Chromosomal Microarray Analysis in Pregnancy Loss: Is It Time for a Consensus Approach?

Affecting up to 10% to 15% of all clinically recognized pregnancies, pregnancy loss has many causes. Half of first-trimester losses come about due to chromosomal abnormalities and then decline in frequency through pregnancy, ultimately reaching a rate of 5% in stillbirths. Recurrent pregnancy loss occurs in 1% to 2% of couples, and identifying the genetic etiology of these losses can provide psychological benefits, eliminate other testing needs, and establish a risk of recurrence. As benefits of performing genetic testing on products of conception (POCs) are clear, various groups have published practice guidelines (American College of Obstetricians and Gynecologists, European Society of Human Reproduction and Embryology, American Society for Reproductive Medicine, and Royal College of Obstetricians and Gynaecologists). However, despite the consensus for the utility of testing on POCs, requirements for these tests vary. Most appropriate testing methodology, obstetric history, and gestational age are all variations that affect the benefit or utility of these tests. As there is a lack of consensus on how to use chromosomal microarray analysis (CMA) in assessment of POCs, this study assessed POC CMA data with available karyotype data and determined the positive CMA rate for POCs. The authors also included within this data set a comprehensive correlation of CMA correlations with pregnancy phenotype, gestational age, and prior prenatal screening results. Finally, the study also assessed the impact CMA results from early losses may have on clinical care.This cohort included 323 POC samples processed by the Brigham and Women's Hospital Center for Advanced Molecular Diagnostics in the Clinical Cytogenetics Laboratory over a consecutive 42-month period. Collected variables included estimated gestational age at time of pregnancy loss, maternal age, pregnancy phenotype results of any orthogonal testing on POC tissue, noninvasive prenatal screen results, CMA findings, and available parental karyotypes. All cases recorded gestational age except 1 sample lacking documentation. Categories of phenotypic data included lymphatic malformations, anatomical abnormalities of the fetus or placenta, twin or higher-level multiple gestations, environmental complications, amniotic fluid disorders, or the combined influence of any of the above. A lack of definitive phenotypic data excluded 3 cases from the analysis. Analyses of CMA results were categorized as normal, abnormal, or variants of uncertain significance (VUSs). The cases defined as normal had no identifiable clinically significant copy number variants, and abnormal cases were defined as pathogenic.Results of the study included gestational age for 322 of 323 cases, which was then used to determine the rate of fetal demise or miscarriage at different gestational ages in the cohort, with first-trimester losses at <14 weeks (n = 230) and later losses defined as occurring at >14 weeks gestational age (n = 92).The results demonstrated that frequency and genetic abnormality types (as identified by CMA) align closely with karyotype studies previously published. This means that an approximate half of the first-trimester loss cohort have a genetic abnormality, with trisomies as the most common finding. Very early losses tend to have a high positive rate of abnormalities (45.6%), which further supported CMA use in first-trimester losses. Inadvertent assessment and reporting of maternal metaphases were reduced, and yield of results improved with the implementation of CMA.