Detection of Chromosomal Mosaicism - The importance of Karyotyping and FISH
Laboratory Genetics and Genomics
-
Primary Categories:
- Clinical Genetics
-
Secondary Categories:
- Clinical Genetics
Introduction:
The use of whole-exome/whole-genome sequencing for the detection of sequence-level variants and copy number variations (CNV) is increasingly being adopted as a primary testing method for genetic disease diagnosis by many institutions. Additionally, long-read sequencing is a promising technique to capture not only sequence level variants and CNVs but also structural chromosomal abnormalities. Significant progress has been achieved in this field; however, the detection of chromosomal mosaicism, including numerical gains and losses, rings, markers, and structural rearrangements – particularly those that are lineage-specific or exist at low levels – might still pose challenges with these emerging technologies. Furthermore, differentiating various mosaic cell lines may not be feasible. Literature review shows that chromosomal mosaicism is estimated to occur in up to 70% of human preimplantation embryos. Clinically significant chromosomal mosaicism is reported in ~3% of products of conceptions (POC), 1-2% of chorionic villus sampling (CVS), 0.1-0.3% of amniotic fluids (AF) samples, and 0.05-0.1% postnatal samples when assessed using conventional cytogenetic methods. To date, there are no studies available that provide data on the accuracy and detection levels of chromosomal mosaicism using third of fourth generation sequencing technologies.
Methods:
We conducted a retrospective review of patients who underwent conventional cytogenetic testing (chromosome analysis via Giemsa staining and fluorescence in situ hybridization (FISH)) for various clinical indications at Children’s Mercy Hospital from 2003 and 2023. Within this period, we received 20,866 constitutional cases of with 4,414 were cytogenetically abnormal. We also evaluated 762 cases submitted for exome sequencing with copy number analysis in 2023. We then examined all mosaic cases.
Results:
We found that the average number of mosaic cases each year was 29 (range: 9-39), accounting for ~3.1% of all clinical cases and ~14% of all abnormal cases per year. Levels of mosaicism were variable, ranging from 4 to 95%. Of all cases (both normal and abnormal) across all years: 4.7% of POCs, 4.3% of AF samples, and 2.7% of peripheral blood (PB) samples had chromosomal mosaicism. In comparison to the 762 PB exome cases, only three were mosaic (two with trisomy 9 and a ring(X)) totaling ~0.4% of postnatal cases.
Numerical abnormalities are the most common form of mosaicism in our cohort, accounting for ~79% of mosaic cases. Of the numerical abnormalities, sex chromosomes are the most frequently implicated, with an average involvement of 44% in POCs, 40% in AF samples, and 79% in PB. The most common numerical abnormality of the autosomes in POCs were affecting chromosomes 22 (7%), 15 (6%), and 16 (5%); in AF samples affecting chromosomes 21 (20%), 18 (7.5%), and 22 (5%); and in PBs affecting chromosomes 21 (11%) and 18 (4%).
Approximately 14% of mosaic cases harbored a structural abnormality of the autosomes. The most common abnormality in PBs involved the Y chromosome (12%), and in POCs involved chromosome 13 (19%). The structural abnormalities in AF samples affected chromosomes 3, 7, 9, 12, 14, 16, 18, and 22 at relatively equal prevalence.
Conclusion:
Review of our data indicates a higher prevalence of chromosomal mosaicism identified through conventional cytogenetic methods across various specimen types compared to existing literature. Furthermore, our limited genomic sequencing data suggests that conventional cytogenetic technologies may detect chromosomal mosaicism more frequently than genomic technologies. These findings underscore the importance of employing karyotyping and FISH for the detection of chromosomal mosaicism, particularly at low levels, which may be missed by genomic sequencing. Additionally, a thorough examination of clinical phenotypes can enhance the suspicion of mosaicism, ultimately enhancing diagnostic rates.
The use of whole-exome/whole-genome sequencing for the detection of sequence-level variants and copy number variations (CNV) is increasingly being adopted as a primary testing method for genetic disease diagnosis by many institutions. Additionally, long-read sequencing is a promising technique to capture not only sequence level variants and CNVs but also structural chromosomal abnormalities. Significant progress has been achieved in this field; however, the detection of chromosomal mosaicism, including numerical gains and losses, rings, markers, and structural rearrangements – particularly those that are lineage-specific or exist at low levels – might still pose challenges with these emerging technologies. Furthermore, differentiating various mosaic cell lines may not be feasible. Literature review shows that chromosomal mosaicism is estimated to occur in up to 70% of human preimplantation embryos. Clinically significant chromosomal mosaicism is reported in ~3% of products of conceptions (POC), 1-2% of chorionic villus sampling (CVS), 0.1-0.3% of amniotic fluids (AF) samples, and 0.05-0.1% postnatal samples when assessed using conventional cytogenetic methods. To date, there are no studies available that provide data on the accuracy and detection levels of chromosomal mosaicism using third of fourth generation sequencing technologies.
Methods:
We conducted a retrospective review of patients who underwent conventional cytogenetic testing (chromosome analysis via Giemsa staining and fluorescence in situ hybridization (FISH)) for various clinical indications at Children’s Mercy Hospital from 2003 and 2023. Within this period, we received 20,866 constitutional cases of with 4,414 were cytogenetically abnormal. We also evaluated 762 cases submitted for exome sequencing with copy number analysis in 2023. We then examined all mosaic cases.
Results:
We found that the average number of mosaic cases each year was 29 (range: 9-39), accounting for ~3.1% of all clinical cases and ~14% of all abnormal cases per year. Levels of mosaicism were variable, ranging from 4 to 95%. Of all cases (both normal and abnormal) across all years: 4.7% of POCs, 4.3% of AF samples, and 2.7% of peripheral blood (PB) samples had chromosomal mosaicism. In comparison to the 762 PB exome cases, only three were mosaic (two with trisomy 9 and a ring(X)) totaling ~0.4% of postnatal cases.
Numerical abnormalities are the most common form of mosaicism in our cohort, accounting for ~79% of mosaic cases. Of the numerical abnormalities, sex chromosomes are the most frequently implicated, with an average involvement of 44% in POCs, 40% in AF samples, and 79% in PB. The most common numerical abnormality of the autosomes in POCs were affecting chromosomes 22 (7%), 15 (6%), and 16 (5%); in AF samples affecting chromosomes 21 (20%), 18 (7.5%), and 22 (5%); and in PBs affecting chromosomes 21 (11%) and 18 (4%).
Approximately 14% of mosaic cases harbored a structural abnormality of the autosomes. The most common abnormality in PBs involved the Y chromosome (12%), and in POCs involved chromosome 13 (19%). The structural abnormalities in AF samples affected chromosomes 3, 7, 9, 12, 14, 16, 18, and 22 at relatively equal prevalence.
Conclusion:
Review of our data indicates a higher prevalence of chromosomal mosaicism identified through conventional cytogenetic methods across various specimen types compared to existing literature. Furthermore, our limited genomic sequencing data suggests that conventional cytogenetic technologies may detect chromosomal mosaicism more frequently than genomic technologies. These findings underscore the importance of employing karyotyping and FISH for the detection of chromosomal mosaicism, particularly at low levels, which may be missed by genomic sequencing. Additionally, a thorough examination of clinical phenotypes can enhance the suspicion of mosaicism, ultimately enhancing diagnostic rates.