Abstract A3: Low-frequency subclones in acute myeloid leukemia samples at diagnosis indicate oligoclonality and may play a role in disease progression

Abstract

Abstract Activating type I mutations provide hematopoietic cells with proliferative and survival advantages. Together with type II abnormalities, which cause a differentiation arrest and an increase in self-renewal properties, they cooperate to cause acute myeloid leukemia (AML). In a recent study on the stability of type I and II mutations in paired initial and relapsed pediatric AML samples, we showed instability of mutations between diagnosis and relapse in 26 out of 69 patients (38%). Gain of FLT3 internal tandem repeat (ITD), WT1 and RAS mutations were significantly associated with a shorter period of time between initial diagnosis and relapse. Vice versa, losses were associated with a longer time to relapse. Moreover, disease outcome was significantly worse in patients that were or became FLT3/ITD or WT1 mutated. The mutation status determined at diagnosis is therefore not sufficient to predict outcome and time to relapse. The occurrence of gains of mutations in a relatively short time (median 7.7, range 5.4-17.3 months) suggests that leukemic clones harboring these mutations may already be present within the initial AML sample and resist therapy. We hypothesize that oligoclonality exists within the leukemic stem cell fraction at initial diagnosis AML. Using flowcytometry based cell sorting techniques, we isolate small numbers (<25 cells) of primitive leukemic cells from the mononuclear cell fraction of AML patients, based on leukemia associated and hematopoietic/leukemic stem cell marker expression. Sensitive screening methods allow us to compare the mutation profiles of the most instable type I/ II genes (FLT3, RAS, WT1) in such low amounts of cells. Informed consents were obtained from all patients. We could analyze sub fractions of the blast compartment in a patient sample of which we had enough material at both diagnosis and relapse. Standard diagnostics on the bulk of mononuclear cells of this patient showed at initial diagnosis a single 21 base pair (bp) FLT3/ITD while at relapse only a 45 bp FLT3/ITD was found. We isolated primitive fractions based on marker expression (CD133, CD117, CD34 and CD38) from the initial leukemic blasts (CD45dim). The progenitor-like CD34+CD38dim fraction, which expressed AML stem cell marker CLL-1, besides the 21 bp FLT3/ITD, also showed the 45 bp FLT3/ITD, that was predominant in the blasts at relapse. In addition the CD133− fraction was enriched for leukemic cells harboring the 45 bp ITD when compared to the CD133+ fraction. A similar enrichment was found in the CD117+ fraction versus CD117−. Remarkably, the CD34+CD38−CLL-1+ fraction, in which leukemic stem cells are presumed to reside, contained the 21 bp but not the 45 bp FLT3/ITD. Sorted T cells from this patient only contained the wild type FLT3. In conclusion, the relatively common (nearly 40%) shifts in type I/II mutations of pediatric AML patients may in part be explained by the expansion of minor leukemic sub-clones during post-treatment progression of the disease. Moreover, the presence of the minor outgrowing clone in a progenitor instead of a stem cell compartment, indicates that leukemia-initiating cells may be present in a progenitor compartment. Although we found indications for clonal selection, in other cases the observed mutational shifts may occur due to the emergence of new leukemic sub-clones, during or after therapy and are related to genetic instability. To elucidate this, future research will focus on the detection and characterization of malignant sub-clones at different stages of the disease. Financially supported by Dutch Cancer Society (VU 2005-3666) Citation Information: Clin Cancer Res 2010;16(7 Suppl):A3