JAK2V617F as a Cooperative Mutation in AML with Trisomy 8 and t(8;21): A Comparative Study on 271 AML and 1103 CMPD Cases.

Over the last 1.5 years a number of studies have shown that JAK2 V617F mutations - even if not the initiating mutation of CMPD - are important in particular with respect to diagnosis and prognosis. However, JAK2 mutations have also been described in some cases of AML. To further work out the significance of JAK2 mutations we performed a comprehensive study in comparison to cytogenetics in 1103 CMPD and 271 AML cases. In CMPD the JAK2 mutation rate was 153/179 (85.5%) in polycythemia vera (PV), 123/205 (60.0%) in essential thrombocythemia (ET), 36/60 (73.3%) in chronic idiopathic myelofibrosis (CIMF), and 345/632 (54.6%) in unclassified CMPD matching data of previous studies. Of 73 available karyotypes 18 were aberrant (24.7%). The most frequent cytogenetic aberration in PV was trisomy 9 in 6 of 18 (33%) cases with aberrant karyotypes followed by 20q-deletions (3/18, 17%). In ET in only 5 of 64 (7.8%) cases aberrant karyotypes were detected which showed different non-recurrent aberrations. In CIMF 3/10 had 20q-deletions and 4/10 complex aberrant karyotpes. The AML cohort was unselected and contained a random distribution of karyotypes and FAB subtypes, de novo AML (n=178), s-AML with MDS prephase (n=26), s-AML after CMPD (n=27), t-AML after therapy of a preceeding malignancy (n=39). S-AML after CMPD were defined as an own entity as they may better be regarded as blast crisis of CMPD. The frequency of JAK2 mutations in s-AML after CMPD (16/27, 59.3%) was in the same range as in CMPD overall. A karyotype was available in 15 cases: 3 were normal, 10 complex aberrant (66.7%), five of these contained a trisomy 9 or trisomy 9p, two had other aberrations. No JAK2 mutation was detected in s-AML after MDS. In de novo AML the rate of JAK2 mutated cases was 6.2% (11 of 178 cases). This is in the range of other activating mutations in de novo AML like FLT3-TKD and NRAS mutations. Similarly 2 of 37 (5.4%) pts with t-AML were JAK2 mutated. The pattern of chromosome aberrations in de novo AML was characterized by an extremely high frequency of trisomy 8: 7 of 11 de novo cases with JAK2 mutation (63.6%) had +8, four of these as sole aberration and three in combination with + 9. Two were normal and two complex aberrant (18% each). In comparison, overall 7 of 30 cases (23.3%) with trisomy 8 sole or +8,+9 were JAK2 mutated. Thus JAK2 may be a newly defined cooperating mutation to trisomy 8 whereas a gain of trisomy 9 is known as a typical progression marker in JAK2 mutated cases. Both JAK2 mutated t-AML were t(8;21) positive. In total 21 cases with t(8;21) were analyzed, 3 with t-AML and 18 with de novo AML. Thus 9.5 % of t(8;21) were JAK2 mutated. This may be an event restricted to t-AML. Based on this study we suggest that JAK2V617F in CMPD and AML after CMPD contributes to progression of disease and is followed by chromosomal abnormalities. In CMIF and AML after CMPD most cases have complex aberrant karyotypes suggesting that JAK2 mutation is one step in multistep mutagenesis in which chromosomal instability seems to play a major role. The pathophysiological role in de novo AML seems to be different with JAK2 mutations occurring less frequently overall. They were highly correlated 1) with trisomy 8 in de novo AML and 2) with t(8;21) in t-AML whereas none of 18 de novo AML1-ETO cases were JAK2 mutated. From this we conclude that in AML JAK2 takes over a function as a typical type 1 mutation like FLT3, KIT or RAS at least in combination with +8 or with t(8;21).