The Value of Next-Generation Sequencing in the Screening and Evaluation of Hematologic Neoplasms in Clinical Practice

Victoria Northrup, MSc; Allison Maybank, MSc; Nancy Carson, PhD; Tarek Rahmeh, MD


Am J Clin Pathol. 2020;153(5):639-645. 

In This Article

Materials and Methods


This study was approved by the Horizon Health Network Research Ethics Board. A total of 291 peripheral blood and bone marrow aspirate specimens were submitted for NGS to the Molecular Diagnostics Division of Laboratory Medicine at the Regional Hospital between November 2017 and November 2018. Among these, we randomly selected a cohort of 178 specimens, excluding cancelled tests due to ordering redundancy. This cohort included 43 bone marrow aspirates and 136 peripheral blood samples. NGS was either requested by the consultant hematologist for a peripheral blood sample or reflex-ordered by the hematopathologist for a bone marrow aspiration. Of the 178 specimens, 156 were collected at the Regional Hospital (in-house cases). The remaining 22 were referred in from other sites of our Health Network. Relevant laboratory and clinical information was retrieved for all in-house cases. This includes hematology consultation reports, CBCs, peripheral blood smear morphologic evaluations, bone marrow pathology reports, cytogenetic studies, and standalone molecular assays. In contrast, and with the exception of a single case, no clinical or laboratory data were available to researchers on the 22 referred-in cases.


DNA and RNA were extracted from the bone marrow aspirate and peripheral blood sample. Library preparation was performed using the Ion AmpliSeq Myeloid Panel (Thermo Fisher) (145 samples) or the Oncomine Myeloid Panel (33 samples). NGS was performed using the Ion Torrent S5 XL instrument. Variants were detected using the Ion Suite and Ion Reporter (version 5.2) software, based on a minimum coverage of ×100, and the SeqNext software (JSI Medical Systems). The sensitivity of the assay is approximately 5% for single-nucleotide variants, 10% for indels in a background of normal sequence, and 15% for fusions. Novel potentially clinically relevant variants and variants that had a frequency of 5% or less were confirmed using Sanger sequencing and the software Minor Variant Finder (version 1.0; Thermo Fisher). Variants with a frequency of 5% or less were reported if confirmed with an alternative methodology (ie, Sanger sequencing).

The following genes are included in the Ion AmpliSeq testing panel: ABL1, ASXL1, BRAF, CALR, DNMT3A, ETV6, EZH2, FLT3, GATA2, IDH1, IDH2, JAK2, KIT, KRAS, MPL, NPM1, NRAS, PTPN11, RUNX1, SETBP1, SF3B1, SH2B3, SRSF2, TP53, U2AF1, CBL, CSF3R, CEBPA, TET2, and ZRSR2. The Oncomine panel included the following additional genes: BCOR, HRAS, IKZF1, MYD88, NF1, PHF6, PRPF8, RB1, STAG2, and WT1.

Variant classification, if not a known pathogenic or benign variant, was determined using the recommended joint guidelines of the College of American Pathologists, the Association for Molecular Pathology, and the American Society for Clinical Oncology.[10] Briefly, variants were classified as pathogenic if they were of a type that had been described previously as being pathogenic, such as frameshift variants in genes known to be tumor suppressor genes. For variants of unknown clinical significance, VarSome[11] was used to assess potential pathogenicity. VarSome is a search engine that automatically classifies variants based on the American College of Medical Genetics guidelines[10] and provides links to various databases that list the variant in question, including ClinVar and gnomAD, as well as results from in silico analyses. In addition, a literature search was performed to look for reports that may provide more information on potential pathogenicity. A variant was classified as likely pathogenic if VarSome called the variant pathogenic or likely pathogenic and/or if functional studies were available that definitively showed an effect on protein function. Otherwise a variant was classified as a VUS.

In our clinical practice, NGS is reported to physicians via a molecular diagnostic report that provides the variants found, their classification, and an interpretation by a molecular geneticist. In cases where NGS is performed on a bone marrow sample, a supplemental report is issued by the hematopathologist to whom the case is assigned to describe the NGS findings and their contributive diagnostic and prognostic value, when applicable.

Statistical Analysis

Descriptive statistics were used to determine the percentage of variants detected for various myeloid neoplasms. Statistical significance of bone marrow vs peripheral blood was determined using a Fisher exact test, and significance for NGS vs cytogenetics was determined using McNemar χ2 analysis. Sensitivity and specificity were calculated based on the presence of a pathogenic variant, likely pathogenic variant, or a VUS that is suspected of being pathogenic corresponding to a myeloid neoplasm phenotype. Sensitivity and specificity were calculated in Excel (Microsoft).