Introduction

Congenital adrenal hyperplasia (CAH, OMIM #201910) is a group of autosomal recessive genetic disorders characterized by deficiencies in multiple enzymes involved in the adrenal steroidogenic pathway, leading to a variety of complex hormonal imbalances (Auer et al., 2023). CAH exhibits high clinical and genetic heterogeneity, which severely impairs patients’ quality of life. Clinically, more than 95% of CAH cases are caused by 21-hydroxylase deficiency (21-OHD), which disrupts the synthesis of cortisol, aldosterone, and androgens. Based on the severity of enzymatic deficiency, 21-OHD can be categorized into three subtypes: classic salt-wasting (SW), classic simple virilizing (SV), and non-classic congenital adrenal hyperplasia (NCCAH) (Merke and Auchus, 2020; Auer et al., 2023). The SW subtype, resulting from deficiencies in both cortisol and aldosterone, presents the most severe phenotype, including hypoglycemia and acute adrenal crisis; if left untreated, it can be fatal in the neonatal period. The SV subtype largely preserves mineralocorticoid synthesis, leading to a less severe clinical presentation; however, under conditions of major physical stress (e.g., trauma and surgery), the risk of salt loss is approximately 10%.The NCCAH subtype retains 20%–80% of in vitro enzyme activity, which may not compromise cortisol synthesis but can cause variable overproduction of adrenal androgens. In addition to 21-OHD, other rare forms of CAH exist: variants in the CYP11B1, CYP17A1, HSD3B2, and StAR genes, which are involved in steroid biosynthesis, can also lead to CAH (Miller, 2018). In China, the incidence of classic CAH is approximately 1 in 23,024 live births, while the incidence of non-classic CAH is considerably higher, at around 1 in 1,000 live births (Falhammar and Nordenström, 2015; Miller, 2018). The carrier frequency of classic CAH alleles in the general population is approximately 1 in 60 (Speiser et al., 2018).

Newborn screening (NBS) is a successful and comprehensive public health initiative that enables the effective identification of infants at high risk of rare childhood-onset disorders and the timely administration of therapeutic interventions prior to disease manifestation. NBS for CAH, which detects elevated 17α-hydroxyprogesterone (17α-OHP) concentrations via fluoroimmunoassay, can effectively identify the severe, life-threatening SW and SV subtypes. However, it fails to detect the less severe SV and non-classic (NC) subtypes due to their obscure phenotypes and marginally increased 17α-OHP levels (Hayashi et al., 2017). Furthermore, 17α-OHP concentrations are easily influenced by birth weight and gestational age at delivery, resulting in a false positive rate as high as 9.4% (Hayashi et al., 2017). Therefore, the development of effective screening techniques is particularly crucial.

With the rapid development of next-generation sequencing (NGS) technology and its application in the molecular diagnosis of genetic diseases, new opportunities have been provided for DNA-based neonatal genetic screening. The CYP21A2 gene, which encodes 21-hydroxylase, is located at the 6p21.3 locus on the short arm of chromosome 6 and consists of 10 exons. Its non-functional pseudogene, CYP21A1P, is tandemly arranged with CYP21A2, with an approximate distance of 30 kilobases (kb) between the two (Witchel, 2017). These two genes share a high degree of nucleotide sequence homology, with 98% homology in exonic regions and 96% homology in intronic regions (Carvalho et al., 2021). In addition, CYP21A2 and CYP21A1P, together with their adjacent genes, constitute the bimodular RCCX haplotype (RP1-C4A-CYP21A1P-TNXA-RP2-C4B-CYP21A2-TNXB) (Saraf et al., 2024). The high homology and tandem repeat sequences between functional genes (RP1, CYP21A2, and TNXB) and their corresponding pseudogenes (RP2, CYP21A1P, and TNXA) significantly increase the incidence of gene conversion, deletion, duplication, and non-functional chimeric gene formation during meiosis (El-Maouche et al., 2017). Consequently, approximately 75% of CYP21A2 variants—particularly single nucleotide variations (SNVs) and small insertions and deletions (indels)—originate from the non-functional pseudogene via microconversion. The remaining 25% are large-fragment deletions/duplications and chimeric genes (CYP21A1P/CYP21A2 chimeras or TNXA/TNXB chimeras (Chen et al., 2012). However, current NGS-based genotyping assays are not suitable for CYP21A2 genotyping due to their short read lengths and relatively low analytical sensitivity in highly homologous sequences and intergenic recombination events.

Long-read sequencing (LRS), also referred to as third-generation sequencing technology, is a novel nucleotide sequencing approach that can directly detect full-length genes and their flanking regions, as well as discriminate between highly homologous genes. Furthermore, LRS is capable of identifying the cis-trans relationships of multiple variants and detecting deletions/duplications and large-fragment conversions. Recent studies have demonstrated that LRS is a powerful tool for the accurate genetic diagnosis of CAH (Liu et al., 2022; Li et al., 2023). In the present study, LRS-based CAH genotyping was integrated into the NBS program to systematically evaluate its clinical application value in neonatal CAH screening.