Back to research

Research paper

Neurobehavioral Side Effects of Corticosteroids During Active Treatment for Acute Lymphoblastic Leukemia in Children are Age-Dependent: Report from Dana-Farber Cancer Institute ALL Consortium Protocol 00-01

Prospective parent-rated study within DFCI ALL Consortium Protocol 00-01. 62 children aged 2–17 on cyclical 5-day prednisone (40 mg/m²/day) or dexamethasone (6 mg/m²/day). Behavior ratings at baseline (Day 0), active steroid (Day 7), post-steroid (Days 14, 21). Preschoolers (<6 y) had clinically significant increases in emotional control, mood, behavior regulation, and executive function problems during the steroid week (p<.001), returning to baseline during the two off-steroid weeks. School-age children (≥6 y) showed no comparable effect. Mechanism: high steroid-receptor density in hippocampal and frontal regions.

Indexed context

Mrakotsky CM, et al.

corticosteroidneurobehavioralpediatricpreschoolprednisonedexamethasoneall-leukemiaage-dependent

Markdown path

content/research/papers/2012-mrakotsky-dfci-all-corticosteroid-neurobehavioral.md

Findings

Prospective parent-rated study within DFCI ALL Consortium Protocol 00-01. 62 children aged 2–17 on cyclical 5-day prednisone (40 mg/m²/day) or dexamethasone (6 mg/m²/day). Behavior ratings at baseline (Day 0), active steroid (Day 7), post-steroid (Days 14, 21). Preschoolers (<6 y) had clinically significant increases in emotional control, mood, behavior regulation, and executive function problems during the steroid week (p<.001), returning to baseline during the two off-steroid weeks. School-age children (≥6 y) showed no comparable effect. Mechanism: high steroid-receptor density in hippocampal and frontal regions.

Why it may matter for Levi

Levi (5.5 y) sits in the dramatically more behaviorally susceptible preschool age bracket. Argues that for cyclical low-dose oral exposure, behavior returns to baseline within ~2 weeks — a partial counter-data point to the 'delayed rebound at 3.5 weeks' hypothesis. Important caveat: a single 3-day high-dose IV pulse (~20 mg/kg/day) is a very different point in the dose-time space than 40 mg/m² oral prednisone; absence of a 2–3-week-later effect in Mrakotsky does not exclude one in Levi's regimen, but it sets a useful prior.

Paper text

Mrakotsky et al. (2011) — Neurobehavioral side effects of corticosteroids in children with ALL

Source

Why this paper is in the corpus

This is one of the few prospective pediatric studies that tracked behavioral effects of corticosteroid courses with weekly parent ratings, making it useful for two questions: (1) how strong are the behavioral effects, and (2) how long do they persist after the steroid course ends.

Key findings

  • 62 children (age 2–17 at diagnosis) on cyclical 5-day prednisone (40 mg/m²/day) or dexamethasone (6 mg/m²/day) courses during the continuation phase of ALL treatment.
  • Parent behavior ratings at baseline (Day 0), active steroid (Day 7), and post-steroid (Days 14 and 21).
  • Preschool children (<6 years): behavioral side effects increased significantly (p<.001) during the steroid week on emotional control, mood, behavior regulation, and executive functions, with reported problems in a clinically significant range.
  • Behavioral ratings in this group returned to baseline during the two off-steroid weeks.
  • School-age children (≥6 years): these pronounced effects were not detected.
  • Mechanism noted: brain regions with high steroid-receptor density (hippocampal and frontal regions) regulate mood, behavior regulation, and memory and are particularly susceptible to corticosteroid effects.

Limitations relevant to Levi

  • Repeated cyclical 5-day oral courses, not a single 3-day high-dose IV pulse — so the 2-week post-cycle "return to baseline" finding is from a very different exposure pattern.
  • Outcome measure window was 14 and 21 days post-cycle; longer post-exposure windows (3–6 weeks) were not assessed.
  • The cohort was free of pre-existing severe neurodevelopmental disorders; Levi's baseline is very different.

Levi-relevant takeaways

  • Provides the cleanest prospective evidence that young children (Levi's age bracket) are dramatically more behaviorally susceptible to corticosteroids than older children — emotional control, mood, behavior regulation, and aggression are the exact axes affected.
  • Argues that for low-dose repeated oral courses, behavior does return to baseline within ~2 weeks of the course ending. This is a partial counter-data point to the "delayed rebound at 3.5 weeks" hypothesis: a clean cyclical exposure pattern does not produce a 2–3-week-later re-emergence in this cohort.
  • Caveat: a single high-dose IV pulse (700 mg/day × 3 days, ~20 mg/kg/day) sits at a very different point in the dose-time space than cyclical 40 mg/m² oral prednisone, so absence of a 2–3-week-later effect in Mrakotsky does not exclude one in Levi's regimen.