Microarray analysis can detect large chromosome imbalances identified by karyotyping as well as smaller alterations less than 10 Mb in size. However, detection rates of clinically significant copy number alterations (CNAs) have varied among studies. The present investigation was designed to demonstrate the usefulness of microarray testing compared with karyotyping for various clinical indications in prenatal diagnosis.
Prenatal samples from amniotic fluid, chorionic villi, fetal blood, or products of conception were examined for cytogenetic diagnosis using various microarrays targeted to known chromosomal syndromes. All data used in the analyses were obtained during the process of clinically approved microarray-based comparative genomic hybridization testing for routine patient care. A total of 5003 samples were tested for a variety of indications. Results were reported as normal with no clinically significant CNA, with or without benign CNAs identified; with a variant of uncertain significance (VOUS); or clinically significant (abnormal). A VOUS is defined as an alteration of unclear clinical relevance that has not been previously identified in a laboratory’s patient population, has not been reported in the literature, has not been found in databases, or does not contain any known disease-causing genes. Data from all cases resulting in abnormal or VOUS results were also captured. Prenatal cases were reviewed and categorized according to the result and indication for study. The stratification of categories included a known abnormal karyotype in which the family desired further characterization of the anomaly (n = 648), family history of a parent known to carry a chromosome rearrangement or imbalance (n = 62), fetal demise (n = 417), abnormal ultrasound (n = 2858), abnormal result on a first- or second-trimester screen (n = 77), other family history of a genetic condition (n = 487), advanced maternal age (n = 346), parental anxiety (n = 95), and other/not specified indications (n = 13). Cases referred for abnormal ultrasound results were further categorized according to the clinical phenotype, including anomalies in multiple organ systems (n = n = 808), anomalies in single organ systems (n = 1773), isolated soft markers (n = 77), other nonstructural anomalies (n = 134), and other/not specified (n = 66). Abnormal results were further stratified based on the size of the alteration. If the abnormality was an unbalanced translocation, the largest chromosomal segment affected by the translocation determined whether the case was placed in the ≥10-Mb or <10-Mb category.
Specimen types were cultured amniocytes (65%), direct amniotic fluid (7%), cultured chorionic villi (17%), direct chorionic villi (1%), fetal blood (0.5%), products of conception (9%), and DNA from unspecified sources (0.3%). The overall detection rate of clinically significant results was 5.3% (207/3876). Fetal demises had a significantly higher detection rate of clinically significant results (8.2%, 34/417) than other indications (5.3%, 207/3876). Of the 5003 prenatal specimens, 56.3% were referred with normal karyotypes, 13% had known abnormal fetal karyotypes at the time of array testing, 16% had karyotyping done concurrently with microarray testing, and 14.7% had an unknown status of karyotypes. Excluding the biased family history and fetal demise categories, the detection rate of clinically significant CNAs was 5.5% (140/2533) among cases with known normal karyotypes. These detection rates from the whole cohort represent the identification of clinically significant CNAs beyond those detected by karyotype analysis. Overall, 71% of the abnormalities were less than 10 Mb and not expected to be identified by conventional cytogenetic analysis; known microdeletion syndromes were detected in 35 of these cases. In addition, 48% (33/69) of cases with abnormalities detected by microarray of 10 Mb or greater were reported to have normal conventional karyotypes from the referral laboratories; 6% had revised, abnormal karyotypes following the receipt of array results. In 12 cases, disagreement between mosaic array results and karyotype could have resulted from placental mosaicism, previously undetected mosaicism, or cultural artifact. Combining all known abnormal karyotypes, 1.4% showed clinically significant CNAs elsewhere in the genome not associated with the known rearrangement.
Microarray-based comparative genomic hybridization allows prenatal identification of a much larger proportion of clinically relevant chromosomal abnormalities. Although significant CNAs were found among most categories of indication for study, they are more likely to be found in certain categories, demonstrating an increased diagnostic utility of microarray testing for particular subsets of patients. Large prospective studies using microarrays are necessary to assess the detection rates of chromosome abnormalities beyond those detected by karyotyping. Because the goal of invasive testing is to identify chromosome anomalies significant to fetal pathology, microarray testing should be considered a primary test for diagnosing fetal cytogenetic aberrations.
Signature Genomic Laboratories, PerkinElmer, Inc, Spokane, WA (L.G.S., M.P.D., J.C., A.M.B., J.W.E, J.B.R., B.S.T., B.C.B., J.A.R.); and Commonwealth Perinatal Services, Richmond, VA (A.J.F.).