M or F Age at Specific Critical Illness Clinical Features HPO Terms HPO Codes Presentation Criteria 1514 F 2 mo Frequent apnea, imminent Bulbar weakness and/or Bulbar palsy, abnormal HP:0001283, HP:0003808, HP: respiratory insufficiency palsy, abnormal muscle muscle tone, central 0011398, HP:0000364, HP: tone, central hypotonia, hypotonia, hearing 0003234, HP:0011972 mixed hearing loss abnormality, decreased plasma carnitine, hypoglycorrhachia 1515 M 2 wk Cardiorespiratory Cardiomyopathy Cardiomyopathy, hypertrophic HP:0001638, HP:0001639, HP: insufficiency, ventilator cardiomyopathy, 0006698 dependent for >4 wk ventricular aneurysm 1519 F 5 d Cardiorespiratory Persistent lactic acidosis, Abnormality of metabolism HP:0001939, HP:0012252 insufficiency, ventilator dysmorphic features and/or homeostasis, dependent for 25 d abnormal respiratory system morphology 1601 M 9 mo Cardiorespiratory Sudden seizures Leukoencephalopathy, HP:0002352, HP:0011400 insufficiency, ventilator abnormal CNS myelination dependent for 9 d 1602 M 9 wk Cardiorespiratory Growth delay, cholestasis, Growth delay, cholestasis, HP:0001510, HP:0001396, HP: insufficiency, ventilator VSDs ventricular septal defects 0001629 dependent for >4 wk 1603 M 4 d Cardiorespiratory IUGR, microcephaly, IUGR, seizures, microcephaly HP:0001511, HP:0001250, HP: insufficiency, ventilator epilepsy, encephalopathy, 0000252 dependent for 13 d contractures 1604 M 2 mo Cardiorespiratory Abnormal muscle tone, Abnormal muscle tone, HP:0003808, HP:0001324, HP: insufficiency, ventilator feeding problems muscle weakness, feeding 0011968 dependent for >4 wk problems 1605 F 3 mo Cardiorespiratory Cardiomyopathy, Cardiomyopathy, HP:0001638, HP:0002240 insufficiency, ventilator hepatomegaly, hepatomegaly dependent for 8 d consanguineous parents (first cousins) CNS, central nervous system; F, female sex; ID, identification number; IUGR, intrauterine growth retardation; IVF, in vitro fertilization; M, male sex; PDA, persistent ductus arteriosus; VSD, ventricular septal defect.

they learned that prenatal testing consent form. One couple (out of to an increased diagnostic yield, ∼ – was available. In the follow-up 24) did not agree to the informed in-patient management, and future period (ranging from 2 years 3 consent. For 1 child, we had a family planning. In our cohort of months), no prenatal tests have been suspected case of nonpaternity, ’ 23 critically ill infants, we made 7 performed yet, but presymptomatic which had to be discussed with the diagnoses in patients with a wide testing was performed for 1 couple. parents because the child s diagnosis range of clinical presentations. All In addition, preimplantation could not otherwise be confirmed. but 1 of these genetic diagnoses diagnostics was chosen by 1 couple. We had set up an independent would not have been made in our Expanded preconception screening review board for incidental findings regular molecular diagnostics by using a panel restricted to rare comprising a patient organization setting because the diseases were serious diseases was offered to all representative, a health care lawyer, not suspected on clinical grounds the consanguineous couples after our and a medical ethics specialist, but and specific genetic testing would regular procedure in clinical genetics31 there was no need to consult it. not have been considered. The mean and was accepted by 1 couple. No Discussion turnaround time of 12 days for our additional risk for these diseases was rapid targeted-genomic diagnostics Iobservedncidental in Findings this couple. suggests that invasive diagnostic Rapid targeted genomics using testing, such as muscle biopsies, can WGS has proved to be feasible in be avoided in these children in the The parents of patients were our multidisciplinary setting of future. The clinical relevance of rapid counseled about the chance of NICU, PICU, and clinical genetics genome diagnostics further lies in the incidental findings, although we department in a university hospital in fact that these results can be used in minimized the detection of such the Netherlands. More importantly, the clinical decisions made in caring findings by excluding late-onset this testing has shown it yields for critically ill children in ICUs, disease genes. This possibility was major added value in a routine in better genetic counseling of the clearly stated on the informed diagnostics setting with respect parents, and in guiding their future Downloaded from www.aappublications.org/news by guest on September 28, 2021 6 van Diemen et al

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF 22 ‍ ‍ 31 ‍ 24 ‍ 23 — — — — — — — — — Reference 10 C — — — — — — — — — — before before; before; ExA 0.0001741 0.000008287 Not reported Not reported Not reported Allele Frequency 18) a a — — — — — — — — — — Variants c.5869+1G>A p.(Gly405Ser) and c.713A>G (p.Asn238Ser) EPG5 c.7475T>G (p.Asn421Thrfs (p.Phe2492Cys) and KLHL41 c.1213G>A Causal Gene and/or RMND1 c.1262_1275del — — — — — — — — — — of Gene recessive, compound heterozygous recessive, homozygous recessive, compound heterozygous Inheritance Mode Autosomal Autosomal b No No No No No No No No No Diagnosis Provisional syndrome myopathy-9 phosphorylation deficiency-11 Vici syndrome Autosomal 1p36 microdeletion Nemaline Combined oxidative 33 × 36 × 38 × 40 × 51 × 29 × 30 × 32 × 39 × 34 × 25 × 38 × 45 × Mean CGD-Based Gene Panel Coverage of 8 13 23 20 13 11 16 13 15 13 13 10 Time, d w), IVF w), +6 +2 Clinical Features Turnaround intracerebral hemorrhage hypoplasia, West syndrome, developmental delay, VSD dystonia movements, tremors, no swallowing perinatal asphyxia pregnancy, cholestasis intraventricular hemorrhage cardiomyopathy, AV- block, PDA, VSD coarctation of the aorta, VSD, sagittal craniosynostosis, bilateral postaxial polydactyly, cleft palate, eventration of diaphragm, hypospadias flexion contractures, supernumerary nipple, cryptorchidism, consanguineous parents glycogenosis, consanguineous parents developmental delay, deafness Presentation Results of Rapid Genetic Diagnostics 23 Critically Ill Children

2 1406 M1407 M 10 d Intractable seizures, 11 mo Microcephaly, vermis 1408 M 7 mo Prematurity (24 15041505 F M 9 d 10 d Intractable seizures, Prematurity (31 1401 F14021405 F 2 mo F Lack of spontaneous 4 mo 1 d Dilated cardiomyopathy Intractable seizures, 14 1501 F1502 F 5 wk Hydrocephaly, 2 wk IUGR, seizures, 1506 M 1 d Bilateral hip dislocation, 1509 F 4 mo Pulmonary interstitial ID M or F Age at 1508 F 5 mo Microcephaly, dystonia, TABLE

Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017 7

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF ‍ 26 ‍ 28 , , 29 ‍ — — — — — — — 25 ‍ ‍ 27 Reference 10 C — — — — — — — ExA 0.000008277 0.00004141 0.00001661 0.00004118; Allele Frequency a — — — — — — — Variants (p.Arg59His) (p.Ala403Val) (p.Arg339Trp) (p.Arg339Trp) (p.Arg194Cys) and c.1208C>T GFER c.580C>T GLB1 c.176G>A EIF2B5 c.1015C>T Causal Gene and/or — — — — — — — of Gene recessive, homozygous recessive, homozygous recessive, compound heterozygous Inheritance Mode Autosomal Autosomal 30 . No No No No No No No Diagnosis Provisional progressive myopathy with vanishing white matter GM1 gangliosidosis Autosomal Mitochondrial Leukoencephalopathy 39 × 45 × 44 × 44 × 37 × 43 × 43 × 27 × 43 × 29 × Mean CGD-Based Gene Panel Coverage of 5 5 7 7 8 8 8 17 15 15 Time, d Clinical Features Turnaround hepatomegaly; consanguineous parents (first cousins) feeding problems dysmorphic features epilepsy, encephalopathy, contractures polyhydramnion, hypotonia, mother muscle weakness and (transient) hyperinsulinism palsy, abnormal muscle tone, central hypotonia, mixed hearing loss ventricular septal defects Presentation Continued

2 1605 F 3 mo Cardiomyopathy, 1604 M 2 mo Abnormal muscle tone, 1519 F 5 d Persistent lactic acidosis, 1603 M 4 d IUGR, microcephaly, 1511 F1513 4 wk F1514 Prematurity (30 w), F 2 wk1515 Hypotonia, cholestasis 2 mo M Bulbar weakness and/or 2 wk Cardiomyopathy 1601 M1602 9 mo M Sudden seizures 9 wk Growth delay, cholestasis, ID M or F Age at Diagnosis was made by routine single nucleotide polymorphism array diagnostics and by using copy number variant calling on WGS data Diagnosis was made by routine single nucleotide polymorphism array diagnostics and using All diagnoses were confirmed by Sanger sequencing. TABLE AV, atrioventricular; ExAC, Exome Aggregation Consortium; F, female sex; ID, identification number; IUGR, intrauterine growth retardation; IVF, in vitro fertilization; M, male sex; PDA, persistent ductus arteriosus; VSD, ventricular septal defect. — , not female sex; ID, identification number; IUGR, intrauterine growth retardation; IVF, atrioventricular; ExAC, Exome Aggregation Consortium; F, AV, applicable. a b

Downloaded from www.aappublications.org/news by guest on September 28, 2021 8 van Diemen et al

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF reproductive choices. In this respect, phenotypic feature and were able to child), we also conducted a the UMCG is now running a trial of diagnose her condition asRMND1. combined retrospective pilot study to analyze preconception screening to detect oxidative phosphorylation deficiency deceased newborns with a suspected autosomal recessive disease genes in because of mutations in genetic disease and their parents couples from the general population32 Rapid genetic diagnostics for using a clinical exome-capturing who wish to start a family. newborns and infants is not gene panel (SureSelect Inherited yet common practice and has Diseases; Agilent, Santa Clara, CA) During the course of our prospective ’ and the same bioinformatics analysis so far mainly been reported3,4 by study, we adjusted and optimized Kingsmore s group. ‍ The median approach as used for our rapid ’ the procedure continuously. turnaround time of our study was targeted-genomics approach. The Because we performed this study comparable to that of Kingsmore s diagnostic yield was similar, with 3 in a multidisciplinary setting of set-up, but our diagnostic yield was diagnoses out of 8 patients. There clinicians, technicians, researchers, lower (30% vs 57%). We think this was no need for Sanger sequencing bioinformaticians, and laboratory difference can mainly be explained to confirm cases of (compound) specialists, the difficult aspects of by our strict inclusion criteria (we recessive disease, and it was easier the procedure came to light quickly investigated only those patients who to pick up de novo mutations during our weekly discussions, and had no clear syndrome diagnosis), (although confirmatory Sanger this led to rapid implementation and we consider this as a strength sequencing remained necessary). of improvements that varied from of our study. For example, we This may prove to be an alternative, modifications in the logistics to excluded patients referred to us with more cost-effective strategy in the dealing with outdated databases. For suspected coloboma, congenital heart future. Future research should also the last 7 patients included in the focus on the cost-effectiveness of ∼ disease, choanal atresia, mental and study, we were able to reduce the growth retardation, genital anomalies genetic diagnostics on the basis of turnaround time from 3 weeks to and ear malformations and hearing targeted genomics in a prospective a maximum of 8 days. A significant loss otherwise known as CHARGE study, including all critically ill impact on our turnaround time arose syndrome and epidermolysis bullosa infants in ICUs. Ideally, such a study should incorporate Next-Generation from the acquisition of 2 new Next- because they could be diagnosed – Generation Sequencing machines using straightforward, regular Sequencing of either exomes or (Illumina NextSeq500) to replace the diagnostic testing. Our results and whole genomes on patient parent Illumina HiSeq2500; these proved the number of diagnoses we achieved trios. Although our current analysis to be more stable than our Illumina illustrate the power of using WGS method is based on sequencing only HiSeq2500, and because we acquired diagnostic testing in practice. Other the index patient, new technologic advances will lower the cost of 2 machines, we had sequencing groups have reported on using – capacity readily available because exome sequencing as a diagnostic sequencing, and we expect WGS on of the redundancy. Finally, one of tool, and they had comparable yields patient parent trios to become the the most crucial determinants of – standard in the near future. of detecting– mutations in known the diagnostic yield proved to be disease genes33 37 (in 25% 30% of their the choice of HPO terms used for patients). ‍ ‍ However, inclusion We are now including more patients filtering. For example, we initially criteria and turnaround times were in our study and following up on missed the diagnosis for patient 1506 all unsolved patients in a research “ ” different from our study. with nemaline deficiency becauseKLHL41 setting, which includes analyzing we used the HPO term myopathy, Our diagnoses were considered to be their full genomes. We are testing which does not include the provisional because we decided to additional copy number variant gene for nemaline deficiency. When confirm them by Sanger sequencing. analyses and a gene-prioritization “ we relaxed the HPO term to the This typically took 1 week. However, method based on gene coexpression ” they were all confirmed. In the KLHL41broader term abnormality of the networks. This has already resulted musculature, the mutations in future, we may relax our mandatory in the diagnosis of 1 patient were readily identified. It confirmation by Sanger sequencing (included after the initial cohort because of the good reproducibility further proved important to closely described here) with congenitalRAPSN monitor and follow-up the phenotype of high-quality sequencing results. myasthenic syndrome caused by of patients. Initially, we had no This would improve the turnaround compound heterozygous diagnosis for patient 1508, but once time and offer a faster opportunity mutations (1 deletion of exon 8 it became evident that she had also for clinical intervention. not detected by the initial analysis developed hearing disabilities, we To assess the potential of targeted and 1 known pathogenic missense reran the analysis including this genomics on trios (parents and mutation c.264C>A, p.Asn88Lys), Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017 9

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF which offered the opportunity for otherwise require regular, sequential bioinformatics pipeline; and Jackie targeted therapeutic intervention. diagnostics lasting 6 months or more. SeniorAbbr eviationsfor editing the manuscript. Furthermore, we are evaluating Rapid genome diagnostics raises the use of parallel RNA sequencing possibilities to adjust the treatment alongside targeted genomics to of critically ill children and perform CGD: Clinical Genomic Database help pinpoint candidate genes presymptomatic and prenatal testing. EIF2B5: eukaryotic translation and mutations by assessing gene Acknowledgments ε initiation factor 2B differential expression, allele-specific subunit expression, and by analyzing the EPG5: ectopic P-granules effect of mutations on splicing. We thank Kim de Lange, Martijn Viel, autophagy protein 5 Conclusions Arjen J. Scheper, and Jos Dijkhuis for homolog technical laboratory work; Yvonne β GFER: growth factor ERV1-like J. Vos and Annemieke H. van der GLB1: galactosidase 1 Rapid targeted genomics using WGS Hout for variant interpretation; Erica HPO: Human Phenotype has proven to be feasible and fast Gerkes and Rolf H. Sijmons for clinical Ontology in our multidisciplinary setting, phenotyping and interpretation; KLHL41: kelch-like family and the results add major value Roan Kanninga, Gerben de Vries, member 41 to the clinical decisions made in Lennart Johansson, Elisa Hoekstra, RMND1: required for meiotic the care of critically ill children. and Marloes Benjamins for data nuclear division 1 Adapting a targeted genomics- analysis; the MOLGENIS team (Bart homolog first approach enables genetic Charbon, Mark de Haan, Erik Winder, UMCG: University Medical Center diagnoses to be reached within Dennis Hendriksen, Fleur Kelpin, – Groningen a median time of 12 days (range Jonathan Jetten, Tommy de Boer, WGS: whole-genome sequencing of 5 23 days) in cases that would and Chao Pang) for developing the

manuscript; Drs Neerincx and van der Velde developed computational pipelines for data analysis, interpreted data, and critically reviewed the manuscript; and all authors approved the final manuscript as submitted. DOI: https://doi.org/10.1542/peds.2016-2854 Accepted for publication Jul 10, 2017 Address correspondence to Cleo C. van Diemen, PhD, Department of Genetics, University Medical Center Groningen, PO Box 30.001, 9700 RB Groningen, Netherlands. E-mail: [email protected] PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). Copyright © 2017 by the American Academy of Pediatrics FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose. FUNDING: No external funding. POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

References 1. Online Mendelian Inheritance in 4. Willig LK, Petrikin JE, Smith LD, et Bioinformatics. 2015;31(12): Man (OMIM). OMIM Entry Statistics. al. Whole-genome sequencing for 2032–2034 Available at: www.omim.org/statistics/ identification of Mendelian disorders 7. McKenna A, Hanna M, Banks E, et entry. Accessed July 22, 2016 in critically ill infants: a retrospective al. The genome analysis toolkit: a 2. Solomon BD, Nguyen AD, Bear KA, analysis of diagnostic and clinical MapReduce framework for analyzing Wolfsberg TG. Clinical genomic findings. Lancet Respir Med. next-generation DNA sequencing database. Proc Natl Acad Sci USA. 2015;3(5):377–387 data. Genome Res. 2010;20(9): 2013;110(24):9851–9855 5. Li H, Durbin R. Fast and accurate 1297–1303 3. Saunders CJ, Miller NA, Soden SE, short read alignment with Burrows- 8. Van der Auwera GA, Carneiro MO, et al. Rapid whole-genome sequencing Wheeler transform. Bioinformatics. Hartl C, et al. From FastQ data to for genetic disease diagnosis in 2009;25(14):1754–1760 high confidence variant calls: the neonatal intensive care units. 6. Tarasov A, Vilella AJ, Cuppen E, genome analysis toolkit best practices Sci Transl Med. 2012;4(154): Nijman IJ, Prins P. Sambamba: fast pipeline. Curr Protoc Bioinformatics. 154ra135 processing of NGS alignment formats. 2013;43:11.10.1–11.10.33

Downloaded from www.aappublications.org/news by guest on September 28, 2021 10 van Diemen et al

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF 9. National Human Genome Research Molecular Pathology. Genet Med. characterization of 14 GLB1 mutant Institute. Clinical genomic database. 2015;17(5):405–424 alleles found in GM1-gangliosidosis Available at: http://research.nhgri.nih. and Morquio B patients. J Lipid Res. 21. K hler S, Doelken SC, Mungall CJ, gov/CGD/. Accessed October 20, 2014 ö 2007;48(10):2275 2282 et al. The human phenotype ontology – 10. ExAC Browser (Beta). Version 0.3. project: linking molecular biology 30. Johansson LF, van Dijk F, de Boer EN, Cambridge, MA: Exome Aggregation and disease through phenotype data. et al. CoNVaDING: single exon variation Consortium; 2015. Nucleic Acids Res. 2014;42(Database detection in targeted NGS data. Hum 11. NHLBI GO Exome Sequencing Project issue):D966–D974 Mutat. 2016;37(5):457–464 (ESP). Exome Variant Server. Available 22. Cullup T, Kho AL, Dionisi-Vici C, et al. 31. Battaglia A, Hoyme HE, Dallapiccola B, at: http://evs.gs.washington.edu/EVS/ Recessive mutations in EPG5 cause et al. Further delineation of deletion 12. Abecasis GR, Auton A, Brooks LD, et al; Vici syndrome, a multisystem disorder 1p36 syndrome in 60 patients: a 1000 Genomes Project Consortium. An with defective autophagy. Nat Genet. recognizable phenotype and common integrated map of genetic variation 2013;45(1):83–87 cause of developmental delay and from 1,092 human genomes. Nature. mental retardation [published 23. Gupta VA, Ravenscroft G, Shaheen R, correction appears in Pediatrics. 2012;491(7422):56–65 et al. Identification of KLHL41 mutations 2008;121(5):1081]. Pediatrics. 13. NCBI. dbSNP Short Genetic Variations. implicates BTB-kelch-mediated 2008;121(2):404–410 Available at: www.ncbi.nlm.nih.gov/ ubiquitination as an alternate SNP/ pathway to myofibrillar disruption in 32. UMCG. Preconception screening trial in nemaline myopathy. Am J Hum Genet. general population. Available at: www. 14. Genome of the Netherlands 2013;93(6):1108–1117 rug.nl/research/genetics/research/ Consortium. Whole-genome pcs-pilot- study. Accessed July 22, sequence variation, population 24. Janer A, Antonicka H, Lalonde E, 2016 structure and demographic history et al. An RMND1 Mutation causes of the Dutch population. Nat Genet. encephalopathy associated with 33. Sawyer SL, Hartley T, Dyment DA, et al; FORGE Canada Consortium; Care4Rare 2014;46(8):818–825 multiple oxidative phosphorylation complex deficiencies and a Canada Consortium. Utility of whole- 15. Kircher M, Witten DM, Jain P, O’Roak mitochondrial translation defect. Am J exome sequencing for those near the BJ, Cooper GM, Shendure J. A general end of the diagnostic odyssey: time Hum Genet. 2012;91(4):737–743 framework for estimating the relative to address gaps in care. Clin Genet. pathogenicity of human genetic 25. Taylor RW, Pyle A, Griffin H, et al. 2016;89(3):275–284 variants. Nat Genet. 2014;46(3):310–315 Use of whole-exome sequencing to determine the genetic basis of 34. Monroe GR, Frederix GW, Savelberg 16. Swer tz MA, Dijkstra M, Adamusiak SM, et al. Effectiveness of whole-exome T, et al. The MOLGENIS toolkit: rapid multiple mitochondrial respiratory chain complex deficiencies. JAMA. sequencing and costs of the traditional prototyping of biosoftware at the diagnostic trajectory in children with 2014;312(1):68–77 push of a button. BMC Bioinformatics. intellectual disability. Genet Med. 2010;11(suppl 12):S12 26. Di Fonzo A, Ronchi D, Lodi T, et al. The 2016;18(9):949–956 mitochondrial disulfide relay system 17. MOLGENIS. Version 1.4.0. Groningen, 35. Valencia CA, Husami A, Holle J, et al. protein GFER is mutated in autosomal- Netherlands: University Medical Center Clinical impact and cost-effectiveness recessive myopathy with cataract and Groningen; 2015. of whole exome sequencing as a combined respiratory-chain deficiency. 18. Hudson-Alpha Institute for diagnostic tool: a pediatric center s Am J Hum Genet. 2009;84(5):594 604 ’ Biotechnology. Combined Annotation – experience. Front Pediatr. 2015;3:67 27. Leegwater PAJ, Vermeulen G, Dependent Depletion (CADD) scores. 36. Stark Z, Tan TY, Chong B, et al; K nst AAM, et al. Subunits of the Version 1.0. Available at: http://cadd. gs. ö Melbourne Genomics Health Alliance. A translation initiation factor eIF2B are washington.edu/download. Accessed prospective evaluation of whole-exome mutant in leukoencephalopathy with April 1, 2014 sequencing as a first-tier molecular vanishing white matter. Nat Genet. 19. Online Mendelian Inheritance in Man test in infants with suspected 2001;29(4):383 388 (OMIM). Morbid Map. 2014. – monogenic disorders. Genet Med. 20. Richards S, Aziz N, Bale S, et al; 28. Liu R, van der Lei HD, Wang X, et al. 2016;18(11):1090–1096 ACMG Laboratory Quality Assurance Severity of vanishing white matter 37. Lazaridis KN, Schahl KA, Cousin MA, Committee. Standards and guidelines disease does not correlate with et al; Individualized Medicine Clinic for the interpretation of sequence deficits in eIF2B activity or the integrity Members. Outcome of whole exome variants: a joint consensus of eIF2B complexes. Hum Mutat. sequencing for diagnostic Odyssey recommendation of the American 2011;32(9):1036–1045 cases of an individualized medicine College of Medical Genetics and 29. Santamaria R, Chabás A, Callahan JW, clinic: the Mayo Clinic experience. Genomics and the Association for Grinberg D, Vilageliu L. Expression and Mayo Clin Proc. 2016;91(3):297–307

Downloaded from www.aappublications.org/news by guest on September 28, 2021 PEDIATRICS Volume 140, number 4, October 2017 11

van Diemen et al https://doi.org/10.1542/peds.2016-2854 October 2017 Rapid Targeted Genomics in Critically Ill Newborns 4 140 Pediatrics 2017 ROUGH GALLEY PROOF Rapid Targeted Genomics in Critically Ill Newborns Cleo C. van Diemen, Wilhelmina S. Kerstjens-Frederikse, Klasien A. Bergman, Tom J. de Koning, Birgit Sikkema-Raddatz, Joeri K. van der Velde, Kristin M. Abbott, Johanna C. Herkert, Katharina Löhner, Patrick Rump, Martine T. Meems-Veldhuis, Pieter B.T. Neerincx, Jan D.H. Jongbloed, Conny M. van Ravenswaaij-Arts, Morris A. Swertz, Richard J. Sinke, Irene M. van Langen and Cisca Wijmenga Pediatrics originally published online September 22, 2017;

Updated Information & including high resolution figures, can be found at: Services http://pediatrics.aappublications.org/content/early/2017/09/20/peds.2 016-2854 References This article cites 28 articles, 5 of which you can access for free at: http://pediatrics.aappublications.org/content/early/2017/09/20/peds.2 016-2854#BIBL Subspecialty Collections This article, along with others on similar topics, appears in the following collection(s): Critical Care http://www.aappublications.org/cgi/collection/critical_care_sub Genetics http://www.aappublications.org/cgi/collection/genetics_sub Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.aappublications.org/site/misc/Permissions.xhtml Reprints Information about ordering reprints can be found online: http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news by guest on September 28, 2021 Rapid Targeted Genomics in Critically Ill Newborns Cleo C. van Diemen, Wilhelmina S. Kerstjens-Frederikse, Klasien A. Bergman, Tom J. de Koning, Birgit Sikkema-Raddatz, Joeri K. van der Velde, Kristin M. Abbott, Johanna C. Herkert, Katharina Löhner, Patrick Rump, Martine T. Meems-Veldhuis, Pieter B.T. Neerincx, Jan D.H. Jongbloed, Conny M. van Ravenswaaij-Arts, Morris A. Swertz, Richard J. Sinke, Irene M. van Langen and Cisca Wijmenga Pediatrics originally published online September 22, 2017;

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://pediatrics.aappublications.org/content/early/2017/09/20/peds.2016-2854

Data Supplement at: http://pediatrics.aappublications.org/content/suppl/2017/09/20/peds.2016-2854.DCSupplemental

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. Pediatrics is owned, published, and trademarked by the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2017 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.

Downloaded from www.aappublications.org/news by guest on September 28, 2021