Background: Risk stratification and subsequent optimization of therapy for acute lymphoblastic leukemia (ALL) is indispensable. Early response to induction therapy has been known as an important prognostic factor; however, impact of myelosuppression during induction phase on prognosis has not been assessed. D-index that reflects both intensity and duration of neutropenia was originally developed to estimate risk of invasive fungal infection in neutropenic patients under chemotherapy. Primary aim of this study was to assess the association between D-index during induction phase and prognosis of children and adolescents with B-cell precursor (BCP) ALL.

Methods: Sixty-nine children or adolescents diagnosed with BCP-ALL treated at our hospital were consecutively enrolled from April 2007 to March 2017. Their median age was 5 years (range, 1-23). Children less than 1 year of age were excluded from this study since they were treated with an infant ALL protocol. We performed multivariate analysis to estimate the effect of D-index and other known factors including age, white blood cell (WBC) count at onset, and early response to prednisolone (the number of day 8 peripheral blasts) on poor prognosis. The D-index was calculated based on a graph that plotted the absolute neutrophil counts over the course of the episode of neutropenia. The D-index was defined as the difference between the observed area under the curve, which was calculated by the trapezoidal method, and the expected neutrophil area (500/mm3 × days with neutropenia) if the patient did not develop neutropenia (Figure). Survival distribution was assessed using the Kaplan-Meier method, and differences were compared using the log-rank test. Multivariate analyses were performed to investigate whether D-index might serve as a prognostic factor, using Cox proportional hazards regression analysis or Fine-Gray proportional hazards regression analysis. The duration of event-free survival (EFS) was defined as the time from the initiation of therapy to either treatment failure (relapse, death, or diagnosis of secondary cancer) or to the last day when the patient was confirmed to be in remission.

Results: D-index during induction phase ranged from 0 to 20,095 with a median of 7,985. The overall survival (OS) for patients with D-index greater than or equal to 3,000 (n=62) was 96.4+/-2.5% at 4 years, whereas that for patients with low D-index less than 3,000 (n=7) was 0% (P <0.001). The EFS for patients with D-index greater than or equal to 3,000 was 88.1+/-4.7% at 4 years, whereas that for patients with low D-index less than 3,000 was 0% (P=0.002). As a result of multivariate analysis, poor OS was associated with low D-index (<3,000) during induction phase (hazard ratio [HR], 24.34; 95% confidence interval [CI], 3.317-178.6; P=0.002). Likewise, poor EFS was associated with low D-index (<3,000) during induction phase (HR, 10.42; 95% CI, 1.655-65.57; P=0.013). Treatment-related mortality (TRM) was not associated with low D-index (<3,000) during induction phases, however cumulative incidence of relapse was associated with low D-index (<3,000) during induction phase (HR, 6.832; 95% CI, 1.640-76.09; P=0.012) as well as age at onset greater than or equal to 10 years of age (HR, 4.960; 95% CI, 1.188-34.25; P=0.014).

Conclusion: The results of our study highlight the importance of a novel risk factor for poor prognosis in children and adolescents with BCP-ALL, not previously considered-low D-index during induction phase. Poor OS and EFS in patients with low D-index might be associated with increased risk for relapse. Lack of a certain level of myelosuppression may correlate with insufficient therapeutic effect. Further research involving a larger multicenter cohort is needed to confirm and explain this finding.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.