Investigators in the Dutch-Belgian Hemato-Oncology Cooperative Group and the Swiss Group for Clinical Cancer Research Collaborative Group recently reported in the journal Blood that autologous peripheral blood stem cell transplantation significantly reduced relapse rate compared with intensive consolidation chemotherapy in acute myeloid leukemia (AML) patients in first complete remission.¹ Five-year relapse-free survival was nonsignificantly improved with autologous stem cell transplant. Nonrelapse mortality was significantly greater in autologous transplant patients. There was no difference between treatments with regard to overall survival, reflecting the fact that more relapsing chemotherapy patients had salvage treatment with late stem cell transplant and achieved second complete remissions.

Study Design

In this multicenter trial, 517 AML patients aged 16 to 60 years in first complete remission after two cycles of intensive induction chemotherapy who were not eligible for allogeneic stem cell transplant were randomly assigned to receive autologous transplant (n = 258) or intensive consolidation chemotherapy with etoposide (100 mg/m² on days 1–5) and mitoxantrone (10 mg/m² on days 1–5) (n=259). Remission induction therapy consisted of one cycle of cytarabine plus idarubicin and one cycle of cytarabine plus amsacrine, with patients being randomly assigned between intermediate- and high-dose cytarabine. Patients at some sites were also randomized between granulocyte colony-stimulating factor (G-CSF [Neupogen]) and no G-CSF priming during induction. Conditioning for autologous transplant consisted of treatment with busulfan (Busulfex, Myleran) and cyclophosphamide.

Patients in the chemotherapy and autologous transplant groups were well matched with regard to gender (49% vs 51% male), age (median, 47 vs 49 years), cytogenetic risk (favorable, 9% vs 7%; intermediate, 70% vs 72%; unfavorable, 6% vs 6%; very unfavorable, 7% vs 4%), induction cytarabine dose level (intermediate, 58% vs 54%), induction with G-CSF priming (25% vs 26%), and cycle of induction during which complete remission was reached (cycle 1, 81% vs 79%). Median follow-up of patients alive was 106 months. A total of 93% of chemotherapy patients received the planned consolidation chemotherapy, and 91% of those randomized to autologous transplant actually underwent transplantation.

Patient Outcomes

Five-year relapse rates were 58% in the transplant group vs 70% in the chemotherapy group (P = .02). A nonsignificant improvement in 5-year relapse-free survival was observed with augologous stem cell transplantation (38% vs 29%, P = .065). Nonrelapse mortality, a measure of treatment procedure–related death, was significantly greater in the transplant group (4% vs 1%, relapse-free survival = .02). Median overall survival was similar in the two groups (44% vs 41%, P = .86).

Second-line treatment was received by 74% of the 156 relapsing patients in the transplant arm, and consisted of autologous transplantation in 1%, allogeneic transplantation in 17%, and chemotherapy in 55%. By comparison, second-line treatment was received by 80% of the 187 relapsing patients in the chemotherapy arm, consisting of autologous transplant in 14%, allogeneic transplant in 25%, and chemotherapy in 40%. Thus, a greater proportion of patients in the chemotherapy arm had the potential for salvage with autologous or allogeneic transplant (39% vs 18%). Second complete remissions were achieved in 27% of the relapsed transplant patients vs 47% of the relapsed chemotherapy patients, resulting in long-term survival in 7% vs 15%.

Predictive Factors

In the total study population, a number of factors were predictive of outcome. Increasing age was associated with reduced relapse-free survival (P = .01) and overall survival (P < .001). Extramedullary disease at diagnosis was associated with reduced relapse-free survival (P = .016). Cytogenetics was strongly associated with both relapse-free and overall survival (P < .001 for both), as was achievement of first complete remission in the first vs second cycle of induction chemotherapy (P < .001 for both). There was no significant treatment group effect for any of these factors. However, it was noted that when patients with the monosomal karyotype (very unfavorable cytogenetics) with very poor relapse-free survival in both groups were excluded from analysis, the overall increase in relapse-free survival with autologous transplant treatment became statistically significant (P = .014).

Neutrophil counts of greater than 0.5 × 10⁹/L were achieved by 32% of autologous transplant patients by day 14 and by 88% by day 28 after transplantation, compared with 1% and 42% of patients in the chemotherapy group, respectively, after the end of consolidation therapy. Platelet counts recovered somewhat more rapidly in transplant patients than in chemotherapy patients over the first month, but significantly more slowly thereafter in patients with counts that had not recovered by this time. A similar pattern was observed for platelet transfusion independence.

There were no differences between groups with regard to grade 3 or 4 bleeding events or infections. However, a number of grade 2 to 4 adverse events were more common in the autologous transplant group, including fever of unknown origin and gastrointestinal, hepatic, and neurologic adverse events. ■

Reference

1. Vellenga E, van Putten W, Ossenkoppele GJ, et al: Autologous peripheral blood stem cell transplantation for acute myeloid leukemia. Blood 118:6037-6042, 2011.