Disease relapse after allogeneic hematopoietic stem cell transplant (HCT) remains a major obstacle to treatment success and cure for patients with myelodysplastic syndromes (MDS) and acute myelogenous leukemia (AML). Early identification of patients at a high risk for disease relapse has significant practical implications allowing early therapeutic interventions. We evaluated a novel approach for an early identification of a post-transplant relapse in patients with MDS and AML.
CD34+ specific donor chimerism analysis was obtained in 6 HCT recipients with CD34+ AML and MDS. Starting at day +30 post HCT CD34+ cells from peripheral blood were positively selected by immunomagnetic isolation and then underwent fluorescence-activated cell sorting. Purified CD34+ cells were subsequently evaluated for percentage of donor DNA contribution by short tandem repeat (STR) analysis. CD34+ specific donor chimerism analysis was repeated monthly until hematologic relapse or death occurred. Simultaneously, conventional donor chimerism analysis was measured in the subpopulations of peripheral blood cells.
CD34+ cells were isolated from peripheral blood with a very high purity >95% and in sufficient quantity for reliable STR analysis in all study patients. Out of 6 evaluated patients, 3 developed hematologic relapse confirmed by bone marrow evaluation. In each relapsed patient CD34+ specific chimerism dropped to <15% prior to the confirmation of hematologic relapse. However, conventional STR analysis in granulocytes, T-cells and B-cells remained >80% (Figure 1). Moreover, obvious and steady decline in CD34+ specific donor chimerism significantly preceded the time of hematologic relapse. In contrast, no decline in CD34+ chimerism was identified in 3 patients without hematologic relapse. (Figure 1).
Though this is a preliminary study, the results suggest that CD34+ specific donor chimerism analysis is a useful technique for timely identification of hematologic relapse in MDS and AML patients after HCT and may provide significantly earlier relapse detection than conventional chimerism analysis.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.