Abstract

Imatinib mesylate (STI571) is a selective Bcr-Abl tyrosine kinase inhibitor used for the treatment of Philapdelphia chromosome positive leukemias and other malignancies. Typical non-hematological side effects of imatinib are mostly moderate whereas hematological side effects include all three lineages and are dose-dependent (e.g. 35% grade 3/4 neutropenia at 800 mg daily). Recently, a significant dose-dependent inhibitory effect of imatinib on normal hematopoietic progenitor cells has been reported. However, since alterations of the hematopoietic system might also result (at least in part) from a possible drug-induced damage of the hematopoietic environment, we studied the effects of imatinib on normal human marrow microenvironmental cells. Marrow stromal progenitor cell growth was considerably inhibited by imatinib in a dose-dependent fashion: CFU-F were reduced to 75 ± 35.5%, 52 ± 29.2%, 32 ± 25.7%, 33.7 ± 23.4%, and 20.6 ± 20.6% of control at 0.04, 0.16, 0.62, 2.5, and 10.0 μM imatinib, respectively (IC50: 0.44 μM). Marrow stromal cells (MSC, culture-derived from normal bone marrow mononuclear cells) were also affected dose-dependently: After two weeks in culture, MSC were reduced to 20.0 ± 2.2% (0.62 μM), 14.6 ± 1.2% (1.25 μM), 10.1 ± 0.3% (2.5 μM), 5.8 ± 0.5% (5 μM), and 2.8 ± 0.3% (10 μM) of controls. After 4 weeks, corresponding data were 3.1 ± 0.5%, 2.2 ± 0.4%, 1.8 ± 0.3%, 1.1 ± 0.2%, and 0.5 ± 0.1%. Furthermore, following a 7-day culture period in imatinib-containing medium (0.62 – 10 μM) and subsequent washing and resuspension, the growth kinetics of imatinib-treated MSC were delayed for up to 2 weeks after imatinib withdrawal. Moreover, the total cellular expansion of imatinib-pretreated MSC after 4 weeks of culture was lower when compared with controls. To assess whether imatinib would impair functional MSC capacities, long-term culture initiating cell assays were set up using imatinib-treated MSC (0 μM, 1.25 μM, 5 μM) as feeder cells and normal CD34pos-PBPC for inoculation. After 5 weeks, numbers of clonogenic progenitors assayed per well were not different (control: 20.0 ± 7.1, imatinib 1.25 μM: 16.2 ± 6.5, imatinib 5 μM: 21.3 ± 6). Experiments addressing a possible role of PDGF- and SCF-receptor signaling revealed that the growth inhibitory effects of imatinib (10 μM) clearly exceeded those of receptor blocking antibodies, thereby indicating that imatinib acts - at least in part - independently of PDGFR- and c-kit-signaling. Cell cycle analysis showed that the fraction of imatinib-treated MSC (0.62 - 10 μM) in S- and G2/M-phase was slightly lower albeit not significantly different when compared with controls. Furthermore, no differences with regard to the fraction of apoptotic cells were observed. Nevertheless, MSC cultures in maximum growth phase showed a markedly higher susceptibility to imatinib compared to those in the steady-state. Taken together, imatinib severely affected marrow stromal cells in-vitro. This effect was at least partly independent of PDGFR- and c-kit signaling and appeared to be related to the proliferative status of the cells. This information might be relevant for the administration of imatinib in situations with increased MSC turnover, such as regeneration after intensive chemo/radiotherapy.

Author notes

Corresponding author