Clinical Utility of Multiple-Gene Mutation Testing by Next-Generation Sequencing in Acute Myeloid Leukemia in Community Hospital Setting

Introduction

The WHO 2016 classification system and the 2017 European LeukemiaNet (ELN) genetic risk stratification system incorporate cytogenetic aberrations and mutations of specific genes to classify acute myeloid leukemia (AML) and stratify prognosis of AML patients. Next-generation sequencing (NGS) of multiple genes associated with AML has been shown to be an important tool in studying AML in research and academic center settings. Our goal is to evaluate the clinical application of NGS in a community hospital setting which has not been well documented leading to rejection of coverage by some insurers.

Methodology

This retrospective study included 109 patients (52 females and 57 males with mean age of 58.7) from November, 2016 to June, 2018 with various races/ethnicities background and confirmed diagnosis of AML at Florida Hospital (FH). DNA was extracted from the peripheral blood/bone marrow for NGS analysis. A total of 54 hematopoietic tumor associated genes were studied: ABL1, ASXL1, ATM, BCOR, BIRC3, BRAF, CALR, CBL, CEBPA, CREBBP, CSF1R, CSF3R, DNMT3A, EP300, ETV6, EZH2, FBXW7, FGFR4, FLT3, GATA1, GATA2, GATA3, IDH1, IDH2, IL7R, JAK2, JAK3, KDM6A, KIT, KMT2A, KRAS, MPL, NF1, NOTCH1, NOTCH2, NPM1, NRAS, NSD1, PAX5, PDGFRA, PDGFRB, PTPN11, RUNX1, SETBP1, SF3B1, SRSF2, STAG2, TERT, TET1, TET2, TP53, TSLP,U2AF1 and ZRSR2. The sequencing was performed at Genomics and Pathology Services laboratory at Washington University in St. Louis and initial informatics analysis was performed by PierianDX using Clinical Genome Working platform. The initial analysis results were delivered to the pathologists at FH to do final interpretation with integration of the results of morphology/flow cytometry/karyotyping studies.

The clinical utility of NGS was evaluated by: 1) comparing the AML classification and ELN genetic risk stratification of each patient with and without NGS data, 2) identifying gene mutations for targeted therapy and 3) other clinical impacts.

Results

NGS results changed ELN risk stratification in 45 cases (41.3%) and WHO AML classification in 37 cases (33.9%) out of 109 cases after considering the mutation status of NPM1, FLT3, RUNX1, ASXL1 and TP53. Among the 45 cases that had changes of ELN risk, 25 cases (55%) were downgraded to favorable risk group all because of mutations in NPM1 gene. While, the rest 20 cases (45%) were upgraded to adverse risk group due to the following mutations (RUNX1 in 7, TP53 in 5, ASXL1 in 4, RUNX1 and ASXL1 in 4). The change of WHO classification was caused by mutations in NPM1 (n=26) or RUNX1 (n=11). Additionally, NGS identified 8 cases with either GATA2 or CEBPA gene mutation, which are considered to increase the risk of myeloid neoplasms with germline predisposition based on WHO classification. Finally, NGS identified 42 patients (38.5%) carrying FLT3-ITD (n=17), FLT3-TKD (n=8) and/or IDH1/IDH2 (n=16) mutations. These mutations are known to have approved targeted therapies (FLT3 inhibitors and IDH2 inhibitors) or ongoing clinical trials (IDH1 inhibitors). Overall, NGS results led to clinically significant impacts in 62/109 (56.9%) patients by changing the AML classification/risk stratification, finding potential germline mutation and/or targeted therapies.

Discussion

Our results indicate that NGS findings provide important information for classification, risk stratification, genetic predisposition, and treatment options in the majority (62/109, 56.9%) of AML patients in a community hospital setting. How the information exactly changes the clinical management by community practicing hematologists is currently ongoing and will provide further insights in regards to the clinical utilities of NGS. In addition, NGS of multiple genes is more cost- and labor-effective than single gene sequencing. Taken together, the insurers should accept NGS testing as a standard of care for AML patients.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.