Outcome of Children with T-Cell Acute Lymphoblastic Leukemia (T-ALL) and Standard Risk (SR) Features: Results of CCG-1952, CCG-1991 and POG 9404.

Children with T-ALL have generally had a poorer prognosis than patients with precursor-B ALL. Patients with T-ALL are more likely than those with B-precursor ALL to be > 9 years and present with WBC > 50,000/ul, bulky lymphadenopathy and mediastinal mass. Since 1996, the former CCG has stratified protocol eligibility for patients with ALL based on NCI defined Standard Risk (SR = age between 1-9.99 years; WBC < 50,000/ul) or High Risk (HR) groups, regardless of immunophenotype. In contrast, the former POG has treated all T-cell patients on protocols separate from those for precursor-B ALL. This report compares the outcomes among children with T-cell ALL treated on recently completed or ongoing CCG and POG phase III trials for ALL. CCG-1952 enrolled 2176 eligible children with SR ALL between 1996 and 2000; 106 (5%) had T-cell immunophenotype. Treatment was a standard BFM regimen with prednisone and 2 phases of delayed intensification (DI) with a 2x2 randomization of IT methotrexate (MTX) v ITT and MP v TG. From June 2000 to March 2004, CCG-1991 has enrolled 1794 SR ALL patients: 80 (4%) had T-ALL. Treatment included a similar BFM backbone with substitution of dexamethasone and a 2x2 randomization between escalating IV v oral MTX and 1 v 2 DIs. POG-9404 (opened 1996; closed 2001) was developed exclusively for T-cell disease and enrolled 363 patients with T-ALL; 84 (23%) fit the NCI SR group criteria. On the latter protocol, patients received treatment similar to that developed by the Dana-Farber Leukemia Consortium for high risk B-precursor ALL, with randomization to ± 4 cycles of high dose MTX (5 Gm/m2) and leucovorin. All patients on 9404 received 1800 cGy prophylactic cranial radiation while CCG patients did not. Outcome for T-ALL on CCG-1952 is substantially worse than for B-precursor ALL, with 5 year event-free survival (EFS) of 73% compared to 82% (p = 0.007). Interim analysis of CCG-1991 also shows a significantly worse outcome for T-ALL compared to B-precursor ALL: 3y estimated EFS 78% v 90%, p = 0.0002. In contrast, estimated 5y EFS for patients with SR T-ALL on POG-9404 is 88% (90% on the superior high dose MTX regimen). Comparison of the SR v HR T-ALL patients treated on POG 9404 shows a significant advantage for the SR group (5y EFS of 90% v 75%, p < 0.004). This is in contrast to comparison of T-ALL patients on CCG-1952 (SR) v the concurrent CCG-1961 HR study where T-ALL patients have similar outcome (5y EFS 73% v 72%, p = 0.77). These data suggest that patients with T-ALL and SR features have better EFS when treated with more intensified chemotherapy regimens. Because early EFS for T-cell patients treated on CCG-1991 is worse than on POG 9404, the former study was closed to further accrual of patients with T-ALL. The COG ALL Committee is developing a study for exclusive enrollment of patients with T-ALL using intensive therapy based on the current COG HR ALL regimen, regardless of SR or HR features.