Tandem Auto-Allo in Acute Myeloid Leukemia (AML) Patients in First Complete Remission (CR).

High and intermediate risk AML can benefit by allogeneic stem cell transplantation in first CR. The use of reduced intensity conditioning regimens (ALLO-RIC) decreases the toxicity even if the relapse rate is more pronounced. To contrast the high relapse rate we hypothyzed that if a better quality of remission could be achieved, the relapse incidence could be lowered.

From 2001 to 2008, 31 AML patients in first CR received a tandem auto-allo program. The median number of white blood cell was 3 × 10e9/l (range 0.9-235), 13% of patients have extramedullary localisations. In 13% AML was secondary to previous CT treatment. Cytogenetic was abnormal in 36% of pts. After one or two induction chemotherapies (CT), all but two patients received a consolidation course with high-dose cytarabine (HD-ARAC) CT, followed by autologous stem cell harvest. Then, HD melphalan (HD-PAM 140 mg/m²) followed by autologous stem cells reinfusion was administered, followed by ALLO-RIC. RIC consisted of fludarabine plus (2 Gy) TBI (3 pts) or fludarabine, oral or intravenous busulfan (8 mg/kg) in two days, and anti thymocyte globulin (2.5 or 5 mg/kg). Graft versus host disease (GVHD) prophylaxis was cyclosporine (CyA) plus mycofenolate mofetil (3 pts) or CyA alone (28 pts). Donors were all but one HLA identical sibling. The median number of allogeneic CD34+ and CD3+ cells was 6.1 × 10e6/Kg (range 1.9-11) and 315 (range 166-609). Prognostic scores (HCT-CI, PAM, EBMT) were retrospectively calculated for each patient. All pts have a performance status ≥ 90%.

The median follow-up was from diagnosis and ALLO-RIC 40 and 34 months, respectively. The median time between last CT and HD-PAM was 51 days (range 30-77) and HD-PAM and ALLO-RIC was 69 days (55-176). Treatment related mortality after HD-PAM was null. Prognostic scores were: HCT-CI score 0-2= 53% (16 pts), ≥3= 47% (14 pts), 1 pt not evaluable; PAM score 9-16= 30% (10 pts), 17-23= 67% (19 pts), 24-30= 3% (1 pt); EBMT score 1= 9% (3 pts), 2= 78% (24 pts), 3= 13% (4 pts). At last follow-up, 42% of pts (n= 13) died: 5 due to disease relapse and 8 because of toxicity. Grade II-IV acute GVHD and chronic GVHD incidence were respectively 26% and 65% (extensive 84%). GVHD was the cause of death in seven pts. Six pts (19%) reactived CMV, without disease, and 1 pt not survived to an interstitial pneumonitis. The 5-year overall survival (OS), relapse free survival (RFS), and 1-year TRM were 60%, 60%, and 15%, respectively. In multivariate analysis, prognostic scores did not influence TRM and OS.

This report showed that i) tandem auto-allo is feasible in AML pts; ii) acute GVHD incidence is not increased iii) prognostic score did not impact on TRM and survival; iv) the TRM is quite low with GVHD as main cause of death. A retrospective comparison with a cohort of pts not receiving HD-PAM is on going.

No relevant conflicts of interest to declare.