Clinical Chemistry - Case Study
Student Discussion Document (pdf)
Qiliang Ding, Kyle T. Salsbery, Noemi Vidal-Folch, Devin Oglesbee, and Linda Hasadsri
A pregnant woman was referred to our laboratory for prenatal GBA1 full-gene analysis after both she and her reproductive partner (hereafter referred to as the mother and father, respectively) were identified as Gaucher disease carriers by an external laboratory, which utilizes a hybridization capture-based next-generation sequencing (NGS) assay, followed by long-range polymerase chain reaction PCR (LR-PCR) and/or Sanger sequencing as needed. The father was reportedly positive for the GBA1 (NM_000157.4):c.1226A>G (p.Asn409Ser) pathogenic variant. The mother’s report, however, only described her as positive for a pathogenic gene conversion of GBA1. After contacting the external laboratory, they verbally clarified that she carried the pathogenic RecNciI allele, defined as 3 single-nucleotide variants (SNVs) in cis (c.1448T>C, c.1483G>C, and c.1497G>C) in exon 10 of GBA1.
The ordered service from our laboratory for this case was LR-PCR followed by nested Sanger sequencing of GBA1 exons. LR-PCR primers were designed in uniquely mappable regions to avoid pseudogene interference, yielding an 8960 bp amplicon. We first tested the parental specimens via this assay to confirm the ability to detect the relevant variants. The paternal c.1226A>G variant was identified. However, none of the RecNciI-defining SNVs, or any other pathogenic variant, were detected in the mother. Notably, this assay had previously detected the RecNciI allele, supporting its analytical sensitivity.
Given the discrepant result we obtained from what was reported by the external laboratory, and because Sanger sequencing cannot detect copy-number variants, we performed GBA1 short-read NGS, including copy-number analysis, on the maternal specimen as follow-up. Briefly, this assay uses Integrated DNA Technologies xGen Exome probes for whole-exome capture, followed by Illumina NGS, with tertiary analyses limited to GBA1. Copy-number analysis was conducted by normalizing the sample’s coverage relative to concurrently sequenced samples. The mean coverage of GBA1 exons for this sample was 233.0x, with no regions below the minimum requirement of 20x. This assay also did not detect any of the RecNciI-defining SNVs (short-read NGS). Nonetheless, it did reveal a pathogenic deletion of GBA1 exons 10–11.
Because compound heterozygosity for a RecNciI allele and a deletion of exons 10–11 would result in perinatal lethal Gaucher disease or be incompatible with life, we interpreted the deletion detected in the mother by our short-read NGS assay as representing the same variant reported by the external laboratory. After confirming the ability to detect both parental variants, we tested the fetal specimen using both Sanger sequencing and short-read NGS. Fortunately, neither pathogenic variant was detected in the fetus.
We further characterized the maternal specimen using Oxford Nanopore Technologies long-read sequencing. Interestingly, this analysis identified a RecNciI allele not generated by gene conversion but by a large structural variant, specifically an approximately 20.6 kb deletion, that spans the region between GBA1 and GBA1LP (GBA1LP is also known as GBAP1). This deletion, likely mediated by nonallelic homologous recombination (NAHR), produced a fusion gene composed of the first nine exons of GBA1 and the 3’-exons of GBA1LP that correspond to GBA1 exons 10–11. According to the Human Genome Variation Society’s 3’-rule, the deletion coordinates were chr1:155 215 100–155 235 726 (GRCh38/hg38). This finding reconciles the discrepancy between our short-read NGS assay and the external laboratory’s report.