3717

JAK2

JTK10;Janus kinase 2;JAK2;Janus kinase 2

JAK-2|Janus kinase 2 (a protein tyrosine kinase)|tyrosine-protein kinase JAK2

9p24

protein-coding

We studied a series of 68 subjects diagnosed with childhood acute myeloid leukemia (AML) using conventional cytogenetics and fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR) to analyze mutations in FLT3 and NPM1 genes, and/or array comparative genomic hybridization (CGH). Cytogenetic/FISH abnormalities were observed in 71% of subjects, FLT3-ITD mutations in 15%, and NPM1 mutations in 13%. The array CGH alterations (average 3.6 per case) were observed in 96% of the tested subjects. The most frequent alterations were gains of 8q24.3 and 11p15.5-p15.4 in 16% of the samples. Six genes (AKT1, RUNX1, LTB, SDC1, RUNX1T1, and JAK2) from the imbalanced regions have been reported to be involved in AML, whereas other 30 cancer genes, not previously reported in an AML context, were identified as imbalanced. They probably correspond to non passenger alterations that cooperate with the recurrent translocations. The clinical data and genetic changes were tested to find out the possible association with prognosis. Genomic instability (four or more genomic imbalances) was correlated with poor patient outcome (p = 0.029).

BACKGROUND: Genome-wide analysis studies have demonstrated that IKZF1, CRLF2, and JAK2 gene alterations correlate with poor prognosis in pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL). However, the prognostic significance forthese gene alterations has not been clarified in Japanese patients. PROCEDURE: Atotal of 194 patients with BCP-ALL enrolled in the Japanese children's cancer & Leukemia Study Group ALL 2004 clinical trial were assessed for the presence of three different gene alterations: IKZF1 deletions, CRLF2 expression and JAK2 mutation. RESULTS: IKZF1 deletions and CRLF2-high expression were identified in 22 of 177 (12%) patients and in 15 of 141 (11%) patients, respectively. However,JAK2 R683 mutation was detected only one of 177 patients. The 4-year event-free survival (4y-EFS) was different when comparing patients with or without IKZF1 deletions (68.2% vs. 85.2%; P = 0.04) and was also different when comparing patients with different CRLF2 expression levels (high, 66.7% vs. low, 88.1%; P =0.03). The differences in 4y-EFS were statistically significant in patients withALL in the National cancer Institute (NCI)-high risk group (HR-ALL) (IKZF1 deletions: yes, 58.3% vs. no, 87.0%, P = 0.02; CRLF2 expression: high, 55.6% vs.low, 85.3%, P = 0.04) but not in patients with ALL in the NCI-standard risk group (SR-ALL; IKZF1 deletions: yes, 80.0% vs. no, 84.4%, P = 0.75; CRLF2 expression: high, 83.3% vs. low, 89.2%, P = 0.77). Coexistence of IKZF1 deletions and CRLF2-high expression associated with poor outcomes. CONCLUSIONS: IKZF1 deletions and CRLF2-high expression predicted poor outcomes in patients with HR-ALL but not in patients with SR-ALL in our Japanese cohort. Pediatr Blood cancer 2013;60:1587-1592. (c) 2013 Wiley Periodicals, Inc.#CI- Copyright (c) 2013 Wiley Periodicals, Inc.