P–558 SNP array versus karyotype for prenatal diagnosis in fetuses with abnormal ultrasound: a systematic review and meta-analysis

Abstract

Abstract Study question Does Single nucleotide polymorphism (SNP) array provide a diagnostic advantage over conventional karyotype in prenatal diagnosis for fetuses with an abnormal ultrasound? Summary answer SNP array in the prenatal setting provides an incremental diagnostic yield over karyotype and the diagnostic accuracy is comparable with combined SNP array and karyotype What is known already Single nucleotide polymorphism and comparative genomic hybridization based arrays (aCGH) are the two chromosomal microarray (CMA) platforms available. Guidelines which recommend offering CMA instead of karyotyping for prenatal diagnosis are mainly based on studies that compared aCGH with karyotype. There is a paucity of reviews that critically appraise the role of SNP array as a prenatal diagnostic tool. We decided to estimate the incremental yield of SNP array over karyotype in detecting chromosomal abnormalities, and to determine the diagnostic accuracy of SNP alone compared with SNP array and karyotype in combination for prenatal diagnosis in fetuses with an abnormal ultrasound. Study design, size, duration We conducted a systematic review of studies comparing SNP array with karyotype for prenatal diagnosis in fetuses with an abnormal ultrasound. We performed a literature search in the electronic databases of EMBASE, PubMed, CENTRAL, CDSR, SCOPUS and Web of science for relevant studies published in the English language between January1996 and May 2020. We also hand searched the referenced list of included studies and performed a google search for grey literature to identify potential studies. Participants/materials, setting, methods The study population was women undergoing prenatal diagnosis for abnormal fetal ultrasound. Studies in which SNP array and karyotyping had been used in fetuses with abnormal ultrasound and which allowed for a 2 x 2 data extraction table were included. We estimated the incremental yields for SNP array over karyotype. For determining the diagnostic accuracy, we considered SNP array alone as the index test and combined karyotype & SNP array as the reference standard. Main results and the role of chance We included six studies for quantitative analysis. After pooling results, incremental yield of SNP array over normal karyotype was 10% (95% confidence interval, CI 4 to 16%) in fetuses with abnormal ultrasound while incremental yield of karyotype over SNP array was 1% (95% CI 0 to 2%). The agreement between SNP array and karyotype was 92%. Variant of uncertain significance (VUS) rates ranged from 4–8%. For SNP array alone versus combined SNP array and conventional karyotype, pooled sensitivity and specificity was 0.96 (95% CI 0.91 to 0.99) and 1.00 (95% CI 0.99 to 1.00) respectively. The Area under curve (AUC) was 0.99 indicating the discriminating ability of the SNP array was very high to identify the fetus with chromosomal abnormalities. Limitations, reasons for caution Only studies published in English were included. There was some degree of heterogeneity in inclusion criteria for the included studies. Wider implications of the findings: The current study suggests SNP array alone can replace conventional karyotype for prenatal diagnosis in fetus with an abnormal ultrasound. Limitations to adoption of SNP testing might include the requirement of requisite expertise to interpret the results and counsel patients appropriately, especially with the propensity of SNPs to identify VUS. Trial registration number Not applicable