6. Chimerism-directed adoptive immunotherapy in prevention and/or treatment of post-transplant relapse of leukemia in childhood

Abstract

Purpose We present the role of detailed and frequent monitoring of hematopoietic chimerism in prediction of post-transplant clinical outcome and our initial experience with adoptive immunotherapy (AI) in prevention and treatment of relapse in children after allogeneic hematopoietic stem cell transplantation (HSCT) for leukemia. Materials and methods Between 1/1997 and 12/2000 we performed a total of 46 unmanipulated allogeneic HSCT from HLA-identical siblings (MSD;23) or matched unrelated donors (MUD;23) in 43 consecutive children with hematological malignancies (ALL 16, AML 15, CML 5, MDS 6, JMML 3, CMML 1) with a median age of 10,5 years. We have analyzed hematopoietic chimerism in peripheral blood samples using polymerase chain reaction (PCR) of variable number of tandem repeats (VNTR) with a maximum sensitivity of 0,5% in 44 evaluable follow-ups. AI was used in 11 patients/12 follow-ups (ALL 3, AML 5, CML 2, JMML 1/2). The treatment was started on the basis of increasing mixed chimerism (inMC;7), in molecular relapse (1) or in hematological relapse (4). Withdrawn of post-transplant immunosuppression was performed in 9 patients, 5 patients received second-line therapy by donor lymphocyte infusion (DLI) and in 3 follow-ups DLI was applied as a front-line therapy. AI was combined with chemotherapy (CHT) in 2 children. CD3+ doses varied between 1×10 5 and 2,4×10 8 /kg body weight according to type of donor and indication for DLI. Results Complete donor chimerism (CC) characterized by the disappearance of recipient cells until day (D)+28 and sustained emergence of donor cells was documented in 25/44 follow-ups. Mixed chimerism (MC) characterized by the re-emergence or persistence of recipient cells after D+28 was found in 19/44 follow-ups. Transient MC (trMC) was detected in 7, inMC in 12 follow-ups, respectively. Transplant related mortality (TRM) until D+100 was 7/25 in CC and 1/19 in MC group, respectively; 36 patients were therefore evaluable for relapse free survival (RFS). At a median follow-up of 19 months (range 2 to 51 months) RFS for CC group was 17/18, whilst that for MC group was 7/18. Hematological relapse was documented in 1/7 children with trMC, in 1 patient molecular relapse was detected and AI was started before the appearance of trMC. InMC was followed by hematological relapse in 9/12 follow-ups, 3/12 patients with inMC responded to AI and re-achieved CC. Complete response to AI was defined as sustained recurrence of CC and CCR was documented in 4/12 children (1 AML, 2 CML, 1 JMML) at a median follow-up of 30 months (range 15 to 40 months), 1 patient (ALL) achieved CC but died due to severe graft-versus-host disease. Three patients (2 AML, 1 JMML) showed transient decrease or disappearance of MC and developed hematological relapse subsequently, 8, 9 and 20 months after the first intervention with AI. No response was seen in 4/12 follow-ups (2 ALL, 2 AML). Conclusions Detailed and frequent monitoring of HC allows identification of patients with high risk for hematological relapse and therefore indicated for AI. At least in some children with leukemia it is possible to prevent hematological relapse using AI when the intervention comes early. This work was supported by grants IGA-MZ CR NC 6512-3 and IGA-MZ CR NC 5902-3 .