Increased Risk of Skeletal Toxicity and Infection in Children 10 Years or Older Treated for Acute Lymphoblastic Leukemia (ALL) with Dexamethasone: Results from the DFCI ALL Consortium.

BACKGROUND: Dexamethasone (dex) may be more efficacious but also more toxic than prednisone (pred) in the treatment of childhood ALL. In order to compare the relative toxicities of these 2 steroids, patients (pts) on DFCI ALL Consortium Protocol 00–01 were randomized to receive dex or pred during post-induction therapy.

METHODS: Pts received 5-day steroid pulses every 3 weeks for 2 years (yrs) (pred 40mg/m2/day or dex 6 mg/m2/day) beginning at week 7 of therapy. High Risk (HR) pts received 3 times these doses during the Intensification Phase (approximately 6 months).

RESULTS: Between 2000–2004, 425 pts (ages 1–18 yrs) were randomized; 423 had evaluable skeletal toxicity data, including 241 Standard Risk (SR) and 182 HR pts. 211 received dex and 212 received pred. Within 26 months of enrollment, 55 pts (13%) experienced at least 1 bony event: 45 pts (11%) with fracture and 17 (4%) with osteonecrosis (ON). Bony events were more common in children ≥10 yrs of age (11% of those <10 yrs vs. 18.5% ≥10 yrs, p=0.007). There was no overall difference in frequency of pts with skeletal toxicity based on steroid type (see table, p=0.89), including no difference in fractures (11% dex vs. 10% pred, p=0.88) or ON (5% dex vs. 3% pred, p=0.23). However, skeletal toxicity was more common in dex-treated older children (≥10 yrs) compared with pred-treated older children (25% vs. 11%, p=0.07) and compared with younger (< 10 yrs) dex-treated pts (25% vs. 9%, p=0.004). There was no difference in frequency of skeletal toxicity by age in pred-treated pts (p=0.31). There was no difference in frequency of skeletal toxicity by steroid type in SR pts (8% dex vs. 14% pred, p=0.22) or in younger HR pts (12% dex vs. 11% pred, p=1.0). In multivariable analysis, children ≥10 yrs treated with dex had 2.7 times the risk of developing skeletal toxicity compared with all other pts (p=0.006). Of 404 pts with evaluable infection data, 37/200 (18.5%) dex-treated pts developed at least 1 infection (bacteremia and/or invasive fungal disease) during intensification or continuation compared to 24/204 (11.8%) pred-treated pts (p=0.07). There was no difference in infection rate by steroid type for those <10 yrs of age (p=0.53). However, for those ≥10 yrs, there were more infections in dex-treated than in pred-treated pts (24% vs. 5%, p=0.01). At 2.8 yrs median follow-up, there was no statistically significant difference in event-free survival (EFS) comparing dex- vs. pred-treated pts (90% vs. 88%, p=0.31), although there was a suggestion of inferior EFS in HR pts randomized to pred.

CONCLUSION: Dexamethasone was associated with increased risk of skeletal toxicity and infection in pts ≥10 yrs of age. Longer follow-up is needed to fully assess EFS differences between the randomized treatment groups.