IKZF1 Deletion Is Strongly Associated with Risk of Relapse in Intermediate Risk Group in JACLS ALL02 Cohort

Despite the progress of the current therapy, approximately 20 % of pediatric patients with B-cell-presursor (BCP)-ALL experience relapse who had no conventional adverse prognostic factor. Recently, alteration of the IKZF1 gene has been reported to be associated with a poor outcome in pediatric BCP-ALL without BCR-ABL. In addition, it is also reported that a significant proportion of these cases with the alteration of IKZF1 shared the point mutation of JAK2 and over-expression of cytokine receptor-like factor 2 (CRLF2). Herein, in order to assess the prognostic value of these genetic abnormalities in Japanese cohort, we conducted genetic analysis of IKZF1, JAK2 and CRLF2 in pediatric BCP-ALL.

Diagnostic bone marrow or peripheral blood samples of 215 pediatric BCP-ALL patients treated according to Japan Association of Childhood Leukemia Study (JACLS) ALL02 protocol from April, 2002 to May, 2008 were examined in this study. All patients were categorized into high risk (HR) group defined as follows;(1) initial white blood cell counts more than 10,000 /μl, (2) age at diagnosis is > 10 years and (3) good response to initial predonisolone (PSL) treatment ( blast counts in peripheral blood is less than 1,000 /μl after one week PSL treatment). Ph+ALL and infantile ALL patients were excluded in this protocol. The patients with Down syndrome were also excluded in this genetic analysis. The deletion of IKZF1 was determined using multiplex ligation-dependent probe amplification (MLPA) in 212 patients whose diagnostic DNA samples were available. The expression of isoform of IKZF1 was determined by RT-PCR in 113 patients whose diagnostic RNA samples were available. The expression level of CRLF2 was also determined by real time RT-PCR in 112 patients and over-expression was defined as over ten times more than median expression value. The presence of P2RY8-CRLF2 fusion was examined by RT-PCR or MLPA in the patients with CRLF2 over-expression or IKZF1 deletion. JAK2 mutations were also determined by direct sequencing of the exon 16, 20 and 21 in the patients with IKZF1 deletion.

The deletion of IKZF1 gene was present in 19 of 212 (9.0 %) patients. In detail, the mono-allelic deletion of entire IKZF1 gene was present in 10 of 19 patients. On the other hand, the expression of dominant-negative IK6 isoform was present in 9 of 112 (8.0%) patients including 12 patients with IKZF1 deletion. The expression of IK6 was present in 4 of 12 (33.3%) IKZF1 deleted patients. Interestingly, 5 of 9 patients with IK6 expression had no alteration of IKZF1 gene. In terms of CRLF2, over-expression was detected in 16 of 112 patients (14.3 %). However, P2RY8-CRLF2 fusion was not detected in these 16 patients with altered CRLF2 expression. Strikingly, none of the patients with IKZF1 deletion (n=19) had either P2RY8-CRLF2 fusion or JAK2 exon 16, 20 and 21 mutation. Patients with IKZF1 gene deletion had significantly worse relapse rate than those without IKZF1 deletion (7/19 vs 22/193, p<0.01). On the contrary, none of the patients with IK6 expression experienced relapse.

This study confirmed that the presence of IKZF1 deletion was strongly correlated with risk of relapse in intermediate risk group in JACLS ALL02 cohort. Thus, we expect that IKZF1 deletion is an independent predictor of treatment outcome and represents a candidate of prognostic marker to be integrated in future algorithms for early risk stratification in pediatric BCP-ALL. Strikingly, point mutation of JAK2 exon 16, 20 and 21 or P2RY8 -CRLF2 fusion was rarely present even in BCP-ALL patients with IKZF1 deletion. There might be unrevealed class I mutation cooperating with IKZF1 alteration in Japanese BCP-ALL cohort.

No relevant conflicts of interest to declare.