Early T-cell precursor acute lymphoblastic leukaemia in children treated in AIEOP centres with AIEOP-BFM protocols: a retrospective analysis

Background: Early T-cell precursor acute lymphoblastic leukaemia was recently recognised as a distinct leukaemia and reported as associated with poor outcomes. We aimed to assess the outcome of early T-cell precursor acute lymphoblastic leukaemia in patients from the Italian Association of Pediatric Hematology Oncology (AIEOP) centres treated with AIEOP-Berlin-Frankfurt-Münster (AIEOP-BFM) protocols.

Methods: In this retrospective analysis, we included all children aged from 1 to less than 18 years with early T-cell precursor acute lymphoblastic leukaemia immunophenotype diagnosed between Jan 1, 2008, and Oct 31, 2014, from AIEOP centres. Early T-cell precursors were defined as being CD1a and CD8 negative, CD5 weak positive or negative, and positive for at least one of the following antigens: CD34, CD117, HLADR, CD13, CD33, CD11b, or CD65. Treatment was based on AIEOP-BFM acute lymphoblastic leukaemia 2000 (NCT00613457) or AIEOP-BFM acute lymphoblastic leukaemia 2009 protocols (European Clinical Trials Database 2007-004270-43). The main differences in treatment and stratification of T-cell acute lymphoblastic leukaemia between the two protocols were that in the 2009 protocol only, pegylated L-asparaginase was substituted for Escherichia coli L-asparaginase, patients with prednisone poor response received an additional dose of cyclophosphamide at day 10 of phase IA, and high minimal residual disease at day 15 assessed by flow cytometry was used as a high-risk criterion. Outcomes were assessed in terms of event-free survival, disease-free survival, and overall survival.

Findings: Early T-cell precursor acute lymphoblastic leukaemia was diagnosed in 49 patients. Compared with overall T-cell acute lymphoblastic leukaemia, it was associated with absence of molecular markers for PCR detection of minimal residual disease in 25 (56%) of 45 patients; prednisone poor response in 27 (55%) of 49 patients; high minimal residual disease at day 15 after starting therapy in 25 (64%) of 39 patients (bone marrow blasts ≥ 10%, by flow cytometry); no complete remission after phase IA in 7 (15%) of 46 patients (bone marrow blasts ≥ 5%, morphologically); and high PCR minimal residual disease (≥ 5 × 10(-4)) at day 33 after starting therapy in 17 (85%) of 20 patients with markers available. Overall, 38 (78%) of 49 patients are in continuous complete remission, including 13 of 18 after haemopoietic stem cell transplantation, with three deaths in induction, five deaths after haemopoietic stem cell transplantation, and three relapses. Severe adverse events in the 2009 study were reported in 10 (30%) of 33 patients with early T-cell precursor acute lymphoblastic leukaemia versus 24 (15%) of 164 patients without early T-cell precursor acute lymphoblastic leukaemia and life-threatening events in induction phase IA occurred in 4 (12%) of 33 patients with early T-cell precursor acute lymphoblastic leukaemia versus 7 (4%) of 164 patients without early T-cell precursor acute lymphoblastic leukaemia. No difference was seen in the subsequent consolidation phase IB of protocol I.

Interpretation: Early T-cell precursor acute lymphoblastic leukaemia is characterised by poor early response to conventional induction treatment. Consolidation phase IB, based on cyclophosphamide, 6-mercaptopurine, and ara-C at conventional (non-high) doses is effective in reducing minimal residual disease. Although the number of patients and observational time are limited, patients with early T-cell precursor acute lymphoblastic leukaemia treated with current BFM stratification and treatment strategy have a favourable outcome compared with earlier reports. The role of innovative therapies and haemopoietic stem cell therapy in early T-cell precursor acute lymphoblastic leukaemia needs to be assessed.

Funding: None.