Sequencing as a first-line methodology for cystic fibrosis carrier screening

Abstract

AbstractPurposeMedical society guidelines recommend offering genotyping-based\n cystic fibrosis (CF) carrier screening to pregnant women or women considering\n pregnancy. We assessed the performance of sequencing-based CF screening relative\n to genotyping, in terms of analytical validity, clinical validity, clinical\n impact, and clinical utility.MethodsAnalytical validity was assessed using orthogonal confirmation and\n reference samples. Clinical validity was evaluated using the CFTR2 database.\n Clinical impact was assessed using ~100,000 screened patients. Three screening\n strategies were compared: genotyping 23 guideline-recommended variants (“CF23”),\n sequencing all coding bases in CFTR (“NGS”),\n and sequencing with large copy-number variant (CNV) identification\n (“NGS\u2009+\u2009CNV”). Clinical utility was determined via self-reported actions of\n at-risk couples (ARCs).ResultsAnalytical accuracy of NGS\u2009+\u2009CNV was 100% for SNVs, indels, and\n CNVs; interpretive clinical specificity relative to CFTR2 was 99.5%. NGS\u2009+\u2009CNV\n detected 58 ARCs, 18 of whom would have gone undetected with CF23 alone. Most\n ARCs (89% screened preconceptionally, 56% prenatally) altered pregnancy\n management, and no significant differences were observed between ARCs with or\n without at least one non-CF23 variant.ConclusionModern NGS and variant interpretation enable accurate\n sequencing-based CF screening. Limiting screening to 23 variants does not\n improve analytical validity, clinical validity, or clinical utility, but does\n fail to detect approximately 30% (18/58) of ARCs.