Evaluation of X-Linked Adrenoleukodystrophy Newborn Screening in North Carolina

IMPORTANCE X-linked adrenoleukodystrophy (X-ALD) is a peroxisomal genetic disorder in which an accumulation of very long-chain fatty acids leads to inflammatory demyelination in the central nervous system and to adrenal cortex atrophy. In 2016, X-ALD was added to the US Recommended Uniform Screening Panel. OBJECTIVE To evaluate the performance of a single-tier newborn screening assay for X-ALD in North Carolina. DESIGN, SETTING, AND PARTICIPANTS This diagnostic screening study was of all newborn dried blood spot specimens received in the North Carolina State Laboratory of Public Health between January 2 and June 1, 2018, excluding specimens of insufficient quantity or quality. A total of 52 301 specimens were screened for X-ALD using negative ionization high-performance liquid chromatography tandem mass spectrometry to measure C24:0and C26:0-lysophosphatidylcholine concentrations. Sanger sequencing of the adenosine triphosphate–binding cassette subfamily D member 1 (ABCD1) gene was performed on screen-positive specimens. EXPOSURES A medical and family history, newborn physical examination, sequencing of ABCD1 on dried blood spot samples, and plasma analysis of very long-chain fatty acids were obtained for all infants with screen-positive results. MAIN OUTCOMES AND MEASURES The prevalence of X-ALD in North Carolina and the positive predictive value and false-positive rate for the first-tier assay were determined. RESULTS Of 52 301 infants tested (47.8% female, 50.6% male, and 1.7% other or unknown sex), 12 received screen-positive results. Of these 12 infants, 8 were confirmed with a genetic disorder: 3 male infants with X-ALD, 3 X-ALD–heterozygous female infants, 1 female infant with a peroxisome biogenesis disorder, and 1 female infant with Aicardi-Goutières syndrome. Four infants were initially classified as having false-positives results, including 3 female infants who were deemed unaffected and 1 male infant with indeterminate results on confirmatory testing. The positive predictive value for X-ALD or other genetic disorders for the first-tier assay was 67%, with a false-positive rate of 0.0057%. CONCLUSIONS AND RELEVANCE This newborn screening pilot study reported results on 2 lysophosphatidylcholine analytes, identifying 3 male infants with X-ALD, 3 X-ALD–heterozygous female infants, and 3 infants with other disorders associated with increased very long-chain fatty acids. These results showed successful implementation in a public health program with minimal risk (continued) Key Points Question What is the analytical and clinical validity of a mass spectrometric method evaluating very long-chain fatty acyl-lysophosphatidylcholine species for the detection of X-linked adrenoleukodystrophy among newborns in North Carolina? Findings In this newborn diagnostic screening study of 52 301 dried blood spot specimens, 3 male infants were confirmed to have X-linked adrenoleukodystrophy, 3 female infants were identified as heterozygous for X-linked adrenoleukodystrophy (carriers), 1 female infant had a peroxisome biogenesis disorder, 1 female infant had Aicardi-Goutières syndrome, 1 male infant received an indeterminate diagnosis, and 3 female infants had false-positive results. Meaning The newborn screening results suggested that detecting increased concentrations of very longchain fatty acyl-lysophosphatidylcholine species in dried blood spots is an effective method for identifying infants with X-linked adrenoleukodystrophy. Author affiliations and article information are listed at the end of this article. Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(1):e1920356. doi:10.1001/jamanetworkopen.2019.20356 (Reprinted) January 31, 2020 1/12 Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 09/22/2021 Abstract (continued) to the population. The findings will support other state laboratories planning to implement newborn screening for X-ALD and related disorders.continued) to the population. The findings will support other state laboratories planning to implement newborn screening for X-ALD and related disorders. JAMA Network Open. 2020;3(1):e1920356. doi:10.1001/jamanetworkopen.2019.20356


Introduction
The X-linked adrenoleukodystrophy (X-ALD) disorder is a peroxisomal disorder caused by a deficiency of adenosine triphosphate-binding cassette transporter protein (adrenoleukodystrophy protein) encoded by the adenosine triphosphate-binding cassette subfamily D member 1 (ABCD1) gene (OMIM 300371). 1,2 The protein transports very long-chain acyl-CoA esters into peroxisomes, the site of very long-chain fatty acid (VLCFA) beta-oxidation. 3 In patients with X-ALD, VLCFAs accumulate in all tissues, primarily affecting the central nervous system and adrenal cortex. The 3 main phenotypes of X-ALD are Addison disease (adrenal insufficiency), adrenomyeloneuropathy, and cerebral adrenoleukodystrophy (CALD). 4 Individuals with X-ALD are typically asymptomatic at birth; however, those with the severe phenotype-the childhood form of CALD-typically present between 2.5 and 10 years of age. 4 Without treatment, patients with childhood CALD rapidly decline, and death typically occurs 2 to 4 years after onset of symptoms. 4 Hematopoietic stem cell transplantation is the recommended treatment of patients with childhood CALD and is performed at the first sign of brain pathology as detected on magnetic resonance imaging (MRI). [4][5][6] The MRI is rated using a Loes score, which indicates the severity of brain lesions. 7 Hematopoietic stem cell transplantation can arrest the progression of cerebral demyelination if the patient is treated during an asymptomatic or early stage. 6,8 Most male infants diagnosed as having X-ALD also have adrenal insufficiency and can experience adrenal function impairment as early as 6 months of age that may lead to morbidity and mortality. 9 Therefore, it is important to identify infants as early as possible and to monitor their serum adrenocorticotropic hormone and cortisol levels in an effort to institute lifesaving hormone replacement therapy if abnormalities are observed.
The Advisory Committee on Heritable Disorders in Newborns and Children reviewed evidence demonstrating effective laboratory technologies available for X-ALD newborn screening (NBS) as well as the benefits of early identification and treatment on health outcomes. Based on its interpretation of the data, this advisory committee recommended the addition of X-ALD to the Recommended Uniform Screening Panel in August 2015, a move that the secretary of the US Department of Health and Human Services approved in February 2016. After X-ALD was added to

Methods Samples
In this diagnostic NBS study, X-ALD screening was performed prospectively between March 5 and July 31, 2018, using 52 301 identifiable consecutive specimens received at NCSLPH between January 2 and June 1, 2018. This report adhered to the Standards for Reporting of Diagnostic Accuracy (STARD) reporting guideline. The RTI and UNC-CH institutional review boards approved the design and methods and determined that this pilot study did not meet criteria for human subjects research; therefore, informed participant consent was waived.

Follow-up of Screen-Positive Specimens
Information about screen-positive cases was referred to a genetic counselor at UNC-CH for follow-up. Subsequent patient care was provided according to the follow-up protocol presented in Figure 2.

Statistical Analysis
Descriptive statistics, such as means and SDs, and graphical representations of the data were derived using Microsoft Excel (Microsoft Corp). Demographic categories were reported by frequency as well as by percentages. Positive predictive value was calculated by dividing the true-positive cases by the sum of the positive and negative cases. The false-positive rate was calculated by dividing the falsepositive cases by the true-positive and true-negative cases.

Specimen Testing Results
The demographic data found on the NBS card, including sex, birth weight, and age at infant at DBS collection, of all the specimens sent to the NCSLPH during the study period are provided in Table 1.    Three male infants were confirmed to have X-ALD, and all received follow-up visits with members of the genetics, endocrinology, and neurology departments at UNC-CH. The VLCFA analysis revealed increased C26:0, C26:1, and C24:0 concentrations and an increase in the ratios of C24:0 to C22:0 and of C26:0 to C22:0 in all cases. Two infants had examinations with results that were within references ranges (patients 8 and 12), but 1 had poor weight gain and brisk reflexes   Table 2. AGS indicates Aicardi-Goutières syndrome; F, female; and M, male.  (patient 1). All had serum electrolyte and cortisol levels within reference ranges, but 2 infants had increased adrenocorticotropic hormone levels at 6 months (patient 8) and 9 months (patient 12) of age. To date, those 3 infants remain asymptomatic for neurological involvement. All 3 families reported no known family history of X-ALD. However, 1 family (patient 8) described nonspecific findings, including attention-deficit/hyperactivity disorder and learning difficulties, in a maternal uncle, although no diagnosis could be inferred from the information provided. Carrier testing was offered to the patients' mothers through genetic testing and to brothers by using VLCFA testing. Only 2 families pursued recommended testing during the pilot study.

JAMA Network Open | Pediatrics
Three female infants were diagnosed as heterozygous for X-ALD (carriers), and all families One male infant had a slight increase in the C24:0 to C22:0 ratio (patient 4), which was interpreted by the diagnostic laboratory as possible liver dysfunction. The patient was examined at 6 months of age and had prolonged neonatal hyperbilirubinemia but no evidence of liver dysfunction; the liver function test results were within reference ranges. An isolated increase in the C24:0 to C22:0 ratio is not consistent with a diagnosis of X-ALD or of a peroxisomal biogenesis disorder; thus, this case is likely a false-positive. However, the parents were advised to return in 2 years for reevaluation.
Three infants, all female, were categorized as having false-positive screening results, with no pathogenic variants, likely pathogenic variants, or VUS detected in ABCD1 and normal levels of VLCFAs (patients 2, 9, and 10). One infant (patient 10) had a slight increase in the C26:0 to C22:0 ratio that was not considered significant. These patients had no further follow-up.

Performance of Screening Assay
The newborn screening study identified 3 male infants with X-ALD and 3 female infants heterozygous for X-ALD. The incidence was estimated to be 1 in 8717 births, a slightly higher frequency than that observed by the New York NBS program 18 but lower than the frequency reported by the Minnesota NBS program. 19 The positive predictive value for the first-tier assay was 67%, and the false-positive rate, which was defined as cases sent to follow-up and diagnostic testing indicating that the infant did not have X-ALD or a related disorder, was 0.0057%. Only 1 borderline case was included in the falsepositive rate. All other borderline cases were not sent to follow-up and were not included in this calculation.

Evaluation of the Screening Algorithm
This pilot newborn screening study used the HPLC-MS/MS method in negative ion mode because it is more selective than alternative published methods; and the initial screening algorithm did not require adjustment during the course of the study. However, the build-up of blood spot matrix in the instrumentation caused a revision in maintenance procedures, including frequent replacement of the analytical and guard columns and mass spectrometer consumable parts. Alternative methods for X-ALD screening combines the analysis of C26:0-LPC with amino acid and acylcarnitine 20 or lysosomal enzyme analysis 21  Of the 45 borderline results, an additional specimen was received for 27 infants, and most of the results from the second specimen were normal. The borderline cutoff value was set conservatively to ensure that no infants with X-ALD were missed, but the evidence from this study suggested that the specimens with borderline results were at a low risk for X-ALD. In the future, with more population screening data, it may be possible to eliminate this category.
The present study also found that the screening target, male infants with X-ALD, showed increases in both analytes, suggesting that measurement of C24:0-LPC in addition to C26:0-LPC is valuable in discerning specimens from patients with X-ALD or other peroxisomal disorders, but this additional analyte should not be considered the primary marker. At the time of the study, no stable isotope-labeled internal standard was available for C24:0-LPC; however, d 4 -C24:0-LPC is now commercially available, and additional studies are needed to evaluate this internal standard on the accuracy and precision of C24:0-LPC quantification.

Short-term and Long-term Follow-up
The follow-up protocol included a thorough family history, and carrier testing was offered to mothers because of the reproductive risks and because 80% of female infants develop signs of neurological dysfunction by 60 years of age. 22 Male siblings were also offered testing because there is a 50% risk that these male siblings will have X-ALD if their mother is a carrier. The parents of female carriers and 1 mother of a male infant with X-ALD did not pursue testing for themselves. The families may not have understood the full implications of the diagnosis despite receiving information, were not planning for more children, or were dissuaded by the cost of testing (not included in the pilot study) or lack of insurance. Additional materials may need to be developed to help families understand the various effects X-ALD may have and how to share information with their extended family.
Long-term follow-up is an important component of a complete NBS system and is particularly important for newborns identified with X-ALD. In total, 40% to 45% of male infants with X-ALD have no symptoms until adulthood. 4 There is no published correlation between genotype and phenotype or age at onset; 23-25 therefore, routine clinical monitoring is critical for the initiation of timely and effective interventions and for assessing the clinical course of cases identified through NBS. The New York State NBS Program published recommendations for follow-up of presymptomatic boys in childhood. 26 These recommendations have been adopted by most states screening for X-ALD and this pilot study. The adapted protocol used to follow-up with patients identified in this study includes monitoring of adrenal function with tests for serum adrenocorticotropic hormone and cortisol every 6 months until 18 years of age and then annually, and a neurology evaluation, including brain MRI without contrast annually, up to 3 years of age, every 6 months from 3 to 10 years of age, and annually from 11 to 18 years of age. Additional brain MRI may be added outside the proposed time frame if there are clinical concerns because of history or neurologic examination findings. In conjunction with follow-up recommendations for presymptomatic male infants, clinical referral networks with expertise in medical genetics, neurology, endocrinology, and the provision of hematopoietic stem cell transplantation need to be established to coordinate follow-up for individuals with presymptomatic X-ALD. 27 Coordinating follow-up care with 3 subspecialties may be challenging for some states because of the lack of available specialists and funding; however, the coordination will facilitate more effective care and proper surveillance of these patients. These centers could potentially be used to track long-term outcomes of patients, which are critical to understanding which subtypes of X-ALD are identified through NBS and whether early identification and treatment of childhood CALD is successful. It will also provide insight into the natural progression of adrenal and neurologic involvement and allow refinement of screening and follow-up protocols to include guidance on infants who have disorders other than X-ALD.

Limitations
The testing method is intended for screening and should be accompanied by diagnostic testing and medical evaluation. This method cannot predict disease severity, and long-term follow-up data were not collected to evaluate long-term outcomes of identified infants.

Conclusions
The North Carolina pilot study implemented a screening program that detected true screen-positive specimens with a high degree of analytical specificity and identified 3 male and 3 female newborns with X-ALD as well as other disorders. Future work is needed to evaluate the long-term data on patients identified with X-ALD through NBS to understand the clinical presentation, course of the condition, effectiveness of early treatment, ability of the health care system to provide follow-up care, and effect on families.