A Uyttebroeck

Long-term results of three randomized trials (58831, 58832, 58881) in childhood acute lymphoblastic leukemia: a CLCG-EORTC report. Children Leukemia Cooperative Group.

We present here the long-term results of three randomized clinical trials conducted on children with newly diagnosed acute lymphoblastic leukemia (ALL) between 1983 and 1998 by the Children Leukemia Cooperative Group (CLCG) from EORTC. In study 58831/32, the overall event-free survival (EFS) rates (± s.e.) at 6 and 10 years were 66% ± 1.8% and 65% ± 1.8%, respectively, and the risk of isolated central nervous system (CNS) relapse was 6% ± 1% and 7% ± 1%, respectively. In patients with a standard risk of relapse the omission of cyclophosphamide had no adverse effect on disease-free survival rates at 10 years (trial 58831). In medium- and high-risk patients the omission of radiotherapy did not increase the risk of CNS or systemic relapse (trial 58832). In study 58881 (1989–1998) the overall EFS rate at 8 years was 68.4% ± 1.2% and the risk of isolated CNS relapse was 4.2% ± 0.5%. In this trial which adressed three randomized questions, the following results were obtained: the combination of cytarabine at high doses with methotrexate at high doses during interval therapy did not improve prognosis. The addition of 6-mercaptopurine iv during maintenance increased the risk of late relapse. E. coli asparaginase was more toxic and has a higher efficacy than erwinia asparaginase. leukocyte counts >100 × 109/l, specific genetic abnormalities, a poor initial response to steroids or a high level of minimal residual disease at early time points were consistently associated with an adverse prognosis in the 58881 trial.

Clinical presentation, evolution, and prognosis of precursor B-cell lymphoblastic lymphoma in trials LMT96, EORTC 58881, and EORTC 58951

In children, lymphoblastic lymphomas represent 30% of Non‐Hodgkin lymphomas (NHL), and approximately 15% are precursor B‐cell lymphomas (PBLL). Our study evaluated their main clinical characteristics, evolution, and prognosis in three trials. From 1989 to 2008, 53 children with PBLL (median age 7·75 years) were included in three protocols: Malignant Lymphoma Therapy (LMT) 96, European Organization for Research and Treatment of Cancer (EORTC) 58881, and EORTC 58951 using Berlin‐Frankfürt‐Münster‐derived acute lymphoblastic leukaemia (ALL) therapy. There were 10 stage I disease, 9 stage II, 9 stage III and 25 stage IV. Clinical presentation was heterogeneous with a majority of bone lesions and cutaneous or subcutaneous manifestations. At diagnosis 23 patients had bone marrow involvement, and only three had central nervous system involvement. The median follow‐up was 74 months. At last follow‐up, 45 patients were in continuous complete remission, whereas eight had progressed or had relapsed (7 Stages IV and 1 Stage III) and died. Two patients had a secondary neoplasia, and are still alive. Disease stage was a major prognostic factor, with better overall survival (OS) and event‐free survival (EFS) (P < 0·05) rates observed in patients with Stage I to III as compared to those with Stage IV. Treatment with protocols derived from ALL therapy are efficient with an 82% EFS and an 85% OS at 5 years.

Value of High-Dose Cytarabine During Interval Therapy of a Berlin-Frankfurt-Munster–Based Protocol in Increased-Risk Children With Acute Lymphoblastic Leukemia and Lymphoblastic Lymphoma: Results of the European Organization for Research and Treatment of Cancer 58881 Randomized Phase III Trial

PURPOSE The European Organization for Research and Treatment of Cancer 58881 study was designed to test in a prospective multicentric randomized trial the value of high-dose (HD) intravenous (IV) cytarabine (Ara-C) added to HD IV methotrexate (MTX) to reduce the incidence of CNS and systemic relapses in children with increased-risk acute lymphoblastic leukemia (ALL) or stage III and IV lymphoblastic lymphoma treated with a Berlin-Frankfurt-Munster (BFM)-based regimen. PATIENTS AND METHODS After completion of induction-consolidation phase, children with increased-risk (risk factor > 0.8 or T-lineage) ALL or stage III and IV lymphoblastic lymphoma were randomized to receive four courses of HD MTX (5 g/m(2) over 24 hours every 2 weeks) and four intrathecal administrations of MTX (Arm A) or the same treatment schedule with additional HD IV Ara-C (1 g/m(2) in bolus injection 12 and 24 hours after the start of each MTX infusion) (Arm B). RESULTS Between January 1990 and January 1996, 653 patients with ALL (593 patients) or lymphoblastic lymphoma (60 patients) were randomized: 323 were assigned to Arm A (without Ara-C) and 330 to Arm B (with Ara-C). A total of 190 events (177 relapses and 13 deaths without relapse) were reported, and the median follow up was 6.5 years (range, 2 to 10 years). The incidence rates of CNS relapse were similar in both arms whether isolated (5.6% and 3.3%, respectively) or combined (5.3% and 4.6%, respectively). The estimated 6-year disease-free survival (DFS) rate was similar (log-rank P =.67) in the two treatment groups: 70.4% (SE = 2.6%) in Arm A and 71.0% (SE = 2.5%) in Arm B. The 6-year DFS rate was similar for ALL and LL patients: 70.2% (SE = 1.9%) versus 76.3% (SE = 5.6%). CONCLUSION Prevention of CNS relapse was satisfactorily achieved with HD IV MTX and intrathecal injections of MTX in children with increased-risk ALL or stage III and IV lymphoblastic lymphoma treated with our BFM-based treatment protocol in which cranial irradiation was omitted. Disappointingly, with the dose schedule used in this protocol, HD Ara-C added to HD MTX, although well tolerated, failed to further decrease the incidence of CNS relapse or to improve the overall DFS.

IKZF1 deletion is an independent prognostic marker in childhood B-cell precursor acute lymphoblastic leukemia, and distinguishes patients benefiting from pulses during maintenance therapy: results of the EORTC Children's Leukemia Group study 58951

The added value of IKZF1 gene deletion (IKZF1del) as a stratifying criterion in B-cell precursor acute lymphoblastic leukemia (BCP-ALL) is still debated. We performed a comprehensive analysis of the impact of IKZF1del in a large cohort of children (n=1223) with BCR-ABL1-negative BCP-ALL treated in the EORTC-CLG trial 58951. Patients with IKZF1del had a lower 8-year event-free survival (EFS, 67.7% versus 86.5%; hazard ratio (HR)=2.41; 95% confidence interval (CI)=1.75–3.32; P<0.001). Importantly, despite association with high-risk features such as high minimal residual disease, IKZF1del remained significantly predictive in multivariate analyses. Analysis by genetic subtype showed that IKZF1del increased risk only in the high hyperdiploid ALLs (HR=2.57; 95% CI=1.19–5.55; P=0.013) and in ‘B-other‘ ALLs, that is, lacking classifying genetic lesions (HR=2.22; 95% CI=1.45–3.39; P<0.001), the latter having then a dramatically low 8-year EFS (56.4; 95% CI=44.6-66.7). Among IKZF1del-positive patients randomized for vincristine-steroid pulses during maintenance, those receiving pulses had a significantly higher 8-year EFS (93.3; 95% CI=61.3–99.0 versus 42.1; 95% CI=20.4–62.5). Thus, IKZF1del retains independent prognostic significance in the context of current risk-adapted protocols, and is associated with a dismal outcome in ‘B-other‘ ALL. Addition of vincristine-steroid pulses during maintenance may specifically benefit to IKZF1del patients in preventing relapses.

CL097 - Les lymphomes lymphoblastiques pré-B dans les protocoles LMT96, EORTC 58881, et 58951

Les lymphomes lymphoblastiques pre-B (PBLL) sont une entite rare qui representent 15 % des lymphomes lymphoblastiques ou 4,5 % des lymphomes malins non hodgkiniens de l’enfant. Objectif de l’etude Decrire les caracteristiques cliniques et l’evolution therapeutique des PBLL inclus dans les protocoles francais LMT96 et francobelges EORTC 58881, 58951. Patients De 1990 a 2008, 53 enfants (âge median 7,75 ans) ont ete inclus et traites selon un schema directement derive de celui des leucemies aigues lymphoblastiques (LAL). Resultats La classification de Murphy retrouvait 10 stades I, 9 stades II, 9 stades III et 25 stades IV (23 atteintes medullaires et 3 atteintes meningees). La plupart des localisations principales etaient extra-nodales (majoritairement osseuses et cutanees). A J70, 51 patients avaient une reponse complete. Lors du dernier suivi (mediane a 77,5 mois), 45 enfants etaient toujours en remission (2 d’entre eux ayant eu un 2nd cancer) et 8 (7 stades IV et 1 stade III) etaient decedes. Le stade de la maladie a un impact pronostic significatif, l’OS et l’EFS etant superieures pour les stades I a III (EFS a 93 % ; OS a 96 %) par rapport aux stades IV (EFS a 71 % ; 0S a 73 %). Conclusion Un traitement prolonge de type LAL est efficace (EFS et OS a 5 ans de 82 et 85 %).