6 Hochhaus A, Kreil S, Corbin AS, La Rosee P, Muller MC, Lahaye T et al. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy. Leukemia 2002; 16: 2190–2196. 7 Nalla VK, Rogan PK. Automated splicing mutation analysis by information theory. Hum Mutat 2005; 25: 334–342. 8 Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kuriyan J et al. Structural mechanism for STI-571 inhibition of abelson tyrosine kinase. Science 2000; 289: 1938–1942. The JAK2 V617F mutation is rare in RARS but common in RARS-T Leukemia (2006) 20, 2060–2061. doi:10.1038/sj.leu.2404373; published online 24 August 2006 In the majority of patients the distinction between myelodys- plastic syndromes (MDSs) and chronic myeloprololiferative disorders is made relatively easily by assessment of clinical, laboratory and morphological characteristics of the peripheral blood and bone marrow. However, the existence of uni- or mulitlineage dysplasia and hyperplastic myelopoiesis in the same patient is well recognized and classified by the World Health Organization (WHO) as myelodysplastic/myeloprolifera- tive disease, unclassifiable. 1 Within this group is a provisional entry referred to as refractory anaemia with ringed sideroblasts associated with marked thrombocytosis (platelets 4600  10 9 /l) (RARS-T). 1 The relationship between dysplasia and the proliferation is not entirely clear, but the recent discovery of a single somatic point mutation in the pseudokinase domain of the Janus kinase 2 (JAK2) resulting in a codon change V617F represents a potential major breakthrough in the pathogenesis of myeloproliferation. Mutation at this site results in the constitutive activation of the JAK2 kinase and thus the activation of the JAK2 pathway. This mutation is present in the majority (495%) of patients with polycythaemia vera and approximately half of those with primary thrombocythaemia and myelofibrosis 2 and others. We have recently reported a high incidence (61%) of the V617F mutation in patients with Budd Chiari syndrome who lack features of a frank proliferative disorder. 3 The prevalence of JAK2 V617F mutation in atypical MPDs or MDS is generally low. 4 Identification of this mutation in MDS identifies a subset of patients with a prominent proliferative component. 5 The lack of a precise molecular definition for most cases of MDS, with the exception of 5q syndrome, testifies to the heterogenous nature of this disorder. We analysed 70 patients from two European centres with a diagnosis of MDS with ringed sideroblasts of X15% (as defined by the WHO classification), for the presence of the JAK2 V617F mutation. A summary of patient’s characteristics is shown in Table 1. DNA samples for analysis were obtained from archived bone marrow aspirate slides or from fresh peripheral blood taken from patients presenting to our centres. Ethical approval was obtained before the study commenced. Genomic DNA was prepared using standard methods (Qiagen, House, Crawley, West Sussex, UK). A modified allele–specific PCR (AS-PCR) was used to screen DNA samples. 1 JAK2–mutated DNA samples and control DNAs were further subjected to pyrosequencing (Figure 1) to confirm the presence of the mutation. 2,3 Three out of four JAK2 mutated DNA samples were confirmed mutated by pyrosequencing. However, one sample appeared fully wild–type at the JAK2 allele. This conflicting data can be explained by the fivefold increased sensitivity of the AS-PCR above the pyrosequencing assay. It is likely that the clone size in the pyrosequencing normal patient is much smaller making it indistinguishable from the background in the pyrosequencing assay. The JAK2 V617F mutation was detected in 4/70 (5.7%) by AS- PCR, which is comparable to previously reported studies. 5–7 The mean platelet count is higher in the mutated cases at 699  10 9 vs 255  10 9 /l (Po0.0001). The JAK2–positive patients were younger at presentation; 60 vs 70 years (P ¼ 0.031). No statistically significant difference in the mean haemoglobin concentration (9.6 vs 10 g/dl, P ¼ 0.647), white cell count (8.39  10 9 vs 6.8  10 9 /l, P ¼ 0.496) or bone marrow cellularity (90 vs 78%, P ¼ 0.357) was noted between JAK2 V617F– mutated compared to wild-type cases, respectively. The number of cases within the RARS cohort presenting with thrombocytosis (RARS-T) is 6/70 (8.6%). RARS-T in this cohort is defined as platelet count 4500  10 9 /l. On analysis of patients with RARS-T, the incidence of JAK2 V617F mutation is significantly higher with 4/6 (67%) cases demonstrating the mutation. Of the 66 JAK2 wild–type cases, only 2/66 (3%) presented with platelet count 4500  10 9 /l. The median age of the RARS-T subgroup was 62 (52–71) years with equal male to female patients distribution. Cytogenetic analysis in all cases of RARS-T was normal. Of the JAK2–mutated cases, 3/4 are alive at Table 1 Summary of patient characteristics JAK2 WT N ¼ 66 JAK2 mutant N ¼ 4 Median age in years (range) 70 (53–86) 60 (52–69)* Sex Male 45 2 Female 21 2 FBC at diagnosis Hb (g/dl) 10.0 (7.2–16.9) 9.6 (7.4–11.9) WBC (  10 9 /l) 6.8 (2.35–28.92) 8.39 (5.2–16.2) Platelets (  10 9 /l) 255 (27–835) 699 (525–1099)* Bone marrow cellularity (%) 78 (40–100) 90 Ringed sideroblasts (%) 50 (20–100) 58 (45–70) Median time to follow-up (months) 46 (0.5–204) 33 (3–58) Transformation No 56 3 AML 4 1 RAEB 6 0 Alive at follow-up Yes 35 3 No 31 1 Abbreviations: AML, acute myeloid leukaemia; FBC, full blood count; Hb, haemoglobin; JAK2, Janus kinase 2; RAEB, refractory anaemia with excess blasts; WBC, white blood cell; WT, wild-type. *Statistically significant difference (Po0.05). Letters to the Editor 2060 Leukemia