Introduction: The outcome of salvage chemotherapy in refractory or relapsed adult acute lymphoblastic leukemia (ALL) is poor. In this situation allogeneic hematopoietic stem cell transplantation (HSCT) remains the only curative therapy, but it is not clear whether reinduction chemotherapy prior to HSCT is required to achieve long-term remissions.

Methods: Here, we present a retrospective analysis of 65 patients (median age 29 years, 17–54) who received allogeneic HSCT for refractory or relapsed ALL from 1995 until 2006 (1st relapse n=50, 2nd or higher relapse n=7, primary refractory n=8, B-lineage ALL n=50, T-lineage ALL n=15). 22/65 patients were transplanted without reinduction chemotherapy prior to HSCT (median blasts in the bone marrow 30%, 7–90), whereas 43/65 patients received reinduction. 23 out of these 43 patients achieved CR2, but 21/43 patients were refractory (median blasts 70%, 10–95). Patients received standard high-dose (n=61) or reduced intensity conditioning (RIC) (n=4) and HLA-matched (n=58) or -mismatched (n=7) HSCT from related (n=26) or unrelated (n=39) donors.

Results: Median follow-up of the surviving patients was 53 months (2–130). Overall survival (OS) of the whole cohort at 6 months, 1, 3 and 5 years was 61%, 40%, 27% and 24%, respectively. Deaths were due to relapse (n=29), GVHD (n=8), infections (n=7) or other causes (n=4). Chronic GVHD (cGVHD) occurred in 28/53 patients (53%) who survived more than 100 days after HSCT. Patients who developed cGVHD had significantly better outcome than patients without cGVHD (p<0.0001) with OS at 6 months, 1 year, 3 years and 5 years being 88% vs. 58%, 72% vs. 25%, 55% vs. 10% and 47% vs. 10%, respectively. There was a trend towards improved OS for the cohort which did not receive reinduction chemotherapy compared to the cohort which received reinduction (p=0.06) with OS at 6 months, 1, 3 and 5 years being 64% vs. 60%, 55% vs. 31%, 45% vs. 17% and 36% vs. 14%, respectively. Importantly, in the reinduction cohort there were no differences in OS for the subgroups of patients transplanted in CR2 or transplanted with active disease. Deaths in the reinduction cohort were mainly due to relapse (64% vs. 50%), suggesting that reinduction therapy might have contributed to the selection of therapy-resistant leukemia clones. In further univariate analyses risk group at diagnosis, number of relapses, time-to-relapse, donor type and ALL lineage did not have significant impact on OS.

Conclusions: Patients who undergo allogeneic HSCT for relapsed ALL can achieve long-term survival, which is projected at about 24% at 5 years. The development of cGVHD leads to increased OS, whereas risk group at diagnosis, number of relapses, time-to-relapse, donor type and ALL lineage do not have significant impact. Importantly, application of reinduction chemotherapy prior to HSCT does not improve outcome, even if a second complete remission is achieved. These results demonstrate the effectiveness of HSCT without prior disease control and again demonstrate the importance of graft-versus-leukemia-reactions in ALL.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution