Analysis of JAK3, JAK2, and C-MPL mutations in transient myeloproliferative disorder and myeloid leukemia of Down syndrome blasts in children with Down syndrome

To the editor:

Children with Down syndrome are predisposed to 2 linked clonal myeloid disorders: transient myeloproliferative disorder (TMD¹ ) and myeloid leukemia of Down syndrome (ML-DS). In addition to trisomy 21, somatic mutations in GATA1 are required events for TMD and ML-DS (reviewed in Gurbuxani et al²  and Hitzler and Zipursky³ ). Because only 20% to 30% of TMD cases progress to ML-DS, additional (epi)genetic events are likely to be important for the progression of TMD to ML-DS.⁴ 

Recently, acquired constitutively activating mutations have been described in the pseudokinase domain of the JAK3 kinase in 2 studies.^5,6  In addition, mutations in JAK2 and C-MPL⁷  have been detected in adults with myeloproliferative disorders (reviewed in Campbell and Green⁸ ). These observations raise the possibility that mutations in JAK2, JAK3, and/or C-MPL may cooperate with mutant GATA1 to promote the development of ML-DS.

We screened patient samples to investigate the frequency of JAK2, JAK3, and C-MPL mutations in TMD and ML-DS. The JAK3 gene was analyzed in 16 samples with hematologically confirmed diagnoses of TMD or ML-DS and one sample from a neonate that later developed ML-DS (groups A and B, Table 1). For group A, genomic DNA corresponding to all 23 coding exons of JAK3 was studied by denaturing high-performance liquid chromatography analysis (WAVE; Transgenomic, Omaha, NE). Genomic DNA corresponding to abnormally migrating polymerase chain reaction fragments was sequenced. In all samples, genomic sequence of the pseudokinase domain (exons 10–17) was determined. For samples within group B, exons 10–17 were analyzed by genomic sequencing. We did not detect potential pathogenetic JAK3 mutations in these 17 samples. Only 2 different heterozygous changes in 2 patients were identified in the coding sequence of JAK3 (corresponding to known single nucleotide polymorphisms (rs 2230589 and rs 35458530; Table 1). Other known single nucleotide polymorphisms were also found in introns in most cases.

No genomic JAK2 mutations (JAK2^617V>F by amplification refractory mutation system (ARMS) polymerase chain reaction and exon 12 mutations by denaturing high-performance liquid chromatography analysis) were present in 19 of 23 patient samples (subset of groups A, B, and C; there was insufficient DNA to analyze 4 of 23 patients). We did not find mutations in exon 10 of C-MPL (MPL W515L/K) in 8 samples (within groups B and C). Our failure to detect JAK2 mutations agrees with a previous report,⁵  whereas this is the first report showing absence of C-MPL mutations in TMD and ML-DS.

Three points are noteworthy. First, failure to detect acquired mutations in JAK2, JAK3, and C-MPL contrasts with detection of GATA1 mutations. This suggests either JAK2, JAK3, and C-MPL mutations are present in a smaller subpopulation of the sample or are not present at all. Second, our findings taken together with previous published data^5,6  indicate that the current incidence of JAK3 mutations in TMD is one in 15 and in ML-DS is 2 in 23. Furthermore, there is no evidence yet that mutations in JAK2 or C-MPL contribute to Down syndrome malignancies. Thus, although the total number of samples analyzed to date is small, we conclude that several different (epi)genetic events could be permissive for the progression of TMD to ML-DS and that these could individually occur at low frequencies.