LAUSR

It has been well known for more than a decade, that iron overload adversely affects the prognosis of patients with lower-risk myelodysplastic syndromes (MDS) (Malcovati et al, 2005). Iron overload may be transfusion-related as well as being a consequence of increased intestinal iron uptake driven by hepcidin downregulation caused by dyserythropoiesis (Park et al, 2019) or genetic factors. In a recent study on neutrophils from 42 MDS patient samples with or without iron overload, the 14 patients with a ferritin level higher than 1000 μg/l, or more than 70% transferrin saturation, had a higher detectable 02. and glutathione, detected by flow cytometry (Ivars et al, 2017). In turn, this excessive production of oxygen radicals contributes to organ damage, with hepatic and cardiac morbidity secondary to iron overload (de Montalembert et al, 2017), and may also impair haematopoiesis by increasing DNA instability. As a consequence, several recommendations have been made for iron chelation in MDS patients with a high transfusion burden or a high ferritin. These are largely based on the efficacy of chelation (Gattermann, 2008) and the survival benefit of chelation in beta-thalassaemia patients. Several groups have reported unexpected haematological responses resulting from iron chelation alone, with several haematological responses and even 3–20% of patients reaching transfusion independence with deferasirox (Gattermann et al, 2012; Angelucci et al, 2014). These observations suggest a beneficial effect of iron chelation on haematopoiesis. In addition, at the last American Society of Hematology meeting, the first results of a randomized prospective study of deferasirox compared to placebo in iron overloaded MDS patients (TELESTO) indicated that overall survival may be longer in patients receiving iron chelation (Angelucci et al, 2018). However, this study was underpowered because of its failure to reach the expected recruitment target, and most study events were related to cardiac morbidity. Therefore, the question of haematological (and outcome) improvement resulting from the reduction of oxidative stress by iron chelators remains an interesting hypothesis to explore. In the work published in this issue of the British Journal of Haematology, Jim enez-Solas et al (2019) explored this question in an elegant work on primary samples from MDS patients. The aim was to compare the ex vivo haematopoietic capacity of 14 primary bone marrow (BM) haematopoietic progenitors from patients with lower risk MDS, before and after 4–10 months of deferasirox treatment, with each patient used as their own control. In parallel, BM from 20 healthy donors was analysed. To assess haematopoiesis, colony-forming unit (CFU) assays were performed alongside detailed flow cytometric characterization with 2’,7’–dichlorofluorescin diacetate (DCFDA) flow cytometry analyses, DNA oxidation and double-strand breaks, and Annexin-V apotosis assessment. The initial finding was that the intracellular reactive oxygen species (ROS) levels were higher in all MDS BM subpopulations, mostly in the CD34, myeloid CD34, granulocyte and monocyte populations. This was associated with DNA oxidation and early apoptosis mostly in the myeloid compartments. The growth of both granulocyte-macrophage and erythroid CFU (CFU-GM and CFU-E, respectively) was also impaired in MDS samples compared to controls. Following deferasirox treatment, global ROS levels were not reduced. However, there was a significant decrease in DNA oxidation and DNA double strand breaks in most BM subpopulations compared to pre-treatment levels. Furthermore, the authors observed that the CFU assays generated an increased number of myeloid colonies after deferasirox treatment, reaching the level of the control sample assays. In addition, the number of clusters, which was higher in the pre-treatment samples, decreased almost to the level of controls. By using a cluster/colony ratio, the authors indicated a strong post-treatment decrease of this marker, that is likely to represent a surrogate of haematopoiesis impairment. *Correspondence: Professor Emmanuel Gyan, Service d’h ematologie et th erapie cellulaire, Hôpital Bretonneau, CNRS ERL 7001, Universit e de Tours, 2 boulevard Tonnell e, 37044 TOURS CEDEX 09, France E-mail: [email protected] editorial comment