AUTHORS
Indira Mehta, PhD, Xiaoyu QU, PhD, FACMG2, Austin Doke3, Margaret C. Skokan4, Derek Stirewalt, MD5, Jerald Radich, MD6, Min Fang, MD, PhD, FACMG7
AFFILIATIONS
1 Fred Hutchinson Cancer Center, Seattle, WA, United States
2 Fred Hutchinson Cancer Center, Seattle, WA, United States, Fred Hutchinson Cancer Center, Seattle, Washington, United States
3 Fred Hutchinson Cancer Center, Seattle, WA, United States
4 Fred Hutchinson Cancer Center, Seattle, WA, United States
5 Fred Hutchinson Cancer Center, Seattle, Washington, United States
6 Fred Hutchinson Cancer Center, Seattle, Washington, United States
7 Fred Hutchinson Cancer Center, Seattle, WA, United States, Fred Hutchinson Cancer Center, Seattle, Washington, United States
ABSTRACT
Introduction:
Optical genome mapping (OGM) is an all-encompassing cytogenomic technology that can detect structural variants (SV) and copy number aberrations (CNA) for hematologic malignancies with a higher resolution than standard-of-care (SOC) testing, including karyotyping, FISH, and chromosomal genomic array testing (CGAT). As part of a comprehensive assessment to validate OGM for clinical use, we performed a validation study comparing SOC test results with OGM results from a variety of hematologic malignancy samples. This report describes the performance characteristics of one subset of a more extensive validation study focusing on acute myeloid leukemia (AML).
Methods:
This cohort consisted of diagnostic AML bone marrow or peripheral blood samples from 17 males and 10 females. Samples were selected based on available material with presence of various SVs from the Fred Hutchinson Leukemia Repository. For OGM, ultra-high-molecular-weight genomic DNA was extracted, labelled, imaged on the Saphyr (Bionano) and initially analyzed using the Rare Variant (RV) pipeline in the Access software (Access/Solve, 1.7.2/3.7) and later using the GA (Access 1.8.2/Solve 3.8.2) and VIA analysis software (VIA 7.1).
Results:
Karyotype analysis was performed on all 27 samples with concurrent CGAT also performed on 9 samples. One sample displayed a normal result by both SOC and OGM testing. The remaining 26 samples displayed a total of 87 SV and CNA calls. Eleven of these calls were subclonal (<5% VAF for SVs, < 20% VAF for CNAs, and <20Mb for AI/AOH) and were excluded from further performance assessment. One translocation SV call, t(2;19)(q35;p13.3), observed at 12.5% expected VAF, was not concordant by OGM and of uncertain clinical significance.
Forty-three of SV/CNA calls were concordant between SOC and OGM, showing the following breakdown: aneuploidies (16/43=37%), translocations (8/43=19%), deletions/CNA losses (10/43=23%), inversions (5/43=12%) and copy number loss of heterozygosity (cnLOH) (5/43=9%) and displaying the following ranges for positive percent agreement (PPA) 80%-98%, negative percent agreement (NPA) 100%, accuracy 88%-98%, sensitivity 80%-98%, and specificity 100%, for the overall and various SV/CNA types. OGM displayed 100% concordance of World Health Organization (WHO) and European Leukemia Net (ELN) guideline defined aberrations: MECOM-r, -5/(del)5q, t(6;9), -7/del(7q), +8, KMT2A-r, and inv(16)/t(16;16), identified by SOC testing.
The remaining 32 calls were additional OGM findings not observed by SOC, showing the following breakdown: translocations (9/32=28%), deletions/CNA losses (9/32=28%), CNA gains (2/32=6%), cnLOH (3/32=9.3%), fusions (7/32=22%), and insertions (2/32=6%). Included among the calls uniquely identified by OGM were clinically relevant TET2 deletion, 13q cnLOH and the following WHO/ELN defined aberrations: KMT2A-PTD, ETV6-r, and NUP98-r, abnormalities which were all missed by SOC testing. The additional OGM findings also clarified ambiguous rearrangements seen on karyotype analysis by delineating breakpoints and detecting additional CNAs and translocations revealing genome complexity.
Comparison of these calls to the initial analysis using the RV pipeline and earlier software version identified 15% more SV/CNA calls, with cnLOHs and fusions constituting most of the call types not identified in the initial analysis.
Conclusion:
OGM identified all clonal abnormalities detected by SOC analysis that were expressed above the expected assay limit of detection, as well as additional SV/CNV calls further clarifying aberrations identified by SOC testing and revealing greater complexity of the abnormalities detected within the genome. The use of the GA pipeline in the newer VIA analysis software detected additional informative SV/CNA not detected by earlier analysis software due primarily to the added ability to detect cnLOHs and enhanced visualization of fusion SVs in VIA. The increased identification of SV/CNA calls, in addition to the ease of use of the integrated visualization and interpretation VIA analysis platform, highlights the benefits of the new software analysis program. These results demonstrate the power of the single OGM assay to provide additional diagnostic and prognostic insight beyond the current existing SOC testing methods for hematologic malignancies.
DISCLOSURES
Indira Mehta, PhD: No relevant conflicts of interest to declare.