Detection of Copy Number Variants by Chromosome Microarray Verses Exome Sequencing at a Single Clinic over an Eight Year Period
Clinical Genetics and Therapeutics
-
Primary Categories:
- Clinical- Pediatric
-
Secondary Categories:
- Clinical- Pediatric
Introduction:
In 2021, the ACMG practice guideline on “Exome and genome sequencing for pediatric patients with congenital anomalies or intellectual disability” strongly recommended that exome sequencing (ES) or genome sequencing (GS) be completed “as a first- or second-tier test… for patients with one or more congenital anomalies with onset prior to age one year or for patients with developmental delay or intellectual disability with onset prior to age 18 years.” Based on uncertainty in the detection of potentially clinically significant copy number variants (CNV) through short-read next generation sequencing utilized in ES, we continued to complete chromosome microarray (CMA) as first-tier testing for these populations followed by ES when GS was not possible to coordinate. We attempted to answer this uncertainty by reviewing outcomes of patients who had a reportable variant through CMA followed by ES.
Methods:
Over an eight-year period (October 2016 to September 2024), 270 patients seen at our center were identified as having a reportable variant, including pathogenic (P), likely pathogenic (LP), and unknown significance (VUS) on CMA and then had reflexive ES testing. Both the CMA (commercially available oligonucleotide array with SNP probes) and ES (proprietary next-generation sequencing with paired-end reads) were completed at a single diagnostic laboratory in this period. By CMA, 13 P/LP CNV were identified in 13 patients and 30 VUS were identified in 25 patients.
Results:
The average size of the 13 P/LP CNV was 1099kb with a range of 34-4316kb and all (13/13) were confirmed by ES. Thirty VUS were reported in 25 patients (three patients had two VUS and one patient had three VUS). The average ratio of VUS per CMA ordered decreased from 0.23 in the first 3 years (3/12; 4/15; 10/46) to 0.07 the last 5 years (2/41; 3/39; 4/65; 2/30; 2/22) (p=0.0001). All but six of the 30 (0.80) VUS found on CMA were recapitulated on ES: four in the first 3 years and two in the last 5 years. Three of the four VUS not identifiable on ES in the first 3 years were single exon deletions, 53-57kb in size, all in OMIM-curated disease-causing genes (IMMP2L, NRXN1, and MBD5). The fourth VUS not identified on ES in the first 3 years and both VUS not found in the last 5 years were on the X chromosome (161-741kb in size) and containing all or part of between two and nine genes, at least one of which was disease-causing.
Conclusion:
First, there has been a change in the analysis of CMA at the testing facility utilized by our center about five years ago (2019-2020), which caused a statistically significant decrease in VUS reported on CMA. This may have a secondary effect of raising the concordance of CNV calls on CMA and ES. Second, discontinuing CMA prior to ES with current protocols at our reference-lab-of-choice has a potential to miss certain CNV (two VUS in 197 specimens from our center in the past 5 years). Based on this limited data review, there appears to be enrichment of X-chromosome CNV escaping detection on ES. Given the high concordance between CMA and ES detection of CNV, we can have confidence to order ES as first-tier test, with consideration of CMA in patients with non-diagnostic ES results, until GS becomes a viable option for more of our patient population.
In 2021, the ACMG practice guideline on “Exome and genome sequencing for pediatric patients with congenital anomalies or intellectual disability” strongly recommended that exome sequencing (ES) or genome sequencing (GS) be completed “as a first- or second-tier test… for patients with one or more congenital anomalies with onset prior to age one year or for patients with developmental delay or intellectual disability with onset prior to age 18 years.” Based on uncertainty in the detection of potentially clinically significant copy number variants (CNV) through short-read next generation sequencing utilized in ES, we continued to complete chromosome microarray (CMA) as first-tier testing for these populations followed by ES when GS was not possible to coordinate. We attempted to answer this uncertainty by reviewing outcomes of patients who had a reportable variant through CMA followed by ES.
Methods:
Over an eight-year period (October 2016 to September 2024), 270 patients seen at our center were identified as having a reportable variant, including pathogenic (P), likely pathogenic (LP), and unknown significance (VUS) on CMA and then had reflexive ES testing. Both the CMA (commercially available oligonucleotide array with SNP probes) and ES (proprietary next-generation sequencing with paired-end reads) were completed at a single diagnostic laboratory in this period. By CMA, 13 P/LP CNV were identified in 13 patients and 30 VUS were identified in 25 patients.
Results:
The average size of the 13 P/LP CNV was 1099kb with a range of 34-4316kb and all (13/13) were confirmed by ES. Thirty VUS were reported in 25 patients (three patients had two VUS and one patient had three VUS). The average ratio of VUS per CMA ordered decreased from 0.23 in the first 3 years (3/12; 4/15; 10/46) to 0.07 the last 5 years (2/41; 3/39; 4/65; 2/30; 2/22) (p=0.0001). All but six of the 30 (0.80) VUS found on CMA were recapitulated on ES: four in the first 3 years and two in the last 5 years. Three of the four VUS not identifiable on ES in the first 3 years were single exon deletions, 53-57kb in size, all in OMIM-curated disease-causing genes (IMMP2L, NRXN1, and MBD5). The fourth VUS not identified on ES in the first 3 years and both VUS not found in the last 5 years were on the X chromosome (161-741kb in size) and containing all or part of between two and nine genes, at least one of which was disease-causing.
Conclusion:
First, there has been a change in the analysis of CMA at the testing facility utilized by our center about five years ago (2019-2020), which caused a statistically significant decrease in VUS reported on CMA. This may have a secondary effect of raising the concordance of CNV calls on CMA and ES. Second, discontinuing CMA prior to ES with current protocols at our reference-lab-of-choice has a potential to miss certain CNV (two VUS in 197 specimens from our center in the past 5 years). Based on this limited data review, there appears to be enrichment of X-chromosome CNV escaping detection on ES. Given the high concordance between CMA and ES detection of CNV, we can have confidence to order ES as first-tier test, with consideration of CMA in patients with non-diagnostic ES results, until GS becomes a viable option for more of our patient population.
)
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
)
){kind=logo}
)
)
)
)
)
)
){kind=logo}
){kind=logo}
){kind=logo}