Validation of an Expanded Carrier Screen that Optimizes Sensitivity via Full-Exon Sequencing and Panel-wide Copy Number Variant Identification

Gregory J. Hogan, Valentina Vysotskaia, Kyle A. Beauchamp, Stefanie Seisenberger, Peter Grauman, Kevin Haas, Sun Hae Hong, Diana Jeon, Shera Kash, H. Henry Lai, Laura M. Melroy, Mark Theilmann, Clement Chu, Kevin Iori, Jared Maguire, Eric A. Evans, Imran S. Haque, Rebecca Mar‐Heyming, Hyunseok P. Kang, Dale MuzzeyClinical 2018 · DOI

Abstract

Purpose: By identifying pathogenic variants across hundreds of genes, expanded carrier screening (ECS) enables prospective parents to assess risk of transmitting an autosomal recessive or X-linked condition. Detection of at-risk couples depends on the number of conditions tested, the diseases’ respective prevalences, and the screen’s sensitivity for identifying disease-causing variants. Here we present an analytical validation of a 235-gene sequencing-based ECS with full coverage across coding regions, targeted assessment of pathogenic noncoding variants, panel-wide copy-number-variant (CNV) calling, and customized assays for technically challenging genes.

Methods: Next-generation sequencing, a customized bioinformatics pipeline, and expert manual call review were used to identify single-nucleotide variants, short insertions and deletions, and CNVs for all genes except FMR1 and those whose low disease incidence or high technical complexity precludes novel variant identification or interpretation. Variant calls were compared to reference and orthogonal data.

Results: Validation of our ECS data demonstrated >99% analytical sensitivity and >99% specificity. A preliminary assessment of 15,177 patient samples reveals the substantial impact on fetal disease-risk detection attributable to novel CNV calling (13.9% of risk) and technically challenging conditions (15.5% of risk), such as congenital adrenal hyperplasia.

Conclusion: Validated, high-fidelity identification of different variant types—especially in diseases with complicated molecular genetics—maximizes at-risk couple detection.

Introduction

Carrier screening is a vital tool for prospective parents, allowing them to identify the risk of passing autosomal recessive or X-linked conditions to their children. Historically, these screenings were limited to specific ethnicities or a narrow set of common mutations. However, the advent of next-generation sequencing (NGS) has enabled Expanded Carrier Screening (ECS), which evaluates hundreds of genes simultaneously regardless of the patient's ancestry.

Despite the breadth of NGS, two significant technical hurdles have historically limited the effectiveness of ECS: the detection of large copy-number variants (CNVs) and the accurate sequencing of "technically challenging" genes. This study validates the Counsyl Foresight™ Carrier Screen, an ECS designed to address these gaps through full-exon sequencing, panel-wide CNV identification, and specialized bioinformatics for complex genomic regions.

Methodology: A Multi-Layered Approach

The validated screen utilizes a hybrid capture-based enrichment and sequencing pipeline on the Illumina HiSeq 2500 platform. To maximize sensitivity, the methodology incorporates several specialized components:

  • Panel-wide CNV Calling: Rather than focusing solely on single-nucleotide variants (SNVs), the pipeline uses custom software to analyze read-depth across nearly every gene on the panel. This allows for the detection of large deletions and duplications that standard NGS pipelines often miss.
  • Specialized Bioinformatics: Certain genes, such as SMN1 (Spinal Muscular Atrophy) and CYP21A2 (Congenital Adrenal Hyperplasia), are notoriously difficult to sequence due to the presence of highly similar pseudogenes. The team developed custom algorithms to distinguish between functional genes and these non-functional "twins."
  • Targeted Assays: For conditions like Fragile X Syndrome (FMR1), which involve CGG repeat expansions that are incompatible with standard NGS, the screen integrates PCR and capillary electrophoresis (CE) workflows.

Validation and Datasets

The analytical accuracy of the screen was tested using a diverse array of samples:

  • Reference Standards: 91 samples from the 1000 Genomes Project and 70 cell lines from the Coriell repository with known pathogenic variants.
  • Clinical Confirmation: 115 mutation-positive blood and saliva samples previously verified via gold-standard methods like Sanger sequencing, qPCR, or MLPA.
  • In Silico Stress-Testing: To validate the detection of rare CNVs, the researchers simulated over 250,000 synthetic CNVs to evaluate the sensitivity of the detection algorithms.
  • Clinical Cohort: A preliminary assessment of 15,177 de-identified patient samples was conducted to model the real-world impact of the screen on disease-risk detection.

Key Results and Clinical Insights

The validation demonstrated exceptional performance, with >99.99% sensitivity and specificity for SNVs and short insertions/deletions (indels). The CNV detection system identified 100% of the empirical CNVs in the validation set, and in silico testing showed 99.9% aggregate sensitivity for CNVs in high-priority genes like CFTR and DMD.

The Impact of CNVs and Challenging Genes

The study highlights that a "one-size-fits-all" NGS approach significantly underestimates risk. The inclusion of panel-wide CNV calling accounted for 13.9% of the total fetal disease risk detected. Furthermore, technically challenging conditions (such as Alpha Thalassemia and CAH) accounted for an additional 15.5% of the total risk. Without these specialized detection methods, nearly 30% of at-risk couples might be missed.

At-Risk Couple Detection

For recessive diseases, the detection rate for couples scales as the square of the individual detection rate. By improving individual sensitivity from 80% to 99%, the couple detection rate increases from 64% to 98%. Based on the clinical cohort, the authors estimate that approximately 1 in 300 pregnancies would be affected by at least one serious condition covered by the 176-gene "Universal" panel.

Discussion and Limitations

While the screen provides industry-leading sensitivity, some limitations remain. For a small subset of genes with extreme technical complexity, such as FMR1, the test targets specific known mutations rather than scanning the entire gene. Additionally, while multi-exon CNV detection is near-perfect, the sensitivity for single-exon deletions across the entire panel (excluding CFTR and DMD) was lower, at 81.8%.

As genomic databases grow, the authors anticipate that in silico simulation will become a standard requirement for validating NGS panels, particularly for rare variants where clinical samples are scarce. This study confirms that high-fidelity identification of diverse variant types—especially in molecularly complex diseases—is essential for maximizing the detection of at-risk couples and providing accurate reproductive insights.

Summary updated

Abstract

Validated high-fidelity identification of different variant types-especially for diseases with complicated molecular genetics-maximizes at-risk couple detection.
BibTeX
@inproceedings{hogan2018validation,
  author = {Gregory J. Hogan and Valentina Vysotskaia and Kyle A. Beauchamp and Stefanie Seisenberger and Peter Grauman and Kevin Haas and Sun Hae Hong and Diana Jeon and Shera Kash and H. Henry Lai and Laura M. Melroy and Mark Theilmann and Clement Chu and Kevin Iori and Jared Maguire and Eric A. Evans and Imran S. Haque and Rebecca Mar‐Heyming and Hyunseok P. Kang and Dale Muzzey},
  title = {{Validation of an Expanded Carrier Screen that Optimizes Sensitivity via Full-Exon Sequencing and Panel-wide Copy Number Variant Identification}},
  booktitle = {Clinical},
  year = {2018},
  doi = {10.1373/clinchem.2018.286823},
}

Loading related content...

Figure