Myelofibrotic transformation in essential thrombocythemia. Author reply

A JAK2-V617F mutation was found in 3 of 45 (6.7%) patients with t(8;21) acute myeloid leukemia (AML), whereas only one of 137 (0.7%) patients with de novo AML other than t(8;21) had the same mutation (p=0.047). We examined the clinical significance of KIT, FLT3 and JAK2 mutations as a collective group. There was a significant difference in the cumulative incidence of relapse: 77% in the 21 patients with the mutations and 26% in 19 lacking mutations respectively (p=0.0083). Our study highlights the importance of JAK2 mutations in addition to KIT and FLT3 mutations as a prognostic factor in t(8;21) AML patients.

RUNX1(AML1)-RUNX1T1(MTG8) generated by t(8;21)(q22:q22) contributes to leukemic transformation, but additional events are required for full leukemogenesis.1,2 Mutations in the receptor tyrosine kinases (RTK) including the KIT and FLT3 genes are the genetic events that appear to cause acute myeloid leukemia (AML) harboring t(8;21) and are associated with unfavorable prognosis.3,4 The activating missense mutation in the pseudokinase domain of the JAK2 cytoplasmic tyrosine kinase has been identified in a significant proportion of patients with myeloproliferative disorders.5 Although the same somatic mutation has been found in a small number of AML patients, a relatively high incidence of JAK2-V617F mutation is often seen in de novo and therapy-related t(8;21) AML patients.6–10 Nevertheless, whether JAK2-V617F mutation is associated with other biological parameters including clinical prognosis in patients with t(8;21) AML remains to be fully determined.

To examine its biological and prognostic impact, we studied the JAK2 mutation in 45 patients with de novo t(8;21) AML. Approval for this study was obtained from the Institutional Review Board of Kumamoto University School of Medicine. The results of KIT, FLT3, N-RAS, K-RAS and PDGFRα mutations in 37 of the 45 patients have been reported previously.3 Of the 45 patients, activating mutations in KIT and internal tandem duplications in FLT3 were observed in 18 (40%) and 3 (6.7%) respectively. Mutations of JAK2-V617F were identified by allele specific RT-PCR and direct sequencing.11 We detected the heterozygous JAK2-V617F mutation in 3 patients (6.7%) with t(8;21) AML, which was consistent with previous studies.6–10 None of the 3 t(8;21) AML patients had a history of previous myeloproliferative disorders. No mutations other than V617F were found in the exons 12–14 of JAK2. Among 137 patients with de novo AML other than t(8;21), there was only one patient who had JAK2 mutation (p=0.047). This patient had M2 with 46,XY,add(7)(q11),del(20)(q13). Thus, the present study confirmed that the JAK2 mutation is highly associated with t(8;21) AML.

Although the occurrence of KIT and FLT3 mutations was mutually exclusive in t(8;21) AML patients,3 one patient harboring a JAK2 mutation also had a KIT mutation and the other patient had a K-RAS mutation (Table 1). Although we cannot exclude the possibility that two different subclones in leukemic cells had each mutation, it is also likely that the same leukemic cells carry both mutations because heterozygous JAK2 and KIT or K-RAS mutations are identified as equivocal peaks in the electro-pherogram of direct sequencing (data not shown). It is of note that a high prevalence of co-operating mutations of FLT3, KIT, or N-RAS in AML patients with the JAK2 mutation has been reported.7–9 In the current study, a total of 23 (51%) patients had mutations in KIT, FLT3 and JAK2, suggesting that activating mutations in the RTK and JAK2 play a critical role as a secondary event leading to the development of t(8;21) AML.

Figure 1.Cumulative incidence of relapse (A) and overall survival ( B ) in patients with t(8;21) acute myeloid leukemia by mutations in KIT, FLT3 and JAK2. Estimation of survival distributions was performed using the Kaplan–Meier method and the differences were compared using the log-rank test.

We examined the clinical significance of KIT, FLT3 and JAK2 mutations as a collective group because the present study was limited to a small number of JAK2 mutated cases for the comparison of clinical features, and these mutations activate the same STAT signal transduction pathway and belong in the same class I mutation.2 There was no significant relationship between the mutations and age, sex, leukocyte counts, platelet counts, CD56 expression, or additional chromosomal aberrations. However, t(8;21) AML patients with an activating mutation in KIT, FLT3 and JAK2 had significantly greater marrow blast percentages and serum lactate dehydrogenase levels than those without a mutation (data not shown) . Considering that the JAK2 mutation confers a proliferative and survival advantage on hematopoietic cells,5 these clinical profiles appear to be associated with these mutations.

Table 1.Clinical profiles of t(8;21) acute myeloid leukemia patients harboring the JAK2 mutation.

A total of 44 patients received intensive chemotherapy based on the Japan Adult Leukemia Study Group (JALSG) protocols in the AML87, AML89, AML92, AML95 and AML97 studies.12 Although patients were treated with different schedules, all received regimens consisting of anthracyclines and cytarabine as induction therapy. Cytarabine plus one of the anthracyclines, high-dose cytarabine, or allogeneic hematopoietic stem cell transplantation (HSCT) was used as post-remission therapy. Patient 1 carrying both KIT and JAK2 mutations did not respond to multiple induction chemotherapies including high-dose cytarabine therapy (Table 1). Patient 2 with the JAK2 and K-RAS mutations achieved a complete remission (CR) but later relapsed. Patient 3 received allogeneic HSCT during the first CR and continued in CR. Twenty-one out of 23 (91%) patients with the mutations achieved CR, while 19 out of 21 (90%) patients lacking mutations obtained CR (p=0.9240). On the other hand, there was a significant difference in the cumulative incidence of relapse: 77% in the 21 patients with the mutations and 26% in 19 lacking mutations respectively (p=0.0083) (Figure 1A). It is likely that the poor outcome cannot be attributable only to JAK2 mutation in patients with a KIT or K-RAS mutation although JAK2 mutation together with other mutations may confer additive effects on the clinical outcome. Illmer et al.8 also showed that 4 of 5 t(8;21) or inv(16) AML patients with a JAK2 mutation had early relapses within 20 months after diagnosis. Taken together, these results suggest that mutations in the JAK2, KIT and FLT genes are associated with unfavorable clinical outcome in patients with t(8;21) AML.

Our study also implies that patients with RTK and JAK2 mutations may benefit from allogeneic HSCT. Three patients with mutations received allogeneic HSCT after relapse and have achieved continuous second CR. Three patients in each group also received allogeneic HSCT at the first CR. As a consequence, 6 out of 9 patients with AML harboring KIT, FLT3 and JAK2 mutations who continued CR received allogeneic HSCT. When patients who underwent HSCT were censored at the date of the HSCT, the 6-year overall survival in patients with mutations was 25% compared to 62% in those without mutations (p=0.1368) (Figure 1B). These findings are of significant clinical import as activating mutations in KIT, FLT3 and JAK2 could be potential therapeutic targets for specific tyrosine kinase inhibitors and JAK2 pathway inhibitors in patients with t(8;21) AML harboring the mutations.

Acknowledgments

this work was supported in part by Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sport, Science and Technology, and Grants-in-Aid for Cancer Research from the Japanese Ministry of Health, Labor and Welfare.