BACKGROUND: A conditioning regimen for HLA-identical allogeneic hematopoietic cell transplantation in relapsed, refractory, or otherwise high risk myeloid malignancies has been developed, called FLAMSA (

Schmid et al.;
J Clin Oncol
). This protocol combines a four-day salvage chemotherapy consisting of daily fludarabine 30 mg/m2, amsacrine 100 mg/m2, and cytarabine 2000 mg/m2, followed by three days of pause, with a reduced-intensity conditioning in the subsequent week, comprising total body irradiation 4 Gy (or busulfan 8 mg/kg), cyclophosphamide 80 or 120 mg/kg and antithymocyte globulin 30 or 60 mg/kg (for related / unrelated donors, respectively). Tapering of immunosuppression until day 90 and prophylactic donor lymphocyte infusions for patients without GvHD are integral parts of FLAMSA. This protocol has since enjoyed widespread use due to its tolerability even in patients of older age or reduced performance, as well as its salvage effect and curative potential even in patients without remission before transplantation. However, the lack of a fully HLA-matched donor might render this last chance unsuitable. Our purpose was to compare outcome after FLAMSA with HLA-identical versus partially HLA-mismatched donors.

PATIENTS AND METHODS: We have employed the FLAMSA protocol in 90 patients between March 2004 and June 2007, of whom 69 (the ident group) had a fully HLA-matched related (8/8 loci) or unrelated (10/10 loci) donor and 21 (the nonident group) a partially HLA-mismatched donor (1 locus in sixteen patients, 2 loci in four and haplo-identical in one). Half were females and half males, with a median age of 54 years (range: 19 to 71 years). 39 had been diagnosed with de novo acute myeloid leukemia (AML), 39 with secondary AML, 11 with myelodysplasia and one with acute lymphoblastic leukemia. 12 were in first and 4 in subsequent complete remission, whereas 74 were untreated, refractory or in relapse. Both the ident and the nonident groups were comparable regarding gender, age, diagnoses, cytogenetic risk group, remission status at transplant, as well as cytomegalovirus and sex match with their respective donor.

RESULTS: With 9.2 months (range: 0.3 to 38.2 months) median follow-up of all patients, 11/21 (52%) nonident patients are alive and 10/21 (48%) in complete remission, as compared to 39/69 (57%) ident patients (not significant). Probabilities of overall and disease-free survival at 2.5 years after allogeneic hematopoietic cell transplantation (Figure) are 43% and 35% for nonident and 45% and 41% for ident patients, respectively (p = 0.54 and p = 0.56; not significant). Treatment related mortality among nonident patients was 6/21 (29%) versus 12/69 (17%) in ident patients, whereas relapse related death occurred in 18/69 (26%) in the ident group with compared to 4/21 (19%) in the nonident group.

CONCLUSION: In our single-center, retrospective comparison with limited median follow-up, both fully HLA-identical and partially HLA-mismatched donors were suitable for the FLAMSA protocol. Confirmation of this finding in a prospective study is warranted.

Author notes

Disclosure: No relevant conflicts of interest to declare.