Spectrum analysis of inborn errors of metabolism for expanded newborn screening in Xinjiang, China

Newborn screening

This is a retrospective study, from November 2018 to December 2021, a total of 41,690 newborn babies were referred for expanded newborn screening by MS/MS at the Neonatal Screening Center of Urumqi Maternal and Child Health Hospital in Xinjiang, China. The procedures were undertaken according to Technical Specifications for Neonatal Disease Screening (2010 Edition) (Ministry of Health of the People’s Republic of China, 2010) in using MS/MS for newborn screening. The study protocol was reviewed and approved by the Ethics Committee of Urumqi Maternal and Child Health Hospital (Ethical approval number: XJFYLL2022003). Written informed consents were obtained from all the infants’ parents.

Instruments and reagents

American Waters Xevo TQD tandem mass spectrometer, ALLSHENG microplate constant temperature incubation oscillator, Henya ultrasonic cleaning machine, Eppendorf pipette. The nonderivatization tandem mass spectrometry screening kit (NeoBase™ Non-derivatized MS/MS Kit) and supporting reagents, including extracts and mobile phase, were produced by Perkin Elmer. The purity of the organic solvents (such as methanol) used in the experiment were all high-performance liquid chromatography (HPLC) grade. The argon gas required for the MS crash tank was purchased from a local gas company, with a purity >99.999%.

Sample collection and delivery

According to the requirements of blood specification for neonatal genetic metabolic disease screening in the Technical Specifications for Neonatal Disease Screening (2010 Edition), sufficient post-lactation sampling should be given within 7 days after 72 h of birth. The requirements include at least three blood spots, each with a diameter of >8 mm, natural penetration of blood drops, consistent blood spots on the front and back of filter paper, no pollution, and no seepage blood ring. After natural drying, the blood tablets were stored at 2–8 °C, and the specimens were delivered to the screening center within 1 week.

MS/MS

MS/MS was used to detect amino acids and acylcarnitines with the NeoBase™ Non-derivatized MSMS Kit. First, 3.2 mm diameter perforating forceps were used to punch the dry blood spot sample and control, and a U-shaped plate provided in the kit was used to put the corresponding blood spots in each hole. Each plate was equipped with indoor quality control samples to ensure the correctness and stability of the experiment. Second, the internal standard and daily work solution were prepared in the amino acid dry powder standard bottle and the acylcarnitine dry powder standard bottle, respectively. Each extract was 1.0 ml and was shaken at room temperature until it was completely dissolved to make an amino acid standard solution and an acyl carnitine standard solution. These were stored at 2–8 °C (they can remain stable for 30 days at this temperature). On the day of the experiment, the extract and internal standard solution were prepared as the daily working solution according to the ratio of the kit instructions. Extraction involved 100 μL of the daily working solution being added to each well using a multichannel pipette, which was sealed with sticky film. Then, the microplate was placed into a constant temperature oscillation incubator at 45 °C (700 rpm) and incubated with shaking for 45 min. The supernatant was transferred onto the V-type bottom heat-resistant plate and the whole 96-well plate was wrapped up with aluminum film. Lastly, the 96-well plate was put into the sample room and the MS detection method was used for testing.

Genomic DNA sequencing

A kit was used to extract genomic DNA from 2 ml venous samples from the children and parents (CWE9600 Blood DNA Kit; CWBIOTECH). Qubit was used for DNA sample quality control. Samples were stored at −20 °C for testing. DNA library preparation was undertaken, which included end repair, connector connection, and polymerase chain reaction amplification. After library preparation, quantification and dilution was completed according to the kit instructions, and then sequencing was carried out through the sequencing platform. Sequencing bioinformatics analysis included using the Burrow-Wheeler Aligner (BWA) sequence comparison method and the quality control sequence contrast human genome reference sequence. In addition, annotation software was used. Specifically, the ANNOVAR annotation mutation site was used, which is a public mutation database and associated with multiple databases, such as genome, NHLBI GO Exome Sequencing Project (ESP6500), The Single Nucleotide Polymorphism Database (dbSNP), the Exome Aggregation Consortium (EXAC), and Human Gene Mutation Database (HGMD). This resource provides information on the mutation site in the normal population frequency, sequence conservation, mutations caused by coding amino acid changes, the location in the protein structure, and mutations that are predicted to be harmful. Finally, combined with the condition of the child, the mutations were interpreted for pathogenicity according to American College of Medical Genetics and Genomics criteria and guidelines (Richards et al., 2015), and the pedigree was verified by Sanger sequencing. In this study, the next-generation sequencing and Sanger sequencing verification were sent to Shanghai Xinhua Hospital, Beijing MyGenostics, and Hangzhou Bosheng Medical Laboratory.

Quality control

The indoor quality control products provided by the kit were used to ensure the accuracy of each experimental result. There were two kinds of indoor quality control products: low concentration value and high concentration value. Each experimental plate was equipped with two holes of high-quality control. The quality control value was recorded and used to generate an L-J quality control map: x ± 2SD was the warning limit and x ± 3SD was the runaway limit. Inter-room quality assessment of MS/MS screening of neonatal genetic metabolic diseases was organized by the Clinical Laboratory Center of the National Health Commission, and amino acid box acylcarnitine compartment quality assessment occurs every year, and all indicators were 100% qualified.

Statistical analysis

SPSS 23.0 software was used for data collation and statistical analysis. The t-test was used for the comparison between the two groups, and the percentile method was used for the non-normal distributed data. Values were considered statistically significant at P < 0.05.

Performance of expanded newborn screening

From November 2018 to December 2021, a total of 41,690 newborn babies were screened by MS/MS, including 21,463 males and 20,227 females; the percentage of preterm infants was 4.60% (1,915/41,690) and that of low-birth-weight infants was 2.74% (14,141/41,690). The median number of days for initial screening was 3 days after birth. The basic information of newborn screening is detailed in Table 1. The results of the screening of neonatal IEMs by MS/MS showed that a total of 1,457 newborn babies needed to be recalled, and the positive rate of primary screening was 3.49%, among which the number of successful recalls was 1,417 and the recall rate was 97.25%. After the recall review, the number of patients still remaining positive was 124, and the infants to be diagnosed needed further differential diagnosis and genetic testing. Finally, 36 newborn babies and one family member were diagnosed with an IEM, and the overall incidence of IEM in Xinjiang was 1:1,126 (37/41,690) (excluding one family, the overall incidence of neonatal IEM was 1:1,158). The incidence of amino acidemia, organic acidemia, and fatty acid oxidation disorder were 1:1,668 (25/41,690), 1:4,632 (9/41,690), and 1:13,896 (3/41,690), respectively (excluding one family, the incidence of abnormal neonatal fatty acid metabolism was 1:20,845). The overall positive predictive value was 29.03% (Table 2).

Incidence and disease spectrum of IEM

In this study, we confirmed a total of eight different IEMs, including one amino acidemia (25 cases; 69.44%), six organic acidemia (nine cases; 25.00%), and two fatty acid oxidation disorders (two cases; 5.56%). In total, 36 newborn babies were confirmed to have an IEM (Table 3). Figure 1 showed the process of MSMS screening and diagnosing positive patients.

All the confirmed cases of amino acidemia were hyperphenylalaninemia (HPA; incidence 1/1,668). These cases including 16 newborn babies with mild hyperphenylalaninemia (the average concentration of Phe was 150.18 μmol/L and Phe/Tyr was 3.26 μmol/L). Four cases had moderate phenylketonuria (PKU) and five cases had classical PKU (the average concentration of Phe was 647.95 μmol/L and the average concentration of Phe/Tyr was 8.01 μmol/L).

The diseases diagnosed for organic acidemia included methylmalonic acidemia (MMA; five cases), propionic acidemia (PA; one case), glutaric acidemia type I (GA-I; one case), isovaleric acidaemia (IVA; one case), and three-methylcrotonyl-coenzyme A carboxylase deficiency (MCCD; one case). MMA was the most common disease for organic acidemia (incidence 1/8; 338); three cases were pure MMA and two cases were MMA combined with homocysteinemia. The mean primary screening concentrations of C3, C3/C2, and C3/C0 in all patients with MMA were 14.12, 0.47, and 0.68 μmol/L, respectively. In addition to a significant increase in C5OH and its corresponding ratio during recall and follow-up, a decrease in C0 to 2.2 μmol/L was also observed in a case with MCCD. After clinical administration of special milk powder feeding and levocarnitine supplementation, C0 returned to normal and C5OH decreased compared with before. At present, the child has good growth and development, and no abnormal clinical manifestations.

A total of two cases had fatty acid oxidation disorder (incidence 1 in 20,845), including one case of primary carnitine deficiency (PCD) and one of short-chain acyl-CoA dehydrogenase deficiency (SCADD). Children with SCADD showed an increase of C4, C4/C2, and C4/C3 detected by MS/MS, and gaschromatography-mass spectrometry (GC-MS) results showed an increase in ethylmalonic acid. During the pedigree validation of the child with SCADD, the elder sister was found to carry two mutation sites in the same ACADS gene as this patient. The follow-up results showed that the children with SCADD and his sister had good intellectual and physical development, with normal clinical blood glucose monitoring and no abnormal clinical manifestations.

Genetic analysis of the confirmed cases

Among the 27 newborn babies who screened positive for Phe by MS/MS, the parents of one newborn baby refused gene sequencing; another family of a newborn baby had a PKU proband, Phe > 120 μmol/L, and Phe/Tyr > 2.0 in initial screening as well as in re-examination. The case could be diagnosed through biochemical methods and special dietary treatment was performed in the early stages of screening diagnosis; therefore, NGS gene sequencing was not performed. The remaining 25 screen-positive infants were tested by NGS sequencing, which detected a total of 22 different mutation sites of the PAH gene: c.158G > A was the most common, followed by c.688G > A, c.331C > T, c.1174T > A, as well as c.739G > A. The detailed gene mutations are shown in Table 4, no novel mutations were found in our study. According to the clinical manifestations and final diagnosis of disease, it was found that the infants with mutations in the locus c.158G > A, c.688G > A, c.331C > T, and c.1174T > A of the PAH gene showed mild to moderate hyperphenylalaninemia clinically.

Genetic analysis of the 11 infants with non-PKU IEMs showed that 91% (43/47) carried two mutated genes associated with specific disease. A total of 21 mutations were detected in eight genes associated with IEM, including 12 missense mutations, six frameshift mutations, two nonsense mutations, and one splice site mutation. According to the American College of Medical Genetics and Genomics classification criteria, 66.67% of variants (14/21) mutations were pathogenic, 28.57% of variants (6/21) mutations were likely pathogenic, and 4.76% of variants (1/21) mutations were of unknown significance.

Among the abnormalities in organic acid metabolism, five newborn babies were diagnosed with methylmalonic acidemia. Two of these cases were methylmalonic acidemia combined with homocysteinemia, carrying four types of mutations in the MMACHC gene, and studies have shown that c.609G > A may be the hot spot mutation carried in local patients with concurrent methylmalonic acidemia (Zhang et al., 2021). This mutation was also found in our study, but whether it is a hotspot mutation in this region cannot be determined and needs further verification with the accumulation of cases and data. Three cases were isolated methylmalonic acidemia, and carried six mutation types in the MMUT gene. In one case with confirmed isovaleric acidemia, only a possible pathogenic mutation site of the IVD gene was found, and the final diagnosis was made based on blood MS/MS values, combined with urinary GC-MS analysis and clinical manifestations. A patient with SCADD was confirmed by positive MS/MS for neonatal screening (persistently high C4 and C4/C3 levels in blood). Combining urine organic acid testing (increased urine ethylmalonic acid) and gene sequencing, we found that the elder sister also had the same ACADS mutation site c.989G > A (paternal source) and c.164C > T (maternal source). During clinical follow-up, neither the child nor his sister had clinical manifestations such as feeding difficulties, hypoglycemia, developmental delay, hypotonia, and epilepsy. The children also had good intellectual development, which is consistent with previous studies (Wang et al., 2020), as SCADD mostly does not have special clinical manifestations.