The future of newborn screening for lysosomal disorders

Abstract

The goal of newborn screening is to enhance the outcome of individuals with serious, treatable disorders through early, pre-symptomatic detection. The lysosomal storage disorders (LSDs) comprise a group of more than 50 diseases with a combined frequency of approximately 1:7000. With the availability of existing and new enzyme replacement therapies, small molecule treatments and gene therapies, there is increasing interest in screening newborns for LSDs with the goal of reducing disease-related morbidity and mortality through early detection. Novel screening methods are being developed, including efforts to enhance accuracy of screening using an array of multi-tiered, genomic, statistical, and bioinformatic approaches. While NBS data for Gaucher disease, Fabry disease, Krabbe disease, MPS I, and Pompe disease has demonstrated the feasibility of widespread screening, it has also highlighted some of the complexities of screening for LSDs. These include the identification of infants with later-onset, untreatable, and uncertain phenotypes, raising interesting ethical concerns that should be addressed as part of the NBS implementation process. Taken together, these efforts will provide critical, detailed data to help guide objective, ethically sensitive decision-making about NBS for LSDs.

1. Overview of newborn screening and lysosomal storage disorders

Newborn screening (NBS), undoubtedly one of the most successful public health programs, began over five decades ago after the discovery that phenylketonuria is an easily diagnosed, preventable cause of intellectual disability [1–3]. The goal of NBS is to detect potentially fatal or serious, treatable conditions in newborns as early as possible, before symptoms manifest. Early detection through NBS enables timely intervention and initiation of treatment. In recent years, screening technologies have led to a remarkable growth of the numbers and types of disorders on NBS panels. As the technical ability to screen continues to expand, determining which disorders are appropriate for mandated NBS is an ongoing concern. The Wilson and Jungner principles, traditionally used to guide the selection of disorders suitable for public screening, focus on screening for disorders that present and require treatment during childhood [4]. While these criteria have been used to guide NBS expansion, it has been the development of innovative screening assays and novel therapies that has largely driven the addition of new disorders [5–7].

The lysosomal storage disorders (LSDs) comprise a group of more than 50 diseases with a combined frequency of approximately 1:7000 [8]. The majority of LSDs are caused by inherited enzyme deficiencies, while a few result from membrane transport and other loss of function effects. Cofactor deficiencies account for a small minority of LSDs. Clinical presentations are often multi-systemic, although many LSDs have a distinct neurologic presentation that may be progressive and disabling. Most LSDs have a broad phenotypic spectrum ranging from severe early onset to less severe later onset phenotypes, where the severity of clinical presentation often correlates with the level of residual biochemical function of the mutated protein. For some LSDs, the later onset phenotypes are more prevalent than the early-onset forms.

There is a wide range of treatment options for LSDs, including enzyme replacement therapy (ERT), pharmaceutical chaperones, substrate reduction therapy (SRT), and hematopoietic stem cell transplantation [9]. A new therapeutic strategy is rapidly gaining traction, with the recent European Commission approval of a lentiviral vector-based, ex-vivo, gene therapy for the treatment of metachromatic leukodystrophy (MLD) [10]. Gene therapy clinical trials for several other neuronopathic LSDs including GM1 gangliosidosis, Krabbe leukodystrophy, and mucopolysaccharidosis types IIIa and IIIb are being planned or are currently underway. Multiple therapeutic interventions are under development for NPC, including intrathecal and intravenous cyclodextrins, arimoclomol, and N-acetyl-L-leucine. A new intravenous ERT is in clinical trial for acid sphingomyelinase deficiency [11], an intraventricular ERT is approved for CLN2, and intrathecal ERT is in trial for MLD. As early treatment is presumed to optimize outcomes, there is interest in screening newborns for LSDs with the goal of reducing disease-related morbidity and mortality through early detection.

When considering NBS for LSDs, it is important to note that some current screening methods are unable to distinguish between early and later onset phenotypes. This means that individuals at risk for later- or adult-onset LSDs may be diagnosed shortly after birth. While not distinct to LSDs (other examples include X-linked adrenoleukodystrophy and cystic fibrosis), this ability to screen infants for disorders that may not present for years or even decades raises interesting ethical considerations. Given that the traditional moral focus of NBS has been on direct benefits to the infant, there are some concerns as to whether NBS for LSD’s should be mandated.

2. Overview of some LSDs that may benefit from NBS

Acid Sphingomyelinase Deficiency (ASMD, aka Niemann-Pick type A/B) results from bi-allelic pathogenic variants in SMPD1, leading to the accumulation of sphingomyelin within cells of the monocyte-macrophage system [12]. The acute neurovisceral form, or Niemann Pick-A, leads to death in infancy; the milder phenotype, also known as Niemann Pick -B, is characterized by hepatosplenomegaly, hyperlipidemia, and pulmonary involvement; and the intermediate form, Niemann Pick A/B, is a slowly progressive neuronopathic disorder [12,13]. Clinical trials with recombinant acid sphingomyelinase are currently underway for the non-neurologic manifestations of ASMD [11].

Ceroid Lipofuscinosis type 2 (CLN2) is a neurologic disorder caused by deficiency of tripeptidyl peptidase 1 resulting from bi-allelic pathogenic variants in TPP1. It typically manifests between two and four years of age with the development of seizures, developmental regression, ataxia, myoclonus, and vision loss [14]. Without treatment, it is usually fatal during adolescence. ERT with intraventricular cerliponase alfa has been shown to stabilize CLN2 symptoms [15].

Fabry Disease is an X-linked disorder of glycosphingolipid metabolism caused by the deficient activity of α-galactosidase A due to hemizygous or heterozygous pathogenic variants in GLA [16]. Classically affected males typically develop painful acroparesthesias, angiokeratomas, corneal opacities, progressive renal failure, cardiovascular and cerebrovascular disease. Later-onset variants of Fabry disease develop renal [17] and cardiac [18] manifestations during adulthood. Heterozygous females have a wide range of disease severity [19]. Fabry disease is treatable with ERT, and a subset of responsive mutations are also responsive to oral chaperone therapy [20–25].

Gaucher Disease is caused by the accumulation of glucosylceramide in cells of the monocyte/macrophage system due to deficient activity of acid beta glucosidase resulting from bi-allelic pathogenic variants in GBA [26]. Clinical manifestations shared by all three types of Gaucher disease include hepatosplenomegaly, skeletal involvement, and cytopenias. Neurologic disease is present in types 2 and 3. Type 2 disease is usually fatal during infancy due to rapid neurodegeneration and severe systemic disease and is currently untreatable. Type 1 Gaucher is effectively treated with ERT and substrate reduction therapy due to the lack of CNS involvement [27]. High dose ERT, SRT, and HSCT may slow the progression of disease in some type 3 patients [28–30].

GM1 gangliosidosis results from bi-allelic pathogenic variants in the GLB1 gene, resulting in insufficient activity of beta galactosidase [31]. The subsequent accumulation of GM1 ganglioside results in progressive developmental disabilities, skeletal abnormalities, enlarged liver and spleen, vision loss, bone disease, and distinctive facial feature. Children with the most severe form of GM1 gangliosidosis usually do not survive past early childhood. There are also less severe forms of GM1 gangliosidosis that progress more slowly. Gene therapy trials are currently underway.

Krabbe Disease is caused by bi-allelic pathogenic variants in GALC resulting in deficiency of galactocerebrosidase (GALC), an enzyme essential for the normal turnover of myelin [32]. Infants with the most severe form, early infantile Krabbe disease, present with irritability, cortical fisting, and stiffness by six months of age [33]. The disease progresses rapidly, with most dying by two years of age [33,34]. Individuals with late infantile Krabbe disease develop symptoms between six and twelve months and follow a progressive neurodegenerative course [35]. Those with later onset phenotypes may present from childhood to adulthood with variable symptoms that may include ataxia, visual disturbances, and dementia [36–38]. At present, the only treatment for Krabbe disease is HSCT which must be done early in the disease course for it to be effective [39,40], although preclinical gene therapy studies are underway.

Lysosomal Acid Lipase deficiency (LAL), caused by bi-allelic pathogenic variants in the LIPA gene, results in an accumulation of lipids in various organs, resulting in hepatosteatosis, liver dysfunction and fibrosis, steatorrhea, developmental delay, and malnutrition. Later onset forms may include atherosclerosis [41]. ERT with sebelipase alfa is approved for children and adults with LAL deficiency [42].

Metachromatic Leukodystrophy (MLD) is caused by pathogenic variants in ARSA, resulting in arylsulfatase A deficiency. Onset ranges from late infancy through adulthood. Affected individuals have progressive and eventually fatal neurologic deterioration, with the younger age of onset associated with the most severe and rapid rate of progression [43]. HSCT has been successfully used in MLD and is more effective when performed early in the disease course [44]. A lentiviral-based, ex-vivo transfection of hematopoietic stem cells was recently approved in Europe [45]. Additional investigational products currently in clinical trial include intrathecal ERT.

Mucopolysaccharidosis type 1 (MPS I), is a multisystemic disorder resulting from deficient activity of alpha-iduronidase caused by bi-allelic pathogenic variants in IDUA. Like all MPS disorders (see below), enzyme deficiency results in accumulation of specific glycosaminoglycans [46]. Severe MPS I presents during infancy with the gradual development of coarse facial features, dysostosis multiplex, hepatosplenomegaly, and intellectual disability. If untreated, it is usually fatal during childhood [46]. Attenuated MPS I has a more variable age of onset, and a broader range of neurocognitive outcome. Therapeutic options for all forms of MPS I include intravenous ERT and HSCT.

MPS II, or Hunter Syndrome, is an X-linked disorder resulting from hemizygous variants in IDS. The resulting iduronate-2-sulfatase deficiency leads to the progressive accumulation of glycosaminoglycans throughout the body [47]. Boys with the severe phenotype have coarsened facial features, hepatosplenomegaly, developmental delay, behavioral abnormalities, typical skin lesions, and dysostosis multiplex [48]. Treatment with intravenous ERT has been shown to stabilize somatic signs of disease [49]. Several clinical trials aimed at the CNS manifestations are currently underway.

MPS III, or Sanfilippo disease, results from deficiency of N-sulpho-glucosamine sulphohydrolase (MPS IIIA, SGSH gene), and alpha-N-acetylglucosaminidase (MPS IIIB, NAGLU gene). MPS IIIC and IIID are more rare. Affected children have developmental delay, mildly coarse facial features, and behavioral problems [50,51]. Clinical trials with gene therapy are underway for MPS IIIA and IIIB.

MPS IVA, also known as Morquio A syndrome, results from deficient activity of N-acetylgalactosamine-6-sulfate sulfatase due to bi-allelic pathogenic variants in GALNS. MPS IVB is caused by mutations in the GLB1 gene. Affected individuals have dysostosis multiplex, cataracts, umbilical hernia, coarse facial features, cardiovascular symptoms, and normal intelligence [49] ERT is approved to treat MPS IVA [49].

MPS VI, or Maroteaux-Lamy syndrome, is caused by deficient activity of arylsulfatase B due to bi-allelic pathogenic variants in ARSB. Phenotypic features include coarsened facial features, hepatosplenomegaly, and dysostosis multiplex, but MPS VI does not affect intelligence [52]. Intravenous ERT has been shown to improve survival and some of the physical manifestations of disease [53].

MPS VII, also known as Sly Syndrome, is caused by pathogenic variants in GUSB resulting in β-glucuronidase deficiency. Clinical manifestations include coarsened facial features, hepatosplenomegaly, cardiovascular disease, cataracts, and dysostosis multiplex [54,55]. Intravenous ERT with recombinant vestronidase alfa is approved for children and adults with MPS VII.

Niemann-Pick Disease Type C (NPC) is a progressive, neurodegenerative disease caused by mutations in NPC1, a transmembrane protein involved in cholesterol transport, or NPC2, a cholesterol-binding protein [56]. NPC has a broad phenotypic spectrum that includes hepatosplenomegaly, neonatal cholestasis, ataxia, seizures, and progressive impairment of motor and intellectual function [57]. There are a few ongoing clinical trials for NPC, including cyclodextrins in various forms and routes of administration, as well as trials with other types of small molecule therapies.

Pompe disease, also known as glycogen storage disease type II or acid maltase deficiency, is caused by deficient activity of the lysosomal hydrolase acid α-glucosidase resulting from bi-allelic pathogenic variants in GAA [58]. The most severe phenotype is a classic infantile-onset disease with hypotonia, cardiomyopathy, and death by two years of age [59]. Patients with milder disease have a progressive myopathy which presents during adolescence or even in the later decades of adulthood. ERT has been shown to be highly effective at altering the course of infantile Pompe disease and [60] stabilization of later onset disease [61].

3. History of NBS for LSDs

Pilot studies are an integral part of the history of NBS. Anonymous retrospective pilot studies in Washington, Italy, Taiwan and Hungary were amongst the first to assess various screening assays for LSDs [62–68]. Pilot studies using de-identified residual NBS dried blood spots (DBS) are often performed to learn about the positive rates of newly developed screening assays, to test the robustness of the NBS assay, and to obtain information about expected incidence rates for various disorders. Because they are de-identified, there is no correlation of screen positivity with clinical phenotype so asymptomatic cases may be included in incidence rates. De-identified pilot NBS studies are also performed to assess novel approaches, such as quantitation of disease-specific biomarkers, to enhance screening accuracy. For example, a recent pilot study for MLD demonstrated that a two-tiered screening approach using quantification of DBS sulfatide levels followed by measurement of ARSA activity (both by mass spectrometry) showed near 100% assay specificity [69]. In addition, a novel two-tiered, mass spectrometry approach using measurement of DBS β-galactosidase followed by analysis of two glycan biomarkers demonstrated accurate detection of GM1- gangliosidosis [70].

Live NBS for LSDs began in 2003, when 37,000 Italian newborns were screened for Fabry disease [62]. Subsequently, Taiwan initiated screening for Pompe disease [68], followed by Fabry disease a few years later [71]. These early programs demonstrated the feasibility of screening and provided evidence about the disease incidence in these populations.

In 2006, New York (NY) was the first state in the US to include a lysosomal disorder as part of its routine NBS panel. Screening for Krabbe disease began after a report suggested that pre-symptomatic HSCT can alter the outcome of early infantile disease, which is otherwise uniformly fatal [40]. To date, more than 3.5 million NY infants have been screened for Krabbe disease.

In 2015, Pompe disease was the first LSD to be added to the Recommended Uniform Screening Panel, or the RUSP [72,73]. The RUSP is a recommended, but not mandated, list of disorders for inclusion on all states’ NBS panels. Disorders are added to the RUSP after extensive evidence-based review by the US Advisory Committee on Heritable Disorders in Newborns and Children. The inclusion of Pompe disease on the RUSP was based on the finding that early initiation of alglucosidase alfa can significantly improve cardiac and motor function as well as overall survival in babies with classic infantile onset disease, a disorder which is universally fatal when untreated [74]. MPS I was added to the RUSP in 2016 [75] based on the presumption that early, pre-symptomatic treatment can improve both the neurologic and systemic features of the disease. At present, about 20 states screen all newborns for Pompe disease and MPS I using biochemical testing often followed by genotyping, although it is anticipated that 38 states will be screening for both disorders within the next few years [76]. In addition to Pompe disease and MPS I, several states have mandated screening for LSDs not currently on the RUSP, including Fabry disease, Gaucher disease, Krabbe disease, ASMD, and MPS II [76]. Italy, Brazil, Australia, Austria, Belgium, Hungary, Spain, Japan, South Korea, and Taiwan also screen newborns for one or more LSDs [77].

Early experiences with live screening for LSDs have reported some common themes. First, NBS has identified infants with severe, infantile onset LSDs who clearly benefited from early detection and initiation of treatment [74,78]. Second, there have relatively high numbers of false positives [79–81]. Third, in some programs, pseudodeficiencies were detected more often than true deficiencies, particularly for MPS 1 and Pompe disease [79]. Pseudodeficiency alleles, genetic variants that alter enzymatic activity without causing disease, may result in DBS enzyme activity below the referral cutoff, despite the individual not being at risk for disease. Fourth, the disease incidences are often not as anticipated. In particular, the incidences of Krabbe, Fabry and Pompe disease have consistently been higher than pre-screening estimates [67,79,80,82], largely because of the fifth finding, which is the detection of higher than expected numbers of individuals at risk for later onset disease [83,79,80,82]. Disease incidence data that has been reported by NBS programs is summarized in Table 1. Lastly, as expected, previously unreported variants and polymorphisms have been identified in newborns who screen positive for an LSD, which can make the predictions of phenotypic severity and age of onset challenging [84].

Table 1.

Summary of Reported Disease Incidences. For Krabbe disease and MPS I, the reported incidences include early onset phenotypes only. For the remainder of the disorders, the reported incidences may include early, intermediate, and later onset phenotypes.

LSD	NBS Programs with Reported Disease Incidence	Range of Overall Disease Incidence
ASMD	Il, Hungary	1:20,012 [64] – 1:109,987 [79]
Fabry	Hungary, Italy, Japan, Taiwan, NY, MO, IL, WA	1:1,250 [67] – 1:13,341 [64]
Gaucher	NY, MO, IL, Hungary	1:4,374 [82] – 1:43,959 [79]
Krabbe	NY	1:400,000 [81]
MPS I	IL, MO	1:14,567–1:219,793 [79]
MPS II	IL	1:113,090 [121]
Pompe	Hungary, Taiwan, NY, MO, IL, WA	1:4,447 [64] – 1:23,596 [78]

4. NBS assays

Measurement of enzyme activity is currently the most commonly used primary screening test for LSDs. Nestor Chamoles was the first to report lysosomal enzymatic activity measurements using dried blood spots (DBS) [85]. These assays made use of a fluorometric enzyme substrate allow single-disease assays using a 96-well fluorescence plate reader. In 2004, Gelb, Scott, and Turecek reported the use of tandem mass spectrometer (MS/MS) for the multiplex assay of several lysosomal enzymatic activities in DBS [86]. The MS/MS method has been commercialized into an FDA-approved NBS kit (NeoLSD) by PerkinElmer, Corp, that allows multiplex assay of 6 LSDs (Pompe, MPS-I, Krabbe, ASMD, Gaucher, and Fabry), and this is used in the majority of states now screening for LSDs in the USA as well as in Taiwan. Fluorometric assays of lysosomal enzymatic activities have been commercialized on a digital microfluidics platform (the FDA-approved Seeker system from Baebies, Corp) [87].

The MS/MS enzymatic assay becomes highly multiplexable when the platform is combined with liquid chromatography (LC-MS/MS). For example, a single multiplex assay has been developed for all 10 types of MPS disorders [88]. An 18-plex LC-MS/MS assay has also been recently reported which allows several enzymatic activities as well as biomarkers (see below) to be analyzed in a single run [89]. The LC-MS/MS assay is currently being used in a new prospective pilot study (ScreenPlus) to simultaneously screen for 14 disorders (11 based on enzymatic activities and 3 based on biomarkers). The Illinois NBS laboratory is also using LC-MS/MS to screen for 7 LSDs simultaneously.

In addition to enzyme-based assays, biomarker-based assays for LSDs may be used for NBS. Biomarker assays may measure the accumulation of substrate that results from a deficiency of the enzyme that acts on it. Biomarker assays may also be helpful when an aberrant metabolite accumulates due to the loss of function of a lysosomal protein. An example of the former is the accumulation of sulfatides in MLD [90], while an example of the latter is accumulation of bile acid B in DBS from Niemann-Pick-C patients [91].

Some efforts, notably the NSIGHT projects, have been evaluating the utility and implementation of genomic sequencing as a form of NBS [92]. The NBSeq project comparing WES to biochemical screening showed that WES had an overall sensitivity of 88% and specificity of 98.4%, compared to 99.0% and 99.8%, respectively for MS/MS, although effectiveness varied among individual inborn errors of metabolism [93]. The presence of variants of uncertain significance (VUS), challenges in predicting the pathogenicity of novel compound heterozygous combinations, and detection of intronic and copy number variants are examples of some concerns that are currently impacting the ability to utilize genomic sequencing as a primary NBS modality. Until these issues are resolved in a way that enables it to be at least as accurate as current primary or multi-tiered NBS methods, genomic sequencing is complementary to biochemical metabolite screening, but cannot yet replace it. When these issues are less problematic, genomic sequencing is likely to become the future norm in NBS, as it enables screening for disorders that may not have a biochemical or enzymatic marker.

5. Enhancing accuracy of screening

Because prior experience with screening newborns for LSDs has demonstrated a high false positive rate [79,81,94], it has become a priority to enhance the accuracy of screening assays to reduce the number of false positive outcomes. One such approach is to use multi-tiered testing, often using a combination of enzyme activity, biomarker quantitation, and/or DNA analysis. Bioinformatics and statistical approaches are also being used to reduce the number of false positive cases.

New York initially began screening newborns for Krabbe disease and Pompe disease using a two tiered approach, where DBS specimens with low enzyme activities were reflexed to Sanger sequencing of the relevant gene [95]. However, this two-tiered approach still resulted in high screen positive and false positive rates mainly resulting from variants of uncertain significance, so the Collaborative Laboratory Integrated Reports (CLIR) was integrated into the screening algorithm to assess the risk for disease. The CLIR tool is a post-analytic interpretive tool that uses multivariate pattern recognition software to enhance accuracy of screening [96]. With CLIR, the number of babies referred for Krabbe disease was reduced by almost 80% and the number of babies referred for Pompe disease was reduced by almost 32%. A similar approach was taken in Georgia using a combination of enzyme activity measurement, comparison with additional enzyme activities, and use of the CLIR tool, and showed similar significant improvements in screening accuracy [97].

Biomarkers are another way to enhance accuracy. Psychosine is now part of the routine NBS for Krabbe disease in NY, IL, TN, and IN. The measurement of psychosine by LC-MS/MS in DBS with low GALC enzymatic activity greatly reduces the false positive rate. Moreover, the level of psychosine enables distinguishing early infantile disease from other forms [98]. Similarly, quantitation of heparan and dermatan sulfatide-derived fragments in DBS greatly reduces the false positives in NBS of MPS-I [96,99,100], accurately discriminates between patients with confirmed MPS I and false-positive cases due to pseudodeficiency or heterozygosity, and increases the specificity of newborn screening for MPS I. Psychosine, heparan sulfate and dermatan sulfate are used as second-tier assays instead of first-tier because these biomarker analyses take 5- to 10-fold longer per sample than the enzymatic activity assay. On the other hand, for MLD, the sulfatide biomarker can be rapidly analyzed and is the preferred first-tier assay, which is followed by measurement of ARSA enzymatic activity as a second tier assay [69]. In the case of Niemann-Pick-C, the only available biochemical assay is the biomarker measurement [91]. Other biomarkers being currently evaluated to support LSD NBS are listed in Table 2.

Table 2.

Biomarkers that are currently in use or are under consideration as first or second tier analytes in NBS.

LSD	DBS Biomarker(s)
ASMD (Niemann Pick A/B)	Lysosphingomyelin, Lyso 509 [116,117]
Fabry disease	Lyso-Gb3 [118]
Gaucher disease	Lyso-Gb1 [119]
GM1 gangliosidosis	Dp5, A2G2 [70]
Krabbe disease	Psychosine [98]
Metachromatic Leukodystrophy	Sulfatides [69]
MPS disorders	Glycosaminoglycans [99]
Niemann Pick C	Bile Acid B, [91] Lyso 509 [116]
Pompe disease	Creatine/creatinine ratio [120]

A novel statistical approach using bivariate analysis is also under investigation. Utilizing a combination of enzyme assay and biomarker concentration, bivariate analysis has shown promise at reducing false positive rates and identifying individuals at risk for early onset forms of Krabbe disease [101] and MPS I [102].

Gene sequencing is used in some NBS programs as part of a multi-tiered algorithm to enhance accuracy. The detection of known pathogenic variants through sequencing enables more accurate prognosis, medical management, and counseling. For example, NY DBS samples with low galactocerebrosidase activity are reflexed to sequencing of the GALC gene, so an infant whose DBS has low GALC activity and bi-allelic 30 kb GALC deletions is assumed to have early infantile Krabbe disease and immediately referred for HSCT. Although non-informative sequencing may not be immediately helpful to the individual patient, ongoing collection of VUS data and correlation with clinical status may yield important information about pathogenicity that can be used to enhance overall screening accuracy.

6. Ethical considerations

NBS for LSDs pose several challenging ethical and social issues that are important to address not only for decision-making about when conditions are ready for addition to state panels, but also to help implement screening effectively while protecting newborns and families from potential physical and psychological harms. Many of the ethical issues associated with LSDs represent challenges to traditional screening, including 1) differential severity and risk 2) more variable ages of onset 3) uncertainty of results 4) challenges in assessing benefits to screen positive newborns with later onset disease. These issues are not entirely novel or exclusive to LSDs, however increased interest in adding these conditions to NBS panels have brought more attention to the social and ethical complexities of expanded screening more generally.

6.1. Differential risk and severity of disease

Experience has demonstrated that for most LSDs, NBS will detect newborns across the full phenotypic spectrum of severity, although not all subtypes benefit equally from early detection. NBS for Krabbe disease, for example, began based on the presumed benefit of early HSCT for infantile onset disease. However, NBS has detected more children at risk for later onset phenotypes [79–83], for which there is less consensus on when and how to treat. Conversely, for some LSDs, the later-onset phenotypes are treatable whereas the severe early-onset forms are currently not. For example, most current Gaucher disease therapies target systemic, non-neurologic disease manifestations. However, the severe, rapidly progressive neurodegenerative manifestations that are typical of type 2 Gaucher disease are not amenable to current therapies. In the future, this situation may be the same for ASMD, where olipudase alfa is currently under development for the chronic visceral phenotype (ASMD type B) [11] but not the acute neurovisceral form (ASMD type A), which is currently untreatable. Screening for conditions with untreatable subtypes may therefore challenge the moral justification for mandatory screening and cause possible psychological burdens for families who may not have wanted information about their newborns when no clinical action is possible. Decision-making about widespread LSD NBS implementation will need to weigh the benefits of early detection of treatable LSDs vs the potential harms of identifying non-treatable forms of disease. Increased parental education about the possible results of LSD screening and/or considerations of non-mandatory consent-based screening may help to address these moral concerns.

6.2. Variable ages of onset

To reduce the risk of missing affected infants, most NBS programs use conservative cutoffs, which may have the effect of detecting milder, later onset forms. In fact, currently, most infants with LSDs detected through NBS are ultimately diagnosed to be at risk for later onset presentations [82,62,79]. The American Academy of Pediatrics recommends against testing children for adult-onset genetic diseases [103], as do essentially all genetics and genetics counseling societies [104–106]. The typical argument is that there is no benefit to be had from testing, that “the unbearable certainty of knowing” is, in fact, a great harm, and that such early detection creates the “patients in waiting” of Timmermans and Buchbinder [107]. Another component to this argument is that of autonomy, where returning findings that indicate an adult onset form of disease may violate that individual’s [108–110] choice to know clinical information about themselves. Nevertheless, the literature on this “right to an open future” [111] in genetics and medicine is evolving. There are new arguments that are challenging our previous assumptions about the value and ethical implications of giving later onset information to parents, particularly when harms may be mitigated, and the potential value of that information may promote the best interests of the child [112].

6.3. Uncertainty of results

A closely related issue involves cases where there is uncertainty with respect to diagnosis and/or phenotypic severity, and whether this uncertainty represents a potential harm of NBS for LSDs [113]. There are two types of uncertainty in NBS; first, in cases when it is uncertain or inconclusive whether or not a child identified through NBS truly has a clinically impactful form of disease, and second, when there is uncertainty about if and/or when clinical manifestations may occur. These scenarios typically occur in the context of the identification of VUS or novel compound heterozygous bi-allelic configurations in the disease gene. While health care providers typically monitor these infants over time for signs and symptoms of disease onset, limited evidence to date suggests that the uncertainty may be challenging and anxiety-provoking for parents as well as health care providers [114]. In the first several years of NY’s screening for Krabbe disease, very few families in this “uncertain” category came for all recommended monitoring visits [81]. Uncertainty itself may not be a significant ethical or social issue for LSD screening, however little data exists on the short- and long-term psychological impact that receiving uncertain results may have on families. Therefore, whether this uncertainty in NBS is an actual harm to the family and/or the infant remains to be determined.

6.4. Assessing benefits

Benefits in NBS traditionally have focused on direct clinical benefits to the individual being diagnosed. However, many later onset LSD cases do not require immediate intervention and for some disorders, it is not yet clear if there is a clinical benefit to early diagnosis [115]. Because some later onset LSDs such as type 1 Gaucher disease are more slowly progressive than the corresponding early onset phenotypes (e.g. type 2 Gaucher disease), there is likely a flexible window of time in which effective, disease-stabilizing or reversing treatment can be initiated. It remains to be seen whether there is a demonstrable difference in long term outcome if NBS permits treatment to be started earlier than it otherwise would have been based on clinical symptomatology. Early detection of later onset LSD phenotypes also raises the question of when to initiate expensive and difficult intervention of ERT or other treatments.

When considering benefits of later-onset disorders, it is important to consider “alternative” potential benefits, such as elimination of the diagnostic odyssey, which is all too familiar with rare diseases such as LSDs. Other possible benefits of pre-symptomatic diagnosis may include empowering an individual to use their diagnosis information to make informed life decisions, enabling parents to use the NBS information to make informed reproductive decisions for future pregnancies, and identifying additional at risk or affected family members. It will be crucial to examine how these “alternative benefits” may challenge traditional purposes of NBS programs and the moral/ethical justification for compulsory screening.

7. Conclusion and future directions

As the therapeutic landscape expands and screening capabilities continue to accelerate, there is increasing interest in screening newborns for LSDs. The use of multi-tiered testing and post-analytic tools are likely to enhance screening accuracy, with an additional goal of being able to predict phenotypic severity, prognosis, and facilitate treatment decisions. Another important future direction is the data-driven exploration of the ethical issues associated with NBS for LSDs and other complex disorders.

There are ongoing efforts to address many of these concerns. For example, there is transparency about LSD NBS programmatic outcomes in several states and countries that have instituted live screening; this data is proving extremely valuable in assessing real word implementation issues, screening accuracy, overall and predicted phenotypic incidence, potential benefits, and areas for improvement [74,79–81,94]. Enhancing screening accuracy using novel biomarkers, bioinformatics, genomic and statistical approaches is the focus of several pilot NBS programs [91,92,96,101]. A new multi-disorder, identified, prospective pilot NBS program, ScreenPlus, is exploring the feasibility of using a multi-tiered approach to screen about 175,000 consented infants for 14 disorders, 13 of which are LSDs, while investigating the ethical issues associated with screening for these disorders. Efforts are underway to reevaluate ethical issues about predictive genetic testing in children in the context of the genomic age [92,112], outcomes of these efforts are likely to impact current NBS for LSDs.

In summary, the goal of newborn screening is to enhance the outcome of individuals with LSDs through early, pre-symptomatic detection. Widespread acceptance and implementation of NBS for LSDs will likely require improvements in screening and prognostic accuracy, advances in treatments, and evaluation of data assessing ethical concerns. Taken together, these efforts will provide critical, detailed data to help guide objective, ethically sensitive decision-making about NBS for LSDs.