The outcome of children with T-cell acute lymphoblastic leukemia: A single institution experience

Document Type





Background: T-cell acute lymphoblastic leukemia (T-ALL) constitutes 10% to 15 % of childhood ALL cases in the Western literature. However, higher proportions, up to 40%, have been reported in certain developing countries. While the outcome of childhood T-ALL has improved dramatically over the last decades, by using intensive multi-agent chemotherapeutic regimens, about 30% to 40% of these patients still experience relapses.
Patients and methods: Medical records of the 411 pediatric patients (0–13 years old) diagnosed with ALL and treated at our institution between January 1999 and December 2004 were retrospectively reviewed.
Result: Fifty-three (12.9%) of the patients had T-ALL. 52 of these were treated according to a multi-drug chemotherapy protocol, based on a modification of the St. Jude Total XIII-B protocol, including high-dose methotrexate (HD-MTX), at 2gm/m2, and non-cross-resistant drug pair administered weekly during the continuation phase. The remaining one patient received a standard ALL therapy with a single delayed intensification and anti-metabolite based maintenance therapy. Early response to induction chemotherapy was assessed using a bone marrow (BM) evaluation at Day-14 of induction. The mean age was 7.1 years and 36 patients (68%) were males. Twenty (39%), out of 51 patients with available results, had CNS involvement. Of these eight (15.5%) were categorized as CNS-3, while 12 (23.5%) were CNS-2. 50% of the patients presented with mediastinal mass. Nine (18%) of the patients with available initial WBC (n=49) had hyperleukocytosis with a WBC >100 × 109/L. 8 out of the 51 patients with a Day-14 evaluation had >5% blasts in the BM. All patients subsequently achieved remission at the end of induction. The 5-year event-free survival (EFS) and overall survival (OS) rates were 63.9% and 76.4%, respectively. 13 (24.5%) patients relapsed at a median time of one year. Six relapses occurred in the BM, 5 isolated-CNS relapses, one CNS+BM and one in the skin+BM. Only 2 of these relapsed patients are long-term survivors. The disease-related and treatment toxicity-related deaths occurred in 11(20.7%) and 2(3.7%) patients, respectively. The initial WBC count, mediastinal mass, CNS status or the Day-14 BM results were not found to impact negatively on the outcome, by multivariate analysis.
Conclusion: Our current treatment strategy has resulted in potential cure for about two-thirds of the patients with childhood T-ALL. However, a significant number of patients do fail treatment and novel strategies need to be devised for them. Unfortunately, as specific clinical features that could determine outcome have not as yet been identified, risk stratification, similar to that utilized for precursor-B cell ALL, is not feasible. Further study into risk determining variables at the molecular level, such as activating NOTCH1 mutations, may be promising in predicting outcome and determining treatment intensity in children with T-ALL.


This work was published before the author joined Aga Khan University