Background: Although outcomes for children and young adults with T-ALL continue to improve, relapse on therapy continues to be a major cause of treatment failure. COG AALL0434 is a phase III study for patients with T-ALL that used a standard 4-drug induction followed by response-based risk stratification at Day 29 with Intermediate and High-risk patients randomized to receive or not receive six 5-day courses of nelarabine during consolidation, delayed intensification and maintenance. All patients are also randomized to receive either Capizzi or high-dose methotrexate during interim maintenance and all patients except those who were Low-risk received cranial irradiation (1200 cGY for CNS1 or 2 and 1800 cGy for CNS3). Risk stratification incorporates MRD detection by flow cytometry at Day 29 using the following cutoffs: Low-risk < 0.1%, Intermediate-risk < 1%, and High-risk > 1%. Recently, the immunophenotype has been associated with a particularly poor outcome in T-ALL, a finding that requires confirmation in a large prospective trial using contemporary therapy.

Methods: 1144 children with T-ALL enrolled on COG AALL0434 between January 22, 2007 and June 30, 2014 were categorized at the time of study enrollment as ETP (n=130; 11.3%), Near-ETP (ETP but with elevated CD5; n=195; 17%) or Not-ETP (n=819; 71.6%) by flow cytometry. MRD was assessed by 8-9 color flow cytometry at Day 8 in peripheral blood (PB) and Day 29 in bone marrow (BM). Flow cytometry to characterize ETP status and MRD was performed in a single central laboratory. Event-free survival (EFS) and Overall survival (OS) were estimated by Kaplan-Maier curve analysis.

Results: Despite significantly higher (P<0.0001) rates of induction failure (>25% blasts by morphology at end induction) for ETP (7.8%) and Near-ETP (6.7%) than Not-ETP (1.1%), all 3 immunophenotypic groups showed excellent 5-year EFS and OS that were not statistically different: ETP (87.0%; 93.0%), Near-ETP (84.4%; 91.6%), and Not-ETP (86.9%; 92.0%). In all 3 groups, most events occurred within 12 months from diagnosis and plateaued after 2 years, although events generally occurred earlier for ETP and Near-ETP than Not-ETP. There were no relapses in ETP patients later than 12 months post diagnosis. Initial white blood cell count (WBC) greater than 200,000/microliter was seen in 9.6% of ETP and 30% of Not-ETP/Near-ETP (P<0.0001) and was associated with inferior EFS and OS for Near-ETP (EFS p=0.0003) and Not-ETP (EFS p=0.012), but not ETP. Day 29 MRD > 0.01% was associated with inferior EFS (76.3% vs. 89.0%; p=0.0001) and OS (86.6% vs. 93.8%; p=0.0008) for the total cohort and was detected in 81.4%, 64.8%, and 30.5% of ETP, Near-ETP and Not-ETP patients, respectively. Interestingly, there was no difference in EFS or OS for Day 29 MRD <0.01% vs. 0.01%-1.0%, suggesting significance largely for MRD > 1.0%. Day 29 MRD >0.01% was also associated with inferior EFS (76.6% vs. 90.8%; p=0.0001) and OS (84.3% vs. 94.0%; p=0.0064) in Not-ETP and inferior EFS (80.2% vs. 94.0%; p=0.0073) in Near-ETP, but no difference was seen for EFS or OS in ETP. Day 8 PB MRD > 0.01% was associated with inferior EFS (80.3% vs. 92.0%; p=0.017) but not OS and lost significance in the subset having Day 29 MRD < 0.01%, suggesting Day 8 does not add prognostic information to Day 29 MRD.

Conclusions: AALL0434 is the largest study of T-ALL ever conducted and shows excellent outcomes for T-ALL. In particular, ETP is found to have an outcome identical to non-ETP patients. WBC count > 200,000/microliter is associated with inferior outcome in non-ETP T-ALL. BM MRD > 1% at Day 29 is able to identify a subset of non-ETP T-ALL patients having inferior outcomes. PB Day 8 MRD does not provide unique prognostic information.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract