Mutations of JAK2 and TET2, but not CBL are detectable in a high portion of patients with refractory anemia with ring sideroblasts and thrombocytosis

The WHO classification of 2008 characterized refractory anemia with ring sideroblasts associated with marked thrombocytosis (RARS-T) by the presence of less than 5% marrow blasts, 15% or more ring sideroblasts and a persistent platelet count over 450×10/L to be in line with the revised classification of essential thrombocytosis (ET). RARS-T forms a provisional entity with clinical and morphological features of both MDS and BCR-ABL negative myeloproliferative neoplasias (MPN).1 Although the identification of the JAK2 mutation was an important first step in distinguishing this entity from other hematologic diseases,2 further genetic characterization is needed.

Figure 1.Deletion of TET2. (A) SNP-A karyogram (Affymetrix Genotyping Console V.3.0.1) of patient 4 showing the deleted region on the long arm of chromosome 4 (region 4q24) (red box). Genomic variants and RefSeq genes within this region are shown. The approximate location of probe RP11-351K6 is shown in green. (B) Fluorescence in situ hybridization on two interphase nuclei (top) and one metaphase (bottom) detects only one signal for probe RP11-351K6 spanning the TET2 locus (patient 4). Blue: DAPI, Green: probe RP11-351K6 (BlueGnome, Cambridge, UK).

Table 1.Molecular details of TET2 mutations found in 5 out of 19 RARS-T patients.

In order to gain further insights into the genetic markers specific for RARS-T, we performed comprehensive cytogenetic and molecular genetic investigations including JAK2, TET2, CBL and MPL, markers common to MPN, as well as single-nucleotide polymorphism array analysis (SNP-A), which allows for the detection of both cryptic chromosomal changes as well as uni-parental disomy (UPD). We analyzed a total of 23 RARS-T patients with platelet counts ranging from 466×10/L to 1500×10/L (median: 680×10/L). The median age of patients at initial diagnosis was 76.1 years (range: 46.8 to 86.8). Compared to a total cohort of 1,674 MDS patients analyzed in our laboratory, there was no difference in age (median age 71.5 years; range: 4.8–92.3 years). However, as compared to 239 ET patients, t test revealed that RARS-T patients were significantly older (median 62.2 years; P=0.005). A melting-curve based LightCycler analysis detected the presence of JAK2 in 15 out of 19 analyzed patients (78.9%). Mutational ratios (JAK2/JAK2) ranged from 0.05 to 1.4 (median 0.44) with ratios above a value of 1 considered as homozygous. The JAK2 mutation was homozygous in 4/15 JAK2 positive patients. Interestingly, higher platelet counts showed a tendency to higher mutational ratios. Those patients negative for JAK2 were screened for MPL mutations. None of these patients was found to have mutated MPL. Also, none of 19 analyzed patients carried mutations in exons 8 and 9 of CBL, which have been detected in a small number of MPN patients.3 Conventional cytogenetics did not reveal any recurrent cytogenetic abnormalities in RARS-T patients. We found one case with loss of the Y chromosome, one patient with t(X;11)(p22;p13) and a potentially constitutional t(11;12)(q25;q14). Szpurka et al. reported UPD(9p) in one out of 18 RARS-T patients and UPD(1p) in 4 out of 18 RARS-T patients.4 To gain further data, we performed SNP arrays in 10 patients but were not able to identify UPD in chromosomal loci containing the JAK2 and MPL genes, 9p or 1p, respectively. Our SNP investigations did not detect additional recurrent chromosomal gains or losses nor did we observe recurring regions of UPD. However, one patient showed a deletion spanning a 1.3Mb region on the long arm of chromosome 4 (start: 105,497,200 bp from pter; end; 106,825,780 bp from pter) (Figure 1A). Interestingly, the deleted region contained TET2, a gene recently found to be altered in many subtypes of myeloid malignancies5–9 including 2 patients with RARS-T, of whom one showed a TET2 missense and the other a frameshift mutation.10 To further clarify the 4q24 deletion detected by SNP arrays, we performed fluorescence in situ hybridization (FISH). Twenty out of 100 analyzed interphase nuclei and three metaphases showed only one signal for the probe spanning the TET2 gene in one patient (Figure 1B). Interphase FISH with the TET2 probe was performed in 9 additional cases not analyzed by SNP arrays due to a lack of material. No additional case showing a deletion was detected. In addition to FISH, we performed TET2 sequencing in 19/23 RARS-T. TET2 mutations were detected in 5 out of 19 patients (26%), of which 3 out of 5 also presented mutated JAK2, whereas the remaining 2 out of 5 showed neither JAK2 nor MPL nor CBL mutations. The 5 patients showed 6 individually different TET2 mutations. Three were nonsense and two missense mutations. One patient displayed a frameshift mutation leading to a premature stop codon (Table 1). All mutations appeared to be heterozygous. The degree of homozygosity may, however, be underestimated due to a mixture of homozygous and healthy cells in the samples. As in other disease entities analyzed so far regarding TET2 mutations, no mutation “hotspot” could be detected in our RARS-T patients. In summary, RARS-T patients show a high frequency of both JAK2 and TET2 mutations. Together with the less common MPL mutations described by others,11,12 RARS-T presents a wide variety of mutations that overlap with the spectrum of mutations seen in MPN and other myeloid malignancies. Therefore, a combination of molecular markers including JAK2 and TET2 should be investigated to provide a more precise description of RARS-T as an independent entity.

Acknowledgments

We thank all clinicians for sending RARS-T samples to our laboratory for diagnostic purposes and for providing clinical information.