Abstract 4098

Clustered in the JAK2 exon 12 region, somatic gain-of-function mutations have been found in patients with JAK2V617F-negative Polycythemia Vera and idiopathic erythrocytosis (by Scott et al). The presence of Jak2 exon 12 mutations is an important clonal marker for the diagnosis of JAK2 V617 negative P. Vera. Detailed analysis of the JAK2 exon 12 mutations has revealed more than 10 different mutations in the JAK2 exon 12 region. JAK2 exon 12 mutations are usually expressed at low levels in granulocyte DNA. In order to obtain more accessible data, several strategies have been developed. DNAs isolated from erythropoietin-independent colony are required in some cases when direct DNA sequencing is used to detect the mutations. Allele specific PCR has great sensitivity for detecting the mutations. However, designing a routine PCR test for the detection of JAK2 exon 12 mutations is difficult because the complexity of the mutation sites in JAK2 exon 12, requiring multiple-tubes PCR using a large number of primers. Most of the JAK2 exon 12 mutations are deletions or insertions, accounting for more than 70≂∼f90% of the cases. We have created a fluorescent PCR fragment analysis for the detection of known or unknown mutations based on DNA fragment analysis. To further increase sensitivity, multiple-allele-specific PCR is also created. It is designed to target the five most common exon 12 mutations: N542-E543del, F537-K539delinsL, E543-D544del, K539L and H538QK539L. The fluorescent PCR fragment analysis's sensitivity can reach 3.3%, while the allele-specific PCR is capable of detecting a heterozygous mutation's presence at less than 1% of the cells in our validation study.

Seven patients with JAK2 exon 12 mutations were detected by our assay. Six patients had Jak2 exon 12 deletions, which included N542-E543del, E543-D544del and H538-K539L mutations. One patient showed K539L point substitution. The specificity of all mutations in the seven patients were evaluated by fluorescent fragment PCR analysis, allele specific PCR, and DNA sequencing. The mutations in all patients were confirmed by at least two methods and corresponding results were observed in all of patients. Six plasmids containing the most common Jak 2 exon 12 mutations were constructed and were used for the controls. The mutations were: F537-K539delinsL, N542-E543del, E543-D544del, H538QK539L, K539L and H538QK539L. The JAK2 exon 12 assay results agreed 100% with DNA sequencing in all six positive controls. In the patients with deletions, the ratio of the mutated peak vs. wild type peak varied from 3.33% to 64%. The mutations detected by DNA sequencing were readable in 4 of the 7 samples, showed traces in two samples, and showed no mutations in one sample. However, fluorescent fragment analysis and allele specific PCR yielded remarkable peaks representing the mutations. For example, in patient with 3.33% mutations, the height of the N542-E543del peak was approximately 8000 relative units in allele-specific PCR. There were no nonspecific peaks found in the results from several hundred clinical samples. All seven positive patients with Jak 2 exon 12 mutations had a hypercellular marrow with erythrocytosis, and a subset lacked the classic atypical megakarocyte clustering and atypia commonly seen in chronic myeloproliferative neoplasms such as P. vera. Therefore, JAK2 exon 12 mutation analysis helped to identify a subset of patients with P. Vera which lacked typical histologic criteria for P.V.

So far, DNA sequencing and allele-specific PCR are the most common methods for the detection of JAK2 exon 12 mutations. Our data shows that direct DNA sequencing analysis is not suitable for JAK2 exon 12 mutations since its limitation in sensitivity. The application of more than 10 allele specific PCR reactions for Jak 2 exon 12 mutations is not practical in clinical laboratories because the procedure is both time-consuming and complex. Our assay uniquely combines fluorescent PCR fragment analysis and allele-specific PCR so that we can dramatically reduce the number of PCR reactions and provide 1% sensitivity for greater than 70% of known mutations and 4% sensitivity for any other mutations. The assay is able of detecting both all known mutations and, in theory, novel JAK2 exon 12 deletions or insertions. The results demonstrate that our assay provides valuable data for physicians in the diagnosis of P.V. patients with Jak2 V617F-negative.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.