Absolute Lymphocyte Counts at the End of Induction Is a Prognostic Indicator in Childhood Acute Lymphoblastic Leukemia

BACKGROUND: Recently, several studies have demonstrated that absolute lymphocyte counts (ALC) after induction therapy predicted treatment outcome. To address this issue, we here assessed the impact of the ALC at the end of induction therapy on outcomes in childhood ALL.

METHODS: We reviewed 141 cases of pediatric ALL with 1-21 years of age treated on the Japan Association Childhood Leukemia Study group ALL-02 series of treatment trials between 2002 and 2013. Patients with Philadelphia chromosome-positive ALL were excluded. Variables retrospectively analyzed included ALC at several time points during remission induction, age at diagnosis, gender, initial white blood cell count (WBC), cytogenetics, immunological phenotype, stratified risk, treatment response for bone marrow (the percentage of blasts at day 15), and outcome. Events in the analysis of event-free survival (EFS) included induction failure, death, relapse and secondary malignant neoplasm. The comparison of categorical variables between groups was performed by chi-square test. The probability of EFS and overall survival (OS) were analyzed with the use of the Kaplan–Meier method and a stratified log-rank test. A multivariate analysis of survival was performed with the use of a Cox proportional-hazard model to evaluate the treatment effect with adjustment for stratification factors.

RESULTS: The subjects included 121 of B-precursor ALL, 10 of T cell ALL, and 4 of acute mixed lineage leukemia/ acute unclassified leukemia. We found high WBC count at diagnosis (>100K/microL) and slow early responder for bone marrow at day 15 to be an unfavorable prognostic indicator, and also the ALC at the end of induction (day29) to be a statistically significant predictor of improved OS and EFS in our cohort. Patients with ALC ≥ 800/microL had a superior 5-year overall survival (100 ± 1.7% vs 88.1 ± 4.3 %, p=0.0001) and EFS (98.3 ± 1.7% vs 81.8 ± 5.0 %, p=0.0001). Multivariate analysis demonstrated that ALC at day29 was an independent, clinically significant predictor of improved EFS and OS after controlling WBC at diagnosis, gender, age at diagnosis, and cytogenetics. Multiple regression analysis adjusting for initial WBC count, peripheral blast counts at day8, and cytogenetics, also revealed an independent relationship (p=0.005) between treatment response (the percentage of blasts at day 15) and ALC at day29.

CONCLUSIONS: ALC is a simple, statistically significant prognostic factor in childhood ALL that may refine current risk stratification algorithms.