Next-generation sequencing (NGS) has enabled the diagnosis and discovery of rare genetic diseases at an unprecedented rate, particularly in children. Since 2011 and the first reported case of a child being diagnosed by whole genome sequencing (WGS) with an early onset inflammatory bowel disease amenable to bone marrow transplant, scores of similar cases of treatable conditions diagnosed by WGS or whole exome sequencing (WES) have emerged (1–6).
Not all WGS or WES workups have such positive clinical outcomes, but families nearly always benefit from having a diagnosis. Defining their child’s condition generally offers families an accurate recurrence risk for genetic counseling purposes and helps them prepare for a medically complicated child, not to mention avoiding a difficult diagnostic odyssey. A molecular diagnosis also provides options for prenatal diagnosis in subsequent pregnancies, among other potential benefits.
In addition to families’ relief in knowing definitively the cause of their child’s problems, many diagnoses directly benefit the child, with about 50% of children diagnosed by WGS or WES undergoing a change in management as a result of these findings. Moreover, an earlier diagnosis leads to curative treatment in some cases (3, 7). Though research has shown that WGS has a high diagnostic rate for genetic diseases that have a direct effect on the management and outcomes of critically ill infants, the benefits of extending this technology to the well-baby nursery or for broader screening are uncertain (7, 8). This article will discuss some of the issues that NGS—be it WGS, WES, or targeted sequencing—presents when combined with newborn screening (NBS).
Changes in Newborn Screening
NBS has been a public health success for more than 50 years, with a proven track record of identifying in asymptomatic newborns conditions for which early diagnosis and management directly benefit the child—reducing or even eliminating associated morbidity and mortality. The very first NBS target, phenylketonuria (PKU), offers a classic and enduring example. PKU was an ideal NBS choice because it met traditional criteria established by Wilson and Jungner that include limiting screening to targets considered severe, well characterized, treatable, and that require urgent intervention to prevent permanent sequelae (9).
Most importantly, under these traditional criteria, screening must directly benefit the newborn. Screening recommendations adhered to these criteria until fairly recently. However, tandem mass spectrometry (MS/MS) has significantly expanded the list of conditions detectable by routine NBS, including several that do not meet traditional screening criteria (10).
To determine which conditions provide enough benefit to warrant uniform inclusion in NBS programs, as well as contend with screening variability among states, the Secretary’s Advisory Committee on Heritable Disorders in Newborns and Children commissioned the American College of Medical Genetics and Genomics (ACMG) to review the available evidence for screening each disorder. These experts reached consensus on the benefits of screening for a core panel of 29 conditions, now called the Recommended Uniform Screening Panel (RUSP) (11). An additional 25 diseases had less agreement surrounding screening benefits, for reasons such as not having a current treatment or being very rare. ACMG recommended these for inclusion as secondary targets.
This decision to include targets that do not meet traditional criteria and that are incidental to the purpose of screening changed the fundamental practice of NBS. NGS-NBS has the potential to identify many more conditions that do not meet traditional screening criteria, as well as a much greater amount of secondary information. This capability has spawned professional discussion as well as research funded by the National Human Genome Research Institute and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (12). Indeed, National Institutes of Health Director Francis Collins has predicted a time in the next few decades when every newborn’s genome will be sequenced, providing the ultimate in personalized medicine (13).
Potential NGS Pitfalls
One advantage posited for NGS-NBS is that laboratories could screen for more childhood diseases without incurring additional costs. This might be true in part on the technical side. However, although the cost of WGS is down to $1,000 in some hands with the right equipment, this figure does not include added costs, such as those associated with analyzing, interpreting, managing, and storing the data, counseling, returning results, or performing follow-up testing to clarify ambiguous results, all of which greatly outweigh the cost of the sequencing. The cost to perform NBS by MS/MS varies, as does the testing menu, from state to state in the U.S., but generally falls in the $30–$100 range. Some states charge a fee for this, while others offer it freely to families (14). The costs to perform NBS-NGS, including downstream costs to the healthcare system, need to be examined carefully. Presently, NGS is not likely to fit within available public healthcare budgets (15).
WGS and WES also have the potential to provide significant information that does not directly benefit newborns, such as their likelihood to develop adult onset conditions or untreatable illnesses, their carrier status, and their harboring hundreds of variants of unknown significance (VUS). This ability to screen for additional conditions already has led some states to screen for targets that do not meet traditional screening criteria, prompting calls for more evidence-based decisionmaking (16). One example is screening for lysosomal storage disorders, including Krabbe disease, which two states currently perform at the behest of parent advocacy groups, despite difficulty in interpreting results and a dearth of evidence that early detection benefits affected children (17).
Currently, laboratories cannot interpret most WGS-generated data, as it remains quite difficult to understand the significance of many variants in the coding regions, which make up only 2% of the genome. The noncoding regions are vast and present an even greater challenge. One option would be to report only pathogenic or likely pathogenic variants, with the goal of minimizing false positives while maximizing identification of affected individuals. However there is not yet consensus on which types of variants to report. Current recommendations call for variants to be curated by experts using the ACMG guidelines for sequence interpretation and submitted to the ClinVar database so they are available to other laboratories performing interpretation (19, 20). Without consensus, interpretation inevitably will differ among laboratories, resulting in inconsistencies that most likely will lead to both under- and over-diagnosis. In addition, genotype/phenotype relationships are not always straightforward, and understanding why, when, or even if a person with a certain profile will develop symptoms is rarely as simple as it is with a disease like PKU, for which the mutation spectrum is fairly well understood and can nonetheless be readily diagnosed via biochemical methods, should the molecular results be unclear.
NGS also raises important questions about future capital and personnel investment required for the nation’s public health laboratories. NBS currently takes place at centralized state laboratories, under standardized conditions using methods that measure enzyme activity, metabolites, or other molecules. DNA testing is not a routine part of the process, with only a few babies in some states receiving a DNA test as a second-tier follow-up on abnormal results, such as for CFTR genotyping after a failed or inconclusive screen for cystic fibrosis by immunoreactive trypsinogen on dried blood spots (21, 22). However, DNA screening is beginning to make its mark. Severe combined immunodeficiency (SCID) now is part of the national recommended panel for newborn screening disorders, and states are using DNA-based assays to detect T-cell receptor excision circles to identify this condition (23). However, implementing NGS testing would require a significant overhaul of equipment and personnel. Few laboratories in the U.S. are currently capable of performing NGS testing at scale (24). Discussions about the path forward for NBS should consider whether state laboratories could continue to adapt to the type of equipment, complicated workflow, and interpretative skill required for NGS-based testing as well as whether a new paradigm should be investigated, such as more centralized testing.
Expanded NGS also would pose another important issue—consent. When NBS identifies a condition that directly benefits the baby by decreasing morbidity and mortality, there is general agreement that such testing may be done without parental consent. However, WGS- or WES-NBS would undoubtedly require parental consent. This could greatly decrease participation, threatening universal NBS (25).
One study revealed parents had variable but overall not very enthusiastic attitudes about the potential future expansions of NBS to include genomic testing. This suggests that parental choice will be an important consideration in future screening programs (26). Other surveys of parents have indicated the majority had a high interest in hypothetical NGS-NBS of their newborn. In reality, the uptake may be much lower and should be studied carefully (27, 28). Policies would be needed to resolve discordant interest between parents in granting consent for screening, which was high in one survey (28). When laboratories offer choices about the type of results to be returned, as is the case with some clinical WGS testing laboratories, parents have another opportunity to disagree about how to proceed. Presenting such choices could create confusion about which disorders the infant has been screened for, since this would vary according not only to state of birth (as it does currently despite the RUSP) but also the preferences of parents. In addition, it is unclear what types of healthcare professionals would provide consent in birthing hospitals, particularly in more rural areas, since the U.S. already has a shortage of genetic counselors. Nurses and other hospital staff members would require a great deal of education to fulfill this role, and parents would need decision aids and other educational materials.
Dealing With Uncertain Results
NBS aims to identify affected individuals while avoiding false positives, which add to parental anxiety and the cost of follow-up. However, a screening test is not meant to be diagnostic and will always have some false positives, regardless of methodology. Although the utility of NGS-NBS has not been fully assessed, one prospective study of 1,696 healthy newborns who underwent traditional NBS and trio WGS, i.e., baby and both parents, for the analysis of 163 disorders found a high rate of uncertain WGS-based results relative to conventional NBS (0.90% versus 0.013%, respectively) (29). Most of these findings were VUS in a gene associated with a dominant disorder; however, screening for a disease affecting 23.6% of the population obviously would not be viable. One solution might be to not report VUS, but this would miss some affected individuals with variants that are truly pathogenic.
This study also found that conventional NBS outperformed WGS in terms of diagnostic rate, identifying four of five affected neonates versus two of five by WGS. Of the three individuals missed by WGS, one had congenital hypothyroidism and a VUS in the associated gene, NKX2-1, which exemplifies the problem of uninterpretable variants in the genome. The second WGS miss involved a patient with a metabolic disorder but only one detectable allele in the corresponding gene, HADHB, despite subsequent extensive clinical testing, including interrogation for deletion/duplications. The third missed case had salt-wasting congenital adrenal hyperplasia with no variant detected by WGS in the corresponding gene, CYP21A2, likely due to pseudogene interference. These cases highlight the current shortfalls of NGS: undetectable or uninterpretable alleles, partial genotypes, and problems sequencing genes with highly homologous sequences.
On the positive side, WGS gave fewer false positives, was able to resolve inconclusive results from conventional NBS, identified the causative variants in affected individuals, and distinguished disorders not resolvable by traditional NBS. In addition, fewer sample collections were needed for preterm infants.
One question still unanswered is the scope of disorders detected by NGS-NBS, particularly for treatable conditions not currently picked up by NBS. The authors suggested that combining the two programs would be highly synergistic, particularly since trio WGS is an effective means of phasing alleles to sort out carriers from those with two defective alleles, thus eliminating false positives. However, the cost would be prohibitive, particularly when adding the cost of sequencing and interpreting parental genomes.
Next Steps for Policymakers and Scientists
NGS-NBS likely will be implemented at some point in the future, whether it replaces traditional NBS or serves as a technological supplement for screening disorders within the existing framework. However, the technology is well ahead of policy, and NGS likely will exacerbate current NBS challenges. These include costs, standards for which genes to include and which variants to report, infrastructure, and education of the medical community.
Further additions to the RUSP, particularly in light of NGS-NBS, should be selected by expert committee and potentially approved by the Secretary’s Advisory Committee, which recently introduced a much more rigorous evidence review process (18). The list of targets should be restricted to genes and variants that have robust disease-gene relationships for which there is agreement that screening would be beneficial. The consequences of the diseases screened should be clear, have well-understood inheritance patterns, penetrance, and variability. The list of included conditions also should undergo periodic re-evaluation. Targeted NGS-NBS could be performed for this set of agreed upon conditions, which would limit any off-target findings and be potentially less costly than WES or WGS. An alternative would be analyzing virtual panels from WES or WGS, in which off-target conditions would be bioinformatically masked.
For the continued success of NBS, policy recommendations should take on these issues before laboratories implement NGS for such screening purposes. As NBS enters the genomics era, thoughtful dialogue and planning will be necessary to preserve this lifesaving 50-year old program.
References
1. Bainbridge MN, Wiszniewski W, Murdock DR, et al. Whole-genome sequencing for optimized patient management. Sci Transl Med 2011;3:87re3.
2. Dinwiddie DL, Bracken JM, Bass JA, et al. Molecular diagnosis of infantile onset inflammatory bowel disease by exome sequencing. Genomics 2013;102:442–7.
3. Soden SE, Saunders CJ, Willig LK, et al. Effectiveness of exome and genome sequencing guided by acuity of illness for diagnosis of neurodevelopmental disorders. Sci Transl Med 2014;6:265ra168.
4. Dixon-Salazar TJ, Silhavy JL, Udpa N, et al. Exome sequencing can improve diagnosis and alter patient management. Sci Transl Med 2012;4:138ra78.
5. Tarailo-Graovac M, Shyr C, Ross CJ, et al. Exome sequencing and the management of neurometabolic disorders. N Engl J Med 2016;374:2246–55.
6. Worthey EA, Mayer AN, Syverson GD, et al. Making a definitive diagnosis: Successful clinical application of whole exome sequencing in a child with intractable inflammatory bowel disease. Genet Med 2011;13:255–62.
7. Willig LK, Petrikin JE, Smith LD, et al. Whole-genome sequencing for identification of mendelian disorders in critically ill infants: A retrospective analysis of diagnostic and clinical findings. Lancet Respir Med 2015;3:377–87.
8. Saunders CJ, Miller NA, Soden SE, et al. Rapid whole-genome sequencing for genetic disease diagnosis in neonatal intensive care units. Sci Transl Med 2012;4:154ra35.
9. Wilson JM, Jungner YG. Principles and practice of mass screening for disease. Bol Oficina Sanit Panam 1968;65:281–393.
10. Chace DH, Kalas TA, Naylor EW. Use of tandem mass spectrometry for multianalyte screening of dried blood specimens from newborns. Clin Chem 2003;49:1797–817.
11. Newborn screening: Toward a uniform screening panel and system. Genet Med 2006;8 Suppl 1:1S–252S.
12. Holm IA. Newborn sequencing in genomic medicine and public health (NSIGHT): Genomic sequencing and newborn screening disorders u19 projects. https://www.nbstrn.org/sites/default/files/holm_nbstrn_2014-09-22_pdf.pdf (Accessed January 2017).
13. Collins F. Francis Collins says medicine in the future will be tailored to your genes. Wall Street Journal 2014 July 7.
14. National newborn screening & global resource center. http://genes-r-us.uthscsa.edu/resources/consumer/statemap.htm (Accessed October 2016).
15. Beckmann JS. Can we afford to sequence every newborn baby’s genome? Hum Mutat 2015;36:283–6.
16. The changing moral focus of newborn screening: An ethical analysis by the president’s council on bioethics. http://hdl.handle.net/10822/559367 (Accessed January 2017).
17. Dimmock DP. Should states adopt newborn screening for early infantile Krabbe disease? Genet Med 2016;18:217–20.
18. Perrin JM, Knapp AA, Browning MF, et al. An evidence development process for newborn screening. Genet Med 2010;12:131–4.
19. Clinvar. https://www.ncbi.nlm.nih.gov/clinvar/ (Accessed November 2016).
20. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: A joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 2015;17:405–24.
21. Ranieri E, Lewis BD, Gerace RL, et al. Neonatal screening for cystic fibrosis using immunoreactive trypsinogen and direct gene analysis: Four years’ experience. BMJ 1994;308:1469–72.
22. Gregg RG, Wilfond BS, Farrell PM, et al. Application of DNA analysis in a population-screening program for neonatal diagnosis of cystic fibrosis (cf): Comparison of screening protocols. Am J Hum Genet 1993;52:616–26.
23. Verbsky JW, Baker MW, Grossman WJ, et al. Newborn screening for severe combined immunodeficiency; the Wisconsin experience (2008-2011). J Clin Immunol 2012;32:82–8.
24. Kwan A, Abraham RS, Currier R, et al. Newborn screening for severe combined immunodeficiency in 11 screening programs in the United States. JAMA 2014;312:729–38.
25. Levy HL. Newborn screening: The genomic challenge. Mol Genet Genomic Med 2014;2:81–4.
26. Kerruish N. Parents’ experiences 12 years after newborn screening for genetic susceptibility to type 1 diabetes and their attitudes to whole-genome sequencing in newborns. Genet Med 2016;18:249–58.
27. Goldenberg AJ, Sharp RR. The ethical hazards and programmatic challenges of genomic newborn screening. JAMA 2012;307:461–2.
28. Waisbren SE, Back DK, Liu C, et al. Parents are interested in newborn genomic testing during the early postpartum period. Genet Med 2015;17:501–4.
29. Bodian DL, Klein E, Iyer RK, et al. Utility of whole-genome sequencing for detection of newborn screening disorders in a population cohort of 1,696 neonates. Genet Med 2016;18:221–30.
Carol J. Saunders, PhD, FACMG, is professor of pediatric pathology at the University of Missouri-Kansas City School of Medicine, and clinical director of the Center for Pediatric Genomic Medicine and director of the molecular genetics and diagnostics laboratory at Children’s Mercy Hospital in Kansas City, Missouri. +Email: [email protected]