No abstract

In a multi-institutional collaborative project, 1473 patients with myeloproliferative neoplasms (MPN) were screened for isocitrate dehydrogenase 1 (IDH1)/IDH2 mutations: 594 essential thrombocythemia (ET), 421 polycythemia vera (PV), 312 primary myelofibrosis (PMF), 95 post-PV/ET MF and 51 blast-phase MPN. A total of 38 IDH mutations (18 IDH1-R132, 19 IDH2-R140 and 1 IDH2-R172) were detected: 5 (0.8%) ET, 8 (1.9%) PV, 13 (4.2%) PMF, 1 (1%) post-PV/ET MF and 11 (21.6%) blast-phase MPN (P<0.01). Mutant IDH was documented in the presence or absence of JAK2, MPL and TET2 mutations, with similar mutational frequencies. However, IDH-mutated patients were more likely to be nullizygous for JAK2 46/1 haplotype, especially in PMF (P=0.04), and less likely to display complex karyotype, in blast-phase disease (P<0.01). In chronic-phase PMF, JAK2 46/1 haplotype nullizygosity (P<0.01; hazard ratio (HR) 2.9, 95% confidence interval (CI) 1.7–5.2), but not IDH mutational status (P=0.55; HR 1.3, 95% CI 0.5–3.4), had an adverse effect on survival. This was confirmed by multivariable analysis. In contrast, in both blast-phase PMF (P=0.04) and blast-phase MPN (P=0.01), the presence of an IDH mutation predicted worse survival. The current study clarifies disease- and stage-specific IDH mutation incidence and prognostic relevance in MPN and provides additional evidence for the biological effect of distinct JAK2 haplotypes.

Bone marrow DNA was screened for isocitrate dehydrogenase (IDH) mutations in 200 patients with chronic (n ¼ 166) or blast (n ¼ 34) phase myeloproliferative neoplasms (MPN). Included among the former were 77 patients with primary myelofibrosis (PMF), 47 essential thrombocythemia and 38 polycythemia vera (PV). Nine IDH mutations (5 IDH1 and 4 IDH2) were detected; mutational frequencies were B21% (7 of 34) for blast-phase MPN and B4% (3 of 77) for PMF. IDH mutations were seen in only 1 of 12 paired chronic-blast-phase samples and in none of 27 concurrently studied acute myeloid leukemia (AML) patients without antecedent MPN. IDH1 mutations included R132C (n ¼ 4; two post-PMF AML, one post-PV AML and one PMF) and R132S (n ¼ 1; post-PMF AML). IDH2 mutations included R140Q (n ¼ 3; one post-PMF AML, one post-PV AML and one PMF) and a novel R140W (n ¼ 1; mutation found in both chronicand blast-phase samples). The entire study cohort was also screened for JAK2 and MPL mutations and JAK2V617F was found in three IDH-mutated cases (two PMF and one PV). This study shows a relatively high incidence of IDH mutations in blast-phase MPN, regardless of JAK2 mutational status, and the occurrence of similar mutations in chronic-phase PMF.

p73, a ®rst p53 relative, was recently identi®ed and shown to be monoallelically expressed in a number of dierent human tissues. To determine the potential role of this gene in human bladder cancer, we investigated p73 expression levels, allelic expression patterns, and analysed p73 mutations in 23 unselected primary invasive bladder cancers with matched normal tissues and in seven bladder cancer cell lines. In a comparison between normal and tumor tissues using quantitative RT ± PCR analysis, we found that p73 was overexpressed in 22/23 bladder cancers, sometimes as great as 20-fold. Allelic expression analysis using a C/T polymorphism in exon 2 and a newly identi®ed T/C polymorphism in exon 5 revealed that p73 was biallelically expressed in both normal bladder and cancer tissues, suggesting that p73 is not imprinted in bladder tissue. Mutation screening of the p73 gene in bladder cancer DNAs using denaturing high-performance liquid chromatography analysis and DNA sequencing revealed no tumor-speci®c mutations in any coding exons of the p73 gene. These data suggest that the p73 is unlikely to be a tumor suppressor gene, but that overexpression of p73 may contribute to tumorigenesis in bladder cancer.