Whole-genome sequencing simplifies workup of AML, myelodysplastic syndrome
Whole-genome sequencing appeared as accurate as and more efficient than conventional cytogenetic tests for genomic profiling of acute myeloid leukemia and myelodysplastic syndrome, according to data in The New England Journal of Medicine.
Moreover, whole-genome sequencing yielded new genetic information from 25% of patients — more than half of whom would have been assigned to a different genetic risk category based on conventional testing — and in a shorter time frame.
“Our motivation was to simplify and improve the accuracy of genetic testing for patients with AML and myelodysplastic syndrome,” David H. Spencer, MD, assistant professor of medicine and medical director of the clinical sequencing facility at McDonnell Genome Institute of Washington University School of Medicine in St. Louis, told Healio.
“Mutations in these cancers are very important for making treatment decisions and include a wide range of genomic changes, from changes involving whole chromosomes to gene mutations at a single genomic position,” he added. “This means that multiple genetic tests are often required to obtain all the information that is needed to decide on the appropriate therapy.”
Conventional metaphase karyotyping is an important genetic test for AML and myelodysplastic syndrome; however, it has limitations, Spencer said.
“The resolution is limited to changes that involve millions of base pairs of the genome, and so some mutations can be missed,” he said. “We reasoned that a streamlined genome sequencing approach could obtain all this information in a single test, which would be simpler and potentially more accurate than current testing methods.”
The streamlined approach, called ChromoSeq, provides comprehensive genomic profiling by assessing mutations in 40 genes, copy-number alterations of more than 5 Mbp and recurrent structural variants in a short turnaround time. The researchers evaluated whether this whole-genome sequencing approach could serve as a replacement for conventional cytogenetic and sequencing approaches — which included cytogenetic analysis, fluorescence in situ hybridization and targeted sequencing — by testing its accuracy, feasibility and clinical utility.
They evaluated blood samples from fresh bone marrow aspirate or peripheral-blood specimens of 263 patients with myeloid cancers, 235 of whom underwent successful cytogenetic analysis for comparison. These included 146 retrospective samples (AML, n = 107; mean age, 53.7 years; 44% women; myelodysplastic syndrome, n = 39; mean age, 59.8 years; 44% women) and 117 prospective samples (AML, n = 68; mean age, 60.6 years; 44% women; myelodysplastic syndrome, n = 42; mean age, 68.9 years; 29% women).
Results showed whole-genome sequencing detected all 40 recurrent translocations and 91 copy number alterations that had been identified by cytogenetic analysis.
Further, whole-genome sequencing identified new clinically reportable genomic events in 40 patients (17%). These included 21 new copy-number alterations in 14 patients, new structural variants in 13 patients, and new copy-number alterations in an additional 13 patients with ambiguous or inclusive results by cytogenetic analysis.
Additionally, the prospective sequencing — performed in a median 5.1 days and, when staffing allowed, in as little as 3 days — provided new genetic information for 29 patients (24.8%), leading to a change in risk category for 19 of them (16.2%).
To determine whether whole-genome sequencing can predict clinical outcomes using existing genetic risk group, researchers evaluated data of 71 patients with AML who did not undergo hematopoietic stem cell transplant. Sixty-three of the patients had risk-group assignments in agreement based on conventional testing and whole-genome sequencing. Of the eight who were reassigned to another risk group, five had new adverse-risk findings identified by whole-genome sequencing.
Cox regression analysis showed identification of patients with adverse risk and poor outcomes was slightly better with whole-genome sequencing (HR for death = 0.32; 95% CI, 0.11-0.92) than conventional risk-group analysis (HR = 0.66; 95% CI, 0.17-1.05).
Researchers then evaluated data of 27 patients with AML who did not undergo HSCT and who could not be assigned to a risk group at diagnosis due to unsuccessful cytogenetic analysis or inconclusive results, hypothesizing that whole-genome sequencing would be most informative for these patients.
Survival analysis showed risk predictions based on whole-genome sequencing correlated with outcomes, with longer OS among the 21 patients with intermediate or favorable risk (20.5 months; 95% CI, 5.6-28.8) than among the six patients with adverse risk (median survival, 3.3 months; 95% CI, 1.7-18.9; P = .03).
“We think the ability to perform testing for chromosomal mutations using DNA instead of live cells, improved resolution, and being able to test for all the types of mutations using a single test are all important benefits of our genome-sequencing approach,” Spencer said.
“The cost of sequencing continues to decrease over time due to advances in sequencing technology,” Spencer added. “This is a major reason why genome sequencing is now practical for clinical testing. Our approach also benefited from many other studies that cataloged the mutations that are important in AML and myelodysplastic syndrome, as well as inherited genetic changes that are benign. These resources are critical for our test because it allows us to focus on the changes that are clinically meaningful while not being distracted by variations in the genome that are not relevant.”
The next step, Spencer said, is to take the genome-sequencing approach into a clinical study to determine if its accuracy in genetic profiling and turnaround time improve clinical outcomes.
Many of the benefits of whole-genome sequencing observed in this study can be directly applied to patients with other cancers, Spencer added.
“This is especially true for cancers with a range of mutation types that are important to test for, like translocations, copy number changes and gene mutations,” he said. “Genome sequencing is one of the best approaches to use to detect all of these changes, and it can be performed using just a small amount of DNA, which makes it a robust technology.”
For more information:
David H. Spencer, MD, can be reached at Division of Oncology, Department of Medicine, Washington University School of Medicine, Campus Box 8007, 660 S. Euclid Ave., St. Louis, MO 63110; email: firstname.lastname@example.org.