Myelodysplastic syndromes (MDS) are usually defined as clonal proliferations of pluripotent stem cells, which retain their capacity to differentiate but do so in an inefficient manner, so that mature cells in the peripheral blood are variably altered and reduced. Various approaches can be used to prove the existence of a clonal population of cells. Analysis of X inactivation-dependent methylation patterns in double heterozygous females has been used in MDS, and this approach showed clonal hematopoietic progenitors capable of granulocytic-monocytic differentiation in most cases. However, in a recent study (

) clonality, defined as a clonal population accounting for 35% or more of total granulocytes, was confirmed in only one third of patients with refractory anemia. We studied X-chromosome inactivation patterns (XCIP) in 50 consecutive female patients with MDS. These included 18 RA, 6 RARS, 12 RCMD, 4 RCMD-RS, 2 MDS with del(5q), 4 RAEB-1 and 4 RAEB-2. The XCIP was established by analysis of the IDS gene expression and of DNA methylation at the HUMARA and PGK loci. In order to assess the clonality status of single hematopoietic lineages, the relative expression of HUMARA alleles in reticulocytes, granulocytes and platelets was evaluated by RT-PCR. Finally, telomere length, as a major determinant of the replicative lifespan of hematopoietic cells, was analyzed on peripheral blood granulocytes by flow-FISH (Dako Telomere PNA kit/FITC). XCIP and telomere length analyses were performed at the diagnosis (21 cases) or during the follow-up (29 cases, median time from diagnosis 10 months, range 1–230). XCIP were informative in 43 patients: 20 displayed clonal or ambiguous XCIP (46.5%), while 23 showed polyclonal XCIP (53.5%). Polyclonal XCIP were found in 20 of 36 informative MDS patients without excess blasts (55.5%), and 3 of 7 (42.9%) with excess blasts. Among MDS patients with polyclonal hematopoiesis, clonal cytogenetic abnormalities were found in 9 out of 17 informative cases, all of them showing a chimeric karyotype with normal and abnormal metaphases. Telomere length was significantly shorter in patients with monoclonal patterns than in patients with polyclonal hematopoiesis (P=.01). A preliminary study of XCIP by expression of HUMARA alleles showed that 6 patients with polyclonal XCIP on PMN had clonal XCIP on both reticulocytes and platelets. In brief, these data provide evidence that more than 50% of patients with low risk MDS had some polyclonal hematopoiesis. However, detection of karyotypic abnormalities in these patients confirmed the clonal nature of the disease, the myelodysplastic clones coexisting with residual normal hematopoiesis. Preliminary data indicate that the myelodysplastic clone may originate in a common erythroid-megakaryocytic progenitor. Finally, the telomere data suggest that monoclonal hematopoiesis is associated with telomere shortening.

Author notes

Corresponding author