Hyperleukocytosis and outcomes in pediatric B-cell acute lymphoblastic leukemia: A report from the REDIAL Consortium

Problem Statement
Hyperleukocytosis, defined as a presenting white blood cell (WBC) count exceeding 100,000/µL, is a known prognostic risk factor in pediatric acute lymphoblastic leukemia (ALL). While current risk-adapted therapies escalate treatment for patients with WBC counts >50,000/µL, the specific prognostic significance of hyperleukocytosis (WBC >100,000/µL) in the context of contemporary therapy remains unclear. It is unknown whether this extreme elevation independently predicts inferior outcomes after adjusting for other high-risk features such as age, cytogenetics, and minimal residual disease (MRD) status, or if it merely reflects a higher burden of disease that is adequately managed by existing risk-stratification protocols. Furthermore, the relationship between WBC count and outcomes as a continuous variable, rather than a discrete threshold, requires further exploration in real-world cohorts.

Methodology
The study utilized a multi-institutional retrospective cohort from the REDIAL (REducing Disparities in Acute Leukemia) Consortium, comprising 1,826 pediatric patients (ages 1–18) diagnosed with B-cell or T-cell ALL between 2005 and 2017 across five US centers. All patients were treated on or according to Children's Oncology Group (COG) protocols.

• Data Collection: Demographic and clinical variables included age, sex, race/ethnicity, BMI, WBC count at diagnosis, cytogenetic risk categories, central nervous system (CNS) disease status, and end-of-induction (EOI) MRD.
• Statistical Analysis: The authors performed separate analyses for B-ALL and T-ALL. They employed multivariable Cox proportional hazards models to assess Event-Free Survival (EFS) and Overall Survival (OS), and logistic regression for EOI MRD positivity and CNS status. Directed acyclic graphs guided covariate selection.
• Adjustments: Models were adjusted for age, cytogenetic risk, CNS status, and treatment site.
• Advanced Modeling: To assess dose-response relationships, the authors analyzed log-transformed WBC counts using restricted cubic splines to evaluate non-linear associations. Subgroup analyses were conducted stratified by MRD status and cytogenetic risk.

Key Results

• Prevalence: Hyperleukocytosis occurred in 12% (211/1,826) of the cohort, with 121 cases in B-ALL and 90 in T-ALL.
• B-ALL Outcomes:
  • Survival: Patients with hyperleukocytosis had significantly inferior 5-year EFS (65% vs. 89%) and OS (78% vs. 93%) compared to non-hyperleukocytosis patients.
  • Independence: After multivariable adjustment, hyperleukocytosis remained an independent predictor of EOI MRD positivity (OR 2.53), inferior EFS (HR 2.44), and inferior OS (HR 2.00).
  • Dose-Response: A continuous, non-linear dose-response relationship was observed. Risk of adverse outcomes increased progressively with WBC count, with a sharp rise in MRD positivity risk starting around 10,000/µL.
  • Subgroups: Inferior outcomes persisted across all cytogenetic risk categories and MRD strata. Even among MRD-negative patients, those with WBC >100,000/µL had worse EFS and OS.
  • Cytogenetics: Hyperleukocytosis was associated with a lower frequency of favorable markers (ETV6::RUNX1, double trisomies) and a higher frequency of high-risk markers (KMT2A rearrangements, BCR::ABL1). However, the prognostic impact of hyperleukocytosis remained significant even after adjusting for these specific genetic lesions.
• T-ALL Outcomes: In contrast to B-ALL, hyperleukocytosis showed no significant association with EFS, OS, or MRD positivity in T-ALL patients after multivariable adjustment.

Significance and Claims
The paper claims that hyperleukocytosis identifies an aggressive B-ALL phenotype that predicts treatment failure and inferior survival, independent of standard risk stratification factors including cytogenetics and MRD status. The authors assert that the association is not merely a result of higher disease burden but reflects a distinct biological aggressiveness, as outcomes remain poor even in patients who achieve MRD negativity.

Key contributions include:

1. Validation of Continuous Risk: The study provides evidence that risk increases continuously with WBC count rather than at discrete thresholds, suggesting current risk stratification (which often uses a 50,000/µL cutoff) may fail to identify a subset of patients at increased risk for treatment failure.
2. Independence from MRD: The finding that hyperleukocytosis predicts poor outcomes even in MRD-negative patients suggests it may be a marker of resistant disease biology or inadequate cytoreduction that is not fully captured by standard induction response metrics.
3. Subtype Specificity: The lack of association in T-ALL suggests distinct biological mechanisms between B-ALL and T-ALL regarding the impact of high blast burden.

The authors conclude that while hyperleukocytosis is a strong independent predictor of poor outcomes in pediatric B-ALL, the underlying genetic and pathophysiological mechanisms remain incompletely understood. They suggest these patients may benefit from novel therapeutic approaches, though they explicitly note that the prognostic significance in the era of frontline immunotherapy (e.g., blinatumomab) remains unknown due to the study's retrospective nature and treatment protocols.