Analysis of Long Term Outcome in Autologous Stem Cell Transplant (ASCT) for Acute Myelogenous Leukemia (AML) in First Remission (CR1) - a 15 Year Experience.

Despite improvement in the proportion of patients achieving durable remission with dose intensive consolidation strategies, leukemic relapse still occurs in 40–70% of patients who achieve an initial remission. Studies utilizing ASCT for AML in CR1 have reported relapse free survival between 34–80% based on cytogenetic risk stratification. We report a 15 year experience of ASCT following high dose cytarabine (Ara-C) ± idarubicin (IDA) consolidation in 168 patients (intent to transplant, ITT) in CR1 treated between 3/89 and 6/03. The treatment consisted of a): consolidation with high dose Ara-C ± IDA followed by marrow (M) harvest (n=64) or collection of G-CSF primed stem cells (PSCs) (n=73), b) ASCT using FTBI 12Gy, Etoposide 60mg/kg, Cytoxan 75mg/kg (n=97) or Busulfan targeted to 1^st dose AUC 700–900, FTBI 1200Gy and Etoposide 30mg/kg (n=40), and c) IL-2 9x10⁶ IU/m²/24 hrs days 1–4 and IL-2 1.6 x 10⁶ IU/m²/24 hrs days 9–18 beginning at hematologic recovery post ASCT (from 1995-present). Patient characteristics at ITT were median age 40.8 yr. (16.3–60.9); cytogenetics (SWOG criteria) favorable = 42(25%) (including 11 M3 pre 1993), intermediate=70(42%), unfavorable=16(10%), indeterminate and unknown=40(23%); WBC <50,000=125(74%), ≥50,000=41(25%), unknown=2(1%); 147 (88%) received one induction cycle and 12% received two cycles; Ara-C induction dose was 100–200mg/m² in 49% and 51% received high dose Ara-C induction. Median time from CR1 to consolidation was 27.6 days. Of the 168 patients consolidated, 137 had ASCT. The 100-day non-relapse mortality (NRM) for consolidation+ASCT was 2%. With a median follow up of 9.5 yrs, the 5 yr OS, DFS relapse rate (RR) for 168 ITT patients is 57%, 51%, and 41% respectively. The 5 yr OS, DFS and RR for each cytogenetic category are: favorable 64%, 63%, 29%; intermediate 58%, 50%, 41%; unfavorable 45%, 31%, 67%; and unknown cytogenetics 49%, 48%, 44% respectively. The 5 yr OS, DFS and RR for the 137 transplanted patients were 62%, 59%, and 36%. The 5 year OS and RR for favorable cytogenetics are 69%, 24%; intermediate 62%, 36%; unfavorable 55%, 58%; and unknown 55%, 42% respectively. Prognostic variables analyzed for each outcome in the ITT group were: age, WBC, cytogenetics, Ara-C induction dose, one vs. two inductions and IDA in consolidation. By multivariate analyses the number of inductions was predictive for OS with hazard ratio=2.2 (95%CI:1.2–4.3) p=0.02. No variables were found to be predictive of DFS or relapse. The same variables with the addition of conditioning regimen, IL-2 and M vs. PSCs were analyzed for the 137 transplant patients. By multivariate analysis IL-2 was predictive for OS and DFS and cytogenetics was predictive of relapse. These results demonstrate that the strategy of high dose Ara-C consolidation followed by ASCT can be performed with low NRM. This approach appears to improve the outcome of patients with all cytogenetic risk groups when compared with multiple cycles of high dose Ara-C. The role of post-transplant IL-2 appears to be beneficial and merits further investigation.