IDH mutation analysis is not suitable for the routine molecular diagnostic algorithm in myeloproliferative and myelodysplastic neoplasms

Acquired somatic mutations of isocitrate dehydrogenases 1 and 2 (IDH1 and IDH2) in preleukemic disorders

Mutations of isocitrate dehydrogenase isoform 1 and 2 (IDH1 and IDH2) genes have been identified in glioblastoma and acute myeloid leukemia (AML). However, little is known about the molecular alterations of IDH genes in preleukemic disorders with a propensity to transform to AML. We performed polymerase chain reaction-denaturing high performance liquid chromatography (PCR-DHPLC) followed by direct sequencing to detect IDH mutations in 237 patients with myeloproliferative neoplasms (MPNs; n=108), myelodysplastic syndrome (MDS; n=22), paroxysmal nocturnal hemoglobinuria (PNH; n=41), and aplastic anemia (AA; n=66). No IDH1 R132 and IDH2 R172 mutations were identified in the entire cohort, whereas IDH1 G105G allele was detected in 4/108 MPN (3.70%), 2/22 MDS (9.09%), and 2/41 PNH (4.88%) patients. Three IDH2 R140Q mutations were found in 2/108 MPN (1.85%) and 1/22 MDS (4.54%) patients, while one IDH2 G145G allele was found in 0.92% (1/108) of MPN patients. Overall, our data suggest that IDH mutations are rare in the preleukemic disorders and may not be the major initial step in AML leukemogenesis.

Serum 2-Hydroxyglutarate Production in IDH1- and IDH2-Mutated De Novo Acute Myeloid Leukemia: A Study by the Acute Leukemia French Association Group

Purpose

Mutated isocitrate dehydrogenases (IDHs) 1 and 2 produce high levels of 2-hydroxyglutarate (2-HG). We investigated whether, in acute myeloid leukemia (AML), serum 2-HG would predict the presence of IDH1/2 mutations at diagnosis and provide a marker of minimal residual disease (MRD).

Patients and Methods

Serum samples from 82 patients at diagnosis of de novo AML (IDH1/2 mutated, n = 53) and 68 patients without AML were analyzed for total 2-HG and its ratio of D to L stereoisomers by mass spectrometry. We measured 2-HG levels and molecular markers of MRD (WT1 and NPM1) in serial samples of 36 patients with IDH1/2 mutations after induction therapy.

Results

In patients with AML with IDH1/2 mutations, 2-HG serum levels were significantly higher than in patients with IDH1/2 wild type (P < .001). Area under the receiver operating characteristic curve was 99%. The optimum diagnostic cutoff between IDH1/2 mutated and normal was 2 μmol/L (sensitivity, 100%; specificity, 79%). Quantification of the D/L stereoisomers increased specificity (100%; 95% CI, 83% to 100%) compared with total 2-HG (P = .031). In patients with IDH2 R172 mutations, 2-HG levels were higher relative to those with other IDH1/2 mutations (P < .05). During follow-up, serum 2-HG levels showed strong positive correlation with WT1 and NPM1 (P < .001). After induction therapy, total 2-HG serum levels < 2 μmol/L were associated with better overall (P = .008) and disease-free survival (P = .005).

Conclusion

Serum 2-HG is a predictor of the presence of IDH1/2 mutations and outcome in these patients. Discrimination between D/L stereoisomers improved specificity.