Bendamustine, but not fludarabine, exhibits a low stem cell toxicity in vitro

Abstract

Purpose: We investigated the in vitro toxicity of bendamustine and fludarabine to hematopoietic progenitors and stem cells from healthy donors.

Methods: Clonogenic agar colony assays, non-clonogenic long-term liquid cultures (LTC) and apoptosis assays were used to assess the cytotoxicity of both the agents.

Results: Total colony-forming units (CFU) were more sensitive to fludarabine than to bendamustine in agar colony assays (IC(50) 0.7 microM/L and 8.5 microM/L, respectively). Using the Bliss independence model and combining the two agents yielded additive inhibition of progenitors. Non-clonogenic assays, including LTC and an apoptosis assay detecting activated caspases showed that stem cells are characterized by low sensitivity to bendamustine. In contrast, fludarabine strongly inhibited the viability and growth of stem cells in LTC.

Conclusions: Our data show that bendamustine is characterized by lower in vitro toxicity to hematopoietic progenitors and stem cells than fludarabine and might thus be preferable in regimens prior to stem cells apheresis.

Fig. 1

Inhibition of total CFU by bendamustine and fludarabine in the agar colony assay. CFU were more sensitive to fludarabine than to bendamustine (a, n = 10), whereas no significant difference was observed between stem cell originating from BM (n = 4) and those from PB (n = 6, b, c). Total CFU included CFU-GM (clusters and colonies), CFU-E, BFU-E and CFU-GEMM. Shown is the number of CFU at a given concentration of either bendamustine or fludarabine compared to the negative control (median values + SD)

Fig. 2

Concentration-depend inhibition of CFU-GM (clusters), CFU-GM (colonies) and BFU-E by bendamustine (a) and fludarabine (b) including stem cells from BM and PB (n = 10). CFU-E and CFU-GEMM are not depicted for statistical reasons since absolute counts for these CFU-subtypes were low (≤4 cells/dish for each CFU-E and CFU-GEMM) in the negative control in all samples. Shown are median values (+SD)

Fig. 3

Inhibition of CFU in the agar colony assay by the combination of bendamustine and fludarabine using a concentration close to the IC₅₀ value (10 μM/L for bendamustine and 1 μM/L for fludarabine; n = 6). Total CFU included CFU-GM (clusters and colonies), CFU-E, BFU-E and CFU-GEMM. CFU-E and CFU-GEMM are not analyzed separately for statistical reasons since absolute counts were low in the negative control (≤4 cells/dish for each CFU-E and CFU-GEMM). Shown is the measured fractional inhibition (FI) of the combination (white bars) compared to the estimated FI according to the Bliss model (black bars) using the FI of the single agents (median values + SD). In the case, that measured and estimated FI do not differ significantly, both agents act additively, whereas greater and lower measured FI represent synergistic and antagonistic effects, respectively

Fig. 4

Viability (a) and growth (b) of PB stem cells in non-clonogenic LTC using the IC₅₀ concentration obtained in the clonogenic CFU assay (Fig. 1). One representative of three experiments is shown

Fig. 5

Sensitivity of PB CD34⁺ cells to bendamustine-induced apoptosis and necrosis after 24 (a, n = 5) and 48 h (b, n = 3) incubation in liquid cultures. Shown is the percentage of early apoptotic, late apoptotic, necrotic and alive cells at a given concentration of bendamustine (median values + SD)