The Outcome of Childhood B-Cell Precursor ALL with ETV6-RUNX1 Treated with Japan Association of Childhood Leukemia Study (JACLS) ALL-02 Trial

Background: It is well known that ETV6-RUNX1 positive pediatric B cell precursor acute lymphoblastic leukemia (B-ALL) shows excellent outcome when they are treated with contemporary chemotherapeutic regimens. However, the prognosis of ETV6-RUNX1 positive B-ALL wasn't determined in Japanese pediatric cohort which was uniformly treated. Japan Association of Childhood Leukemia Study (JACLS) ALL-02 protocol achieved 83.7±1.1% of 4-year event free survival of pediatric B-ALL (Hasegawa et al. 58^th ASH annual meeting) which was comparable to that in major recent pediatric ALL clinical trials, although risk stratification based on minimal residual disease (MRD) was not employed. Herein, we report the outcome and analyze the prognostic factors of ETV6-RUNX1 positive B-ALL treated with JACLS ALL-02 protocol.

Patients and treatment: Between April 2002 and March 2008, newly diagnosed de novo ETV6-RUNX1 positive ALL with 1-18 years of age were enrolled to JACLS ALL-02 trial. Patients were stratified into 3 groups: standard-risk (SR), high-risk (HR), extremely high-risk (ER), based on initial prednisolone (PSL) response and the modified National Cancer Institute(NCI) workshop criteria. Prednisolone poor response (PPR) was determined after 7 days of monotherapy with prednisone and one intrathecal dose of methotrexate. PSL good responders (PGR; < 1,000 blasts/μL) were divided into SR and HR according to the modified NCI workshop criteria by WBC 10K and age 10, and received conventional therapy. BCP-ALL with PPR (≥ 1,000 blasts/μL) were assigned to ER and received intensified post-induction therapy. Estimation of event free survival (EFS) and overall survival (OS) was performed using the Kaplan-Meier method and the differences were compared using the log-rank test. Multivariate analysis was performed using a Cox regression model.

Results: 205 newly diagnosed ETV6-RUNX1 positive ALL patients were enrolled in JACLS ALL-02 clinical trial, which included 108 patients in SR group (52.7%), 89 in HR group (43.3%) and 8 in ER group (3.9%). The median age at diagnosis was 4.3 years (range, 1.0 to 15.0 years) and the median WBC count was 6,800/μL (range, 370 to 384,700 /μL). The median follow-up time was 6 year and 6 months. 4y/8y EFS and OS (±SE) are 92.5±2.6/ 89.8±2.5% and 96.5±1.5/ 95.5±1.6%, respectively. 4y / 8y EFS of the patients of NCI-SR (n=167) and NCI-HR (n=38) are 95.6±1.6% / 91.3±2.7% and 89.5±5.0% / 83.0±6.4%, respectively (log rank p=0.04). 4y / 8y OS of the patients with PSL good response (n=197) was 97.9±1.1% / 97.0±1.3%, while those of the patients with PPR (n=8) was 87.5±11.7% / 58.3±18.6% (log rank p<0.001). Eighteen of 205 (8.8%) patients experienced relapse. According to BFM risk stratification of relapsed ALL (S1 to S4), 13 of 18 patients were in S2 and ten of them were rescued by salvage therapy. However, 5 patients in S3 or S4 group died of disease progression. Although two patients showing PPR belonged to S2, all of them died of disease progression. Multivariate analysis identifies NCI-HR (HR=2.92, p=0.05) as a predictor of poor prognosis in EFS, and PPR (HR=12.0, p<0.01) as predictor of poor prognosis in OS.

Discussion: Although JACLS ALL-02 protocol didn't employ MRD based risk stratification, good outcome of the 4y EFS and OS (92.5±2.6 % and 96.5±1.5 %) was achieved for ETV6-RUNX1 positive B-ALL patients. Because NCI-SR and PGR were predictors of excellent outcome for ETV6-RUNX1 positive B-ALL, less intensive treatment should be explored for this group. On the other hand, NCI-HR and PPR indicated poor outcome. We hope that MRD based monitoring might improve outcome of this subgroup in our next clinical trial.