High frequency of the JAK2 V617F mutation in patients with thrombocytosis (platelet count > 600 × 10⁹/L) and ringed sideroblasts more than 15% considered as MDS/MPD, unclassifiable

We wish to confirm recent reports showing that JAK2 V617F^1-5  is commonly detectable in patients with thrombocytosis who also have ringed sideroblasts in their bone marrow.^6-9  For such patients, the WHO classification provides the category of “myelodysplastic/myeloproliferative disease, unclassifiable,” including a subtype of refractory anemia with ringed sideroblasts associated with marked thrombocytosis (RARS-T),¹⁰  which requires at least 15% ringed sideroblasts in the marrow and a platelet count of more than 600 × 10⁹/L.

These criteria were fulfilled by 21 of 2986 patients in the Düsseldorf MDS Registry. For 6 of these patients, archived material was available. Another 4 cases were identified at the Hematology Laboratory of Sint-Jan's Hospital, Brugge, Belgium, between 1998 and 2006. Samples from 10 patients with RARS and normal platelet count served as a control group. Archived material consisted of unstained bone marrow smears, frozen pellets of nucleated bone marrow cells, and DNA extracted from fresh bone marrow aspirates. Two methods were used for mutation detection. First, all samples were analyzed by allele-specific polymerase chain reaction (PCR), according to Baxter el al.¹  The assay was slightly modified and generated a 203-bp mutant PCR product and a 364-bp product from both mutant and wild-type alleles serving as an internal PCR control. The detection limit of the assay was validated as 1% mutant allele. In addition, direct sequencing of JAK2 exon 12 genomic DNA was performed in both directions on all samples.

Both methods yielded negative results for JAK2 V617F in the control group. In patients with RARS-T, allele-specific PCR was, as expected, the more sensitive technique, yielding a positive result in 9 of 10 patients. Direct DNA sequencing detected JAK2 V617F in 7 patients, suggesting that the proportion of mutant DNA was less than 15% in 2 of the patients (nos. 2637 and G329) tested positive by PCR.

The patient with negative results by both methods (no. 2644) had platelet counts ranging between 425 × 10⁹/L and 664 × 10⁹/L, the latter being the only one higher than 600 × 10⁹/L. That threshold, to be exceeded repeatedly, seems appropriate for making a diagnosis of RARS-T.

Three of 9 PCR-positive patients harbored a clone homozygous for JAK2 V617F because DNA sequencing indicated a proportion of mutant DNA of higher than 50% (Table 1) In the other patients, the proportion was lower than 50%, which did not allow us to distinguish between an intermediate proportion of heterozygous cells and a low percentage of homozygous cells, the latter scenario being less likely.

Four of 10 patients had an elevated WBC count. This is unusual for myelodysplastic syndrome (MDS) and speaks for classifying these cases as myeloproliferative disease (MPD). Since it is hard to decide whether the dysplastic feature of ringed sideroblasts on the one hand, or elevated cell counts and the molecular finding of JAK2 V617F on the other hand, should prevail when trying to classify such disorders, the WHO category of MDS/MPD, unclassifiable, is perhaps more than a makeshift solution, as it may properly reflect the mixed underlying biology.

Finally, the lack of leukemic transformation in our patients, as well as in similar patients reported previously, suggests that JAK2 V617F is a relatively benign proliferation signal.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Norbert Gattermann, Klinik für Hämatologie, Onkologie u klin Immunologie, Universitätsklinikum Düsseldorf, Moorenstr 5, D-40591 Düsseldorf, Germany; e-mail: gattermann@med.uni-duesseldorf.de

We thank Sabrina Pechtel and Nadine Mestdagh for expert technical assistance. Part of this work was funded by vzw Wetenschappelijk Fonds Hematologie, AZ Sint-Jan Brugge.