Increased Infection-Related Mortality for Children with Down Syndrome (DS) in Contemporary Children’s Oncology Group (COG) Acute Lymphoblastic Leukemia (ALL) Clinical Trials.

In prior COG ALL trials, children with DS have had increased treatment-related morbidity without major increases in mortality, when compared to children without DS. Recently, COG opened phase III, randomized trials for the treatment of children with high risk (AALL0232) and standard risk (AALL0331) precursor-B ALL. COG AALL0232 uses an augmented BFM backbone regimen with randomized comparisons of 28 days of prednisone (PDN) 60 mg/m2 vs. 14 days of dexamethasone (DXM) 10 mg/m2 during induction (IND), and high dose methotrexate with leucovorin rescue (HD MTX) vs. escalating IV MTX without rescue plus Peg asparaginase using the Capizzi schedule (CMTX) during a two-month interim maintenance (IM) phase of therapy. Children with DS participated in the induction steroid randomization, but were non-randomly assigned to CMTX arm due to known increased sensitivity to HD MTX. During the first 22 months of accrual to AALL0232, there were 4 deaths among 20 patients with DS, 3 of these 4 deaths occurred in pts receiving DXM in IND yielding an IND mortalaity rate of 15%, which is significantly higher than the 4.3% rate in CCG 1961 (the prior high risk trial). COG AALL0331 utilizes a standard 3-drug IND consisting of DXM 6 mg/m2 for 28 days, vincristine and PEG asparaginase. Post-IND therapy includes a 2 × 2 randomization between intensified vs. standard consolidation, and escalating CMTX combined with augmented delayed intensification (DI) vs. standard IM and DI phases. On AALL0331, there were 6 deaths among 26 children with DS during the first 10 months of accrual. The deaths occurred in IND (3), intensified consolidation (1) and standard DI (2). The IND mortality rate for DS children on AALL0331 is 11.5% compared to 1.3% in CCG 1991 and 0% in CCG 1952 which used identical IND therapy. During DI, patients on CCG 1991 received DXM on days 1–7 and 15–21, whereas those on AALL0331 receive DXM on days 1–21; this difference could contribute to an increased risk of infection in the DS population during the DI phases of the latter trial. The majority of deaths have been due to overwhelming infection with cardiovascular collapse during times of neutropenia. Both COG AALL0232 and AALL0331 suspended enrollment to DS patients pending subsequent therapy changes and expanded supportive care guidelines. This experience emphasizes that children with DS carry a significant risk of infection-related mortality and mandates increased vigilance for early detection and treatment of symptoms suggestive of infection.