Introduction:
In pregnancies with ultrasound anomalies, chromosomal microarray analysis (CMA) can identify key genetic anomalies as well as rarer microdeletions and microduplications not detectable by karyotype. In approximately 6-10% of pregnancies with a fetal structural anomaly, CMA reveals copy number variants (CNVs) that may be clinically relevant yet challenging to classify and interpret (PMID: 23215555). A lack of standardization exists in classification, interpretation and reporting of genetic variants found on prenatal CMA. Whereas robust guidelines exist for postnatal microarray classification and reporting, additional clarity would be beneficial for prenatal cases.

Methods:
In 2016, the California Prenatal Screening Program (PNS) introduced chromosomal microarray as an authorized service for patients undergoing amniocentesis or chorionic villus sampling after screening positive for Trisomy 21 or 18 on maternal serum screening and exhibiting certain ultrasound findings. Data was abstracted on all microarrays performed from April 1, 2016 through June 30, 2019 to determine the comparative yield of anomalies detected on CMA but not karyotype, the proportion of clinically relevant findings possibly detectable by karyotype alone, the degree of agreement in classification and reporting of variants (Pathogenic, Likely Pathogenic, Likely Benign, Variant of Unknown Significance, Would Not Report, Carrier Status for Autosomal Recessive Condition, or Region of Homozygosity) by a panel of cytogeneticists, and the impact of reporting results to clinicians and patients.     

 

Results:
In the study period, 397 chromosomal microarrays were ordered through the PNS Program. One resulted in culture failure and microarray was not repeated. Of the 396 successful microarrays, 393 were ordered due to ultrasound indications and 3 were ordered due to karyotype abnormalities.

There were 257 normal results (65%) and139 abnormal results (35%). Of the abnormal results, 106 (76%) were detectable by karyotype, including 83 aneuploidies (33 T21, 20 T18, 13 T13,12 45,X, 2 47,XXX,1 47,XXY, 1 T16,1 T22), 13 triploidies, and 10 other abnormalities large enough to be detectable by conventional cytogenetic methods (3 balanced translocations, 3>10Mb duplications or marker chromosomes, 4>10Mb deletions).

The 10 abnormal results that were not aneuploidy or triploidy and the remaining 33 abnormal CMA results (24%) that were not detectable by karyotype were reviewed by the cytogenetic panel members for classification, reportability and relationship to their referring ultrasound indications. 

Of these 43 abnormal CMA results, most cytogeneticists (>75%) classified 25 of them as Pathogenic or Likely Pathogenic and 5 as Likely Benign.  There was no clear agreement on the reporting of Variants of Unknown Significance. Cytogenetic panel members indicated 30 results (70%) as unlikely/ uncertain to be related to the referring ultrasound indication. These 43 cases were further reviewed by a clinical geneticist who classified 23 (54%) of them as not likely to be related to the referring ultrasound indication. A majority (73%) of sub-microscopic abnormal CNVs were considered incidental.

 

Conclusion:
Our results indicated need for more structured guidelines for classifying and reporting prenatal CMA results to improve consistency.  Standardization is important to a statewide screening program to ensure equitable results for all participants, regardless of geography or laboratory utilized for analysis. This study prompted the California Prenatal Screening Program’s Cytogenetics Committee to create Consensus Reporting Guidelines for Prenatal Microarray to foster more uniform classification and reporting. The study also highlighted the critical need for genetic counseling as many of the results were incidental and unrelated to the referring ultrasound indication. Additional questions identified included the effects of reporting variants with variable expressivity, adult/unclear timing of onset, or incomplete penetrance on reproductive decision making, the relative utility of CMA vs. karyotype, and how to manage costs of follow-up parental testing. Further work is required to optimize the incorporation of prenatal CMA results into public health programs.