Abstract

Abstract 3561

Background:

Adolescents (15–19y) with acute lymphoblastic leukemia (ALL) markedly benefit from pediatric rather than adult chemotherapy regimens, as highlighted by many retrospective studies (Boissel et al., JCO 2001). In young adults with ALL, there are now strong evidences that pediatric or pediatric-inspired protocols may also improve long-term outcome (Huguet et al., JCO 2009). However, whether care environment (adult vs pediatric unit) may impact this outcome remains an open issue. To address this question, we explored the outcome of adolescents and young adults (AYA, 15–29y) treated in adult hematology departments by the pediatric FRALLE 2000-BT protocol.

Methods:

From February 2001 to June 2011, 89 AYAs with Ph-negative ALL including 60 adolescents (15–19y) and 29 YAs (20–29y) were treated according to the pediatric FRALLE 2000-BT protocol in 13 French and Belgian adult hematology units (median follow-up, 5 years). After a prednisone prephase and a four-drug induction (prednisone, daunorubicin, vincristine and L-asparaginase), patients achieving CR received four 8-week phases of treatment: consolidation, 1st delayed intensification (DI), interphase, 2nd DI, and maintenance. According to FRALLE protocol recommendations, each phase should begin after hematological recovery of previous phase (ANC>1 G/L, Platelets>100 G/L). Twenty-five patients (18 adolescents and 7 YAs) underwent allogeneic transplantation after consolidation phase or 1stDI.

Results:

Apart age (median age, 18 vs 23y; p<.0001), adolescent and YAs cohorts had similar characteristics, including WBC, B/T lineage distribution, and high-risk karyotype (t(4;11)/MLL-AF4, hypodiploidy)), except for initial CNS involvement (1/60 vs 4/29 respectively, p=.02). Initial peripheral response to prednisone (PDN) at D7 and bone marrow response to chemotherapy at D21 were better in the adolescent subgroup, without reaching significance (slow PDN-responders: 30% vs 38%, p=.42; slow chemo-responders: 13% vs 21%, p=.37). Overall CR rate was 99% and did not differ between both cohorts (98% in adolescents, 100% in YAs). 5y-OS and 5y-EFS were estimated respectively at 66% and 61%. Despite similar CR rates, 5-year EFS was significantly shorter in adolescents than in YAs (47% vs 79%, p=.02). A non significant shorter 5y-OS was also observed in adolescents (58% vs 81%, p=.11). To explain this difference, we looked at dose-intensity disparities between both subgroups. During induction phase, no differences in drug cumulative doses were observed. To further explore potent difficulties to respect protocol schedules in younger patients, we compared intervals from induction D1 to start of different trial phases between both subgroups. In comparison to YAs, intervals from induction D1 to consolidation, 1st DI, 2nd DI and maintenance phases were progressively delayed in adolescents: 4 days for consolidation (p=.24), 6 days for 1st DI (p=.19), 9 days for 2nd DI (p=.005), and finally 25 days for maintenance phase (p=.0002). In univariate analysis, among age (adolescents vs YAs), B/T lineage, WBC, cytogenetics, and time to 1st DI, only age and time to 1st DI significantly affected EFS (p=.05 and p=.001 respectively). In bivariate analysis, age and time to 1st DI were identified as independent prognostic factors for EFS (p=.03 and p=.004, respectively). The same observation was done with interval time to 2ndDI.

Conclusion:

The outcome of AYAs treated in adult units with the pediatric FRALLE 2000 protocol is encouraging, particularly in YAs. The poorer outcome observed in adolescents may be related to increasing intervals from initial induction to the different phases of treatment. Prospective trials are required to register individual reasons that lead physicians and patients to progressively delay therapy. This study suggests that trial itself is not sufficient to improve the outcome in adolescents with cancer and that AYA programs are needed to explore the difficulties encountered by patients and care teams to adhere to therapy.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract