Comparison of Intensive Chemotherapy (CHT), Allogeneic (ALLO) or Autologous (AUTO) Stem Cell Transplantation (SCT) as Post-Remission Treatment for Adult Patients with High-Risk Acute Lymphoblastic Leukemia (HR-ALL). Final Results of the PETHEMA ALL-93 Trial.

The optimal post remission therapy for HR-ALL patients (pts) is not well established. This multicenter randomized trial was addressed to compare three options of post remission therapy in adults with HR-ALL: intensive consolidation plus maintenance chemotherapy (CHT), ALLO or AUTO SCT. Inclusion criteria: one or more of the following: age 30–50 yr., WBC>25x10⁹/L, t(9;22) or BCR/ABL, 11q23 or MLL and t(1;19). Treatment schedule: 5-drug (VCR, DNR, PDN, ASP, CPM) induction therapy over 5-weeks, followed by 3 1-wk cycles of early intensification CHT including HD-MTX, HD-ARAC and HD-ASP over 3 mo. Patients with an HLA-identical sibling were assigned to ALLO SCT and the remaining were randomized to AUTO SCT or to delayed intensification CHT (the same three cycles given in the post-remission period) followed by standard maintenance CHT (MP+MTX) up to 2-yr. in continuous complete remission (CR).

Study period: 1993–2004, 254 pts included, 222 evaluable, 35 hospitals, 131 males, mean (SD) age 29(10) yr, WBC count 60(98) x10⁹/L, early pre-B: 43 (19%), common+pre-B: 113 (51%), T: 66 (30%). Cytogenetics: 161 (73%) valid cases after central review, normal: 67 t(9;22): 37, 11q23: 6 cases, t(1;19): 2, other rearrangements: 49 cases. Response to therapy: CR 183 (82%). Slow response to therapy (>10% blasts in BM aspirate at day 14 of induction therapy) in 40% of cases. 84 pts were assigned to ALLO SCT, 50 randomized to AUTO SCT and 48 to CHT, 1 pt. removed from protocol after the end of induction. With a median follow-up of 70 mo.(range 24–133), medians for DFS and OS for the whole series were 17 and 23 mo., and 5-yr DFS and OS probabilities were 35±5% and 34±6% (37±6% and 35±6% after removing Ph+ ALL pts from analysis). Groups of ALLO, AUTO and CHT were comparable for the main clinicobiologic characteristics and the rate of response to therapy. Intention-to-treat analysis showed no differences in DFS and in OS for donor vs. no donor comparison as well as for comparison of AUTO vs. CHT. This lack of differences persisted after removing Ph+ ALL pts from the analysis. The same results were observed when the analysis was performed on the basis of effectively treated pts (ALLO SCT performed in 70%, AUTO SCT in 62% and CHT in 73% of the pts). By multivariate analyses, age>30 yr. was associated with death in induction, slow response to induction therapy and Ph+ ALL were associated with a lower CR, and these three variables had a negative influence on DFS and on OS probabilities.

No differences in outcome were found in pts assigned to ALLO SCT or randomized to AUTO SCT or CHT as post-remission therapy in HR-ALL. DFS from ALLO or AUTO SCT or from intensification therapy was also identical in effectively treated patients. Adults with Ph+ ALL or slow response to induction therapy have been identified as a very-high-risk ALL subgroup for whom specific or experimental therapies are justified.