Diagnostic Testing for IDH1 and IDH2 Variants in Acute Myeloid Leukemia: An Analgorithmic Approach Using High Resolution Melting Curve Analysis.
Keyur P Patel, Bedia A Barkoh, Zhao Chen, Deqin Ma, L Jeffrey Medeiros, Rajyalakshmi Luthra. The University of Texas M.D. Anderson Cancer Center, Houston
Background: Mutations at R132 of IDH1 and at R140 and R172 of IDH2 and, polymorphism in G105 of IDH1 are clinically significant in acute myeloid leukemia (AML).The frequency of these mutations in AML is 5-15%. The widely used labor intensive Sanger sequencing method is impractical for routine detection of such low frequency mutations in clinical laboratories. Alternative reported methods such as pyrosequencing, high resolution melting (HRM) curve and restriction enzyme digestion assays detect only IDH1R132 and IDH2R172. Therefore, there is a need to develop screening assays that detect all four clinically significant variants including the G105 polymorphism in IDH1.
Design: We developed two clinical assays using HRM to screen all four variants i.e., G105 and R132 in IDH1 and, R140 and R172 in IDH2. An M13 sequence was incorporated in to PCR primers to allow direct Sanger sequencing of the amplicons using the same set of M13 primers for all four variants following HRM analysis. The IDH1 assay was optimized using the HT1080 cell line containing homozygous G105 polymorphism and heterozygous R132C mutation. Since no cell line containing IDH2 mutations was available, patient samples showing IDH2R140 or IDH2R172 mutations were used to optimize IDH2 assay. The OCI-AML3 Cell line was used as a negative control. We compared the results of the HRM assays with Sanger sequencing in 146 AML bone marrow samples.
Results: We observed a high concordance for all positive and negative results for IDH1 (98%) and IDH2 (94%) between the HRM assay and Sanger sequencing. Significantly, there were no false negative results by HRM. Table 1 shows the distribution of all mutant, wild-type and indeterminate calls by HRM and Sanger sequencing. The sensitivity of the assays in serial dilution studies using patient DNA was 7.9% for IDH1 assay and 7.3% for IDH2 assay.