Substitution with doxorubicin for daunorubicin during induction for high risk pediatric acute lymphoblastic leukemia results in increased toxicity

To the Editor: Widespread drug shortages have affected the availability of chemotherapy agents commonly used to treat pediatric ALL [1]. Induction chemotherapy for high risk acute lymphoblastic leukemia (HR ALL) consists of vincristine, PEGaspargase, daunorubicin, and a glucocorticoid with associated low rates of morbidity and mortality, and a high rate of remission efficacy [2]. The shortage of daunorubicin, which began in the spring of 2011, compromised the delivery of standard induction chemotherapy to pediatric HR-ALL patients. Daunorubicin and doxorubicin are historically considered dose equivalent in the treatment of ALL based upon in vitro testing showing similar sensitivity in untreated pediatric ALL samples [3]. Two clinical studies in acute myelogenous leukemia (AML) comparing the two anthracyclines suggested that daunorubicin has a lower toxicity profile than doxorubicin [4,5]. During the time of daunorubicin shortage, our institution used doxorubicin at a 1:1 substitution. To examine the differences in toxicity and efficacy between doxorubicin and daunorubicin during induction therapy for HR ALL, we conducted a retrospective case control study. We performed a retrospective chart review of all HR ALL patients who received induction therapy at Seattle Children’s Hospital from January 01, 2009 until July 31, 2011. During the study period 45 patients met eligibility criteria, nine treated with doxorubicin and 37 with daunorubicin. The two groups had similar clinical characteristics; including age, initial white blood cell count, gender, and pre-B or T cell disease (Table I). The patients who received doxorubicin were more likely to have missed or delayed doses of doxorubicin/daunorubicin (22% vs. 5%) or vincristine (22% vs. 0%). CNS directed therapy was not affected by the doxorubicin substitution. Although all of the patients in the doxorubicin treated group had an M1 marrow (<5% residual lymphoblasts by morphology) by Day 15, 44.4% were Day 29 minimal residual disease positive ( 0.01% disease) versus 27% in the daunorubicin treated group (P 1⁄4 0.188). Toxicity was increased in the doxorubicin treated group of patients, with increased rates of mucositis (33.3% vs. 10.8%) and typhilitis (22.2% vs. 0%; Table I). One patient with typhlitis died with disseminated fungal (Zygomycetes) infection at Day 19 of induction therapy. Overall, microbiologically documented fungal infections were more common in the doxorubicin treated group (33.3% vs. 5.4%), although the frequency of bacteremia was similar. Our data suggest that a 1:1 substitution of doxorubicin for daunorubicin will not result in equivalent toxicity and that doxorubicin increases the rates of mucositis and typhlitis when used in a similar induction regimen. However, the appropriate dose conversion is not clear. The small size of our doxorubicin-treated population prevents a firm conclusion about the impact on remission induction, including end-induction minimal residual leukemia levels. Doxorubicin substitution during induction was associated with chemotherapy omissions, likely related to the increased toxicity associated with doxorubicin in this induction regimen. Several cooperative groups and institutions routinely administer doxorubicin during induction for HR ALL but on a different schedule [6–8]. In the event of future daunorubicin shortage, alternative induction options could be considered, including induction regimens with two consecutive daily doses of doxorubicin rather than weekly dosing [6].