Donor Lymphocyte Infusions to Prevent Relapse in Acute Leukemias after Allogenic Stem Cell Transplantation.

Purpose: 115 patients with acute lymphoblastic (n = 56) and myeloid leukemia (n = 59) underwent allogenic stem cell transplantation (SCT) from HLA-matched related or unrelated donors between 1998 and 2003. 52/115 patients (45 %) received donor lymphocyte infusions (DLI) for prophylaxis or treatment of relapse after SCT. Subject of this study was to assess efficacy and toxicity of DLI in these patients.

Methods: 83/115 patients were transplanted after standard high-dose radio- and chemotherapy (standard SCT) and 32/115 after reduced intensity conditioning (RIC) with fludarabine, busulfan and ATG. 47/115 patients were in CR-1, 22/115 were in CR-2 and 46/115 patients had advanced stages at the time of transplant. DLI were given to 52 patients, 17/52 patients after RIC and 35/52 patients after standard SCT. Indications for DLI were 1. prophylaxis of relapse in 40/52 patients either after RIC (n = 14) or in high-risk leukemias after standard SCT (n = 26; Ph+ ALL in CR-1, AML and ALL > CR-1) and 2. treatment of hematologic relapse in 12/52 patients. DLI were given in escalating doses every 4 weeks. Chimerism analysis of leukocyte subpopulations were performed before and after DLI by multiplex PCR. Patients with refractory leukemia or those who died before day 60 after SCT were not included in this analysis.

Results: 52/115 patients received 103 DLI with a median of 2.7 x 10⁷ CD3+ cells/kg. 29/52 patients (56 %) are alive with a median follow-up of 22 months (range 3 – 69 months); 24 (46 %) are in CR and 5 (10 %) are alive in relapse. The 3 year-probability of overall survival (OS) for all patients treated with DLI is 53 % versus 66 % for patients not treated with DLI (not significant). Patients with ALL treated with DLI have a 3 year probability of overall survival (OS) of 68 % versus 46 % for ALL-patients, who did not receive DLI (p=0.03). For patients with AML, 3 year probability of OS is 41% after DLI versus 78% for patients not treated with DLI (p=0.03). The differences in patients with AML can probably be attributed to the fact that 55 % of those receiving DLI were transplanted in relapse compared to only 30 % of those not receiving DLI. In ALL patients equal number of patients were transplanted in relapse (DLI: 43%; no DLI: 38%). 35/49 patients with available chimerism analysis had mixed chimerism before 1. DLI. After DLI 16/30 patients (53 %) with available analysis had converted to full donor chimerism. All patients who received DLI for leukemic relapse died due to disease progression. 34/52 patients (65 %) developed acute GVHD after administration of DLI, with 7/52 patients (13 %) having grade III/IV.

Conclusions: In retrospective analysis DLI seem to be an effective treatment strategy with acceptable toxicity to prevent relapse in high-risk leukemias. To monitor the effect of prophylactic DLI Chimerism analysis is a valuable tool. Patients transplanted with advanced disease have a very high relapse risk independent of the administration of DLI. DLI are not suited to treat leukemic relapse.