Chromosomal Microarray Versus Karyotyping for Prenatal Diagnosis

Array-based molecular cytogenetic techniques have improved the detection of small genomic deletions and duplications not seen on karyotyping. These copy-number variants can be benign or pathogenic, depending on their location and genetic content. Whether microarray analysis reliably detects all chromosome abnormalities diagnosed on standard karyotyping and how often abnormal microarray results are not detected on karyotyping are unknown. This large, prospective study was performed at 29 prenatal centers to assess the capability of microarray analysis to diagnose common chromosome abnormalities and to determine the amount of additional data possible using microarray compared with standard karyotyping. Women presenting with a singleton fetus for chorionic villus sampling or amniocentesis for indications including advanced maternal age, a positive aneuploidy screening result, and ultrasound-detected anomalies were eligible. Each sample was split into 2 portions, one for karyotyping and the other for microarray analysis. Each microarray result was considered as true positive, true negative, false positive, or false negative relative to the karyotype findings. Karyotyping was considered the standard against which the chromosomal microarray was judged for identifying common autosomal and sex-chromosome aneuploidies. Karyotyping and microarray assays were performed according to standard protocols. Variants determined not to be of clinical importance were classified as “likely benign” and were not reported to the participant. The rest were determined to be of potential clinical significance and were reported to the participant. Adequate samples were obtained from 4406 of 6537 women screened. Samples for microarray were available from 4391 women (99.7%), and analysis was successful in 4340 cases (98.8%). The primary analysis included 4282 samples, in which 317 (7.4%) common autosomal and 57 (1.3%) sex-chromosome aneuploidies were identified by standard karyotyping. Microarray analysis identified all of these aneuploidies. All 22 unbalanced rearrangements also were identified by microarray. None of the apparently balanced rearrangements identified on karyotyping were identified by microarray analysis, suggesting that these rearrangements were truly balanced. Seventeen triploid samples (0.4%) were present, none of which were identified on microarray. On microarray, 1399 samples had copy-number variants, of which 1234 (88.2%) were classified as common benign. Thirty-five copy-number variants (0.9%) among the 3822 fetuses were on a predetermined list of pathogenic copy-number variants. Of 130 samples, 36 (27.7%) were likely benign, and 94 were considered of uncertain significance. Sixty-one (64.9%) of these 94 had sufficient clinical relevance to be reported to the participant. Overall, 96 (2.5%) of the 3822 fetal samples with normal karyotypes had a microdeletion or duplication of clinical significance. From fetuses with suspected growth or structural anomalies, 45 (6.0%) of the 755 samples had clinically relevant findings on microarray that were not found on karyotyping. Thirty-four (1.7%) of 1966 women without ultrasound-identified anomalies tested because of older age had a normal karyotype and a clinically relevant finding on microarray as did 12 (1.6%) of 729 women who tested positive on Down syndrome screening. Microarray analysis is equivalent to standard karyotype analysis for prenatal diagnosis of common aneuploidies and can be beneficial for prenatal testing. Patients should be aware of the risks of invasive testing, the frequency and severity of clinically relevant microarray findings, and the more limited identification of common aneuploidies obtainable by noninvasive screening. Still under study are how much incremental information should be sought by prenatal testing and how that information should be introduced into care.