Anaplastic large cell lymphoma (ALCL) in children: equal efficacy but greater toxicity of chemotherapy includin methotrexate (MTX) 1g/m2 in 24 hour infusion with intra-thecal injection (IT) than chemotherapyu with MTX 3g/m2 in 3 hours infusin without IT: results of the ALCL99-R1 randomization trial

Objectives: to compare the impact of a different dose and schedule of administration of MTX on the outcome of pediatric patients with ALCL Background: BFM-K2 protocol with 6 alternating courses A (MTX-ifosfamide-etoposide-cytarabine-dexamethasone) and B (MTX-cyclophosphamide-doxorubicin-dexamethasone) has been shown to be highly efficient in childhood ALCL. In this protocol MTX was given at a dose of 1g/m2 in 24 hour infusion with intra-thecal injection (IT). Methods: ALCL99 study is a multicentric international prospective study for childhood ALCL, involving most European groups and Japan, comparing BFM-K2 chemotherapy with MTX 1 g/m2 in 24 hour infusion with IT (M1-arm) and BFM-K2 with MTX 3g/m2 in 3 hours infusion without IT (M3-arm). Children classified as high risk patients (HR) due to mediastinum, lung, liver, spleen involvement or skin lesions were eligible for a second randomisation, testing the impact of adding Vinblastine during induction treatment and weekly as a maintenance treatment for a total of one year (factorial design). Endpoints: Event-free Survival (EFS), Overall survival (OS), CNS relapse, toxicity. Results: 375 patients recruited between November 99 and December 2005, from 175 centres, 12 countries: 187 in the M1-arm and 188 in the M3-arm. Median age: 11.3y (range 4 mo–20y). National or international pathology review already available for 86% patients, 91% Alk+. 232 (62%) classified as HR patients (114 M1, 118 M3), 266 (75%) as St Jude stage 3 or 4 (132 M1, 134 M3). Overall 92% of patients randomised in R1 received protocol treatment, in terms of number of induction courses, dose of MTX and IT injection. Median follow-up: 31 months. Outcome: EFS curves were superimposable with 2-y EFS 73% (95%CI, 66–80%) and 73% (65–79%) in M1 and M3 arms respectively, hazard ratio=1.0 (0.60 to 1.7), p=.98, and 2-y OS 88% (82–93%) and 94% (89–97%) respectively. CNS relapses: 2 versus 0 (p=.25). Toxicity assessed in 2313 induction courses: haematological toxicity grade 4 was observed after 82% M1 and 68% M3 courses (p<.0001), infection after 51% M1 and 31% M3 courses (p<.0001), grade 3–4 stomatitis after 22% M1 and 6% M3 (p<.0001), 17 and 11 patients experienced at least one suspected unexpected adverse event (p=.25), 2 and 3 toxic deaths occurred (p=1.0) respectively. Results of R2 randomisation are presently immature. Conclusion: Outcome appears very similar in both groups whereas severe toxicity was more frequent in the MTX 1g/m2 24h arm as compared to the MTX 3g/m2 3h arm. Thus in childhood ALCL, there is no benefit of using MTX 1 g/m2 in 24 hour infusion with IT as compared to MTX 3g/m2 in 3 hours infusion without IT.