Short Telomeres Before Lenalidomide Treatment Predict Leukemic Progression In Patients with Myelodysplastic Syndrome and Deletion 5q

Abstract 30

Lenalidomide has been shown to induce transfusion independence and cytogenetic response in a high proportion of patients with MDS and isolated 5q deletion (del5q). However, some of these patients progress into acute myeloid leukemia, particularly those who do not show an erythroid or cytogenetic response. Yet the mechanisms inducing the leukemic transformation in patients with MDS and del(5q) are mainly unclear. Since dysfunctional telomeres play an important role in the development of genetic instability and shorter telomeres are associated with advanced MDS and AML, we reasoned whether telomere shortening may contribute to leukemic progression in patients with MDS and del5q. This study included 14 patients with transfusion-dependent anemia and low- or intermediate-1-risk MDS enrolled in the studies MDS-003 (n=42) or MDS-004 (n=260) who were treated with lenalidomide according to the study protocols. Written informed consent was provided according to the Declaration of Helsinki. Seven patients progressed into RAEB-1 or AML while on study and 7 patients did not. Time span between initial diagnosis and study entry was similar (median 51.3 and 44.4 months respectively in the patients with and without leukemic progression). Combined fluorescence R-banding and T/C-FISH analysis to determine the telomere length of each individual chromosome was performed as described earlier (K Lange et al, Genes Chromosomes Cancer, 2010). Telomere length at study entry was 6.1 kb (range 4.8 to 7.9 kb) in patients with MDS and del5q who later underwent leukemic transformation. Patients without later disease progression had a median telomere length of 9.3 kb (range 8.4 to 10.2 kb). Thus, patients who progressed after a median of 29 months (range 5 to 51 months) had significantly shorter telomeres than patients who had a cytogenetic response or stable disease (p<0.001). Notably, during treatment with lenalidomide, the group of patients who underwent leukemic transformation had a median increase in telomere length of 2.4 kb (range: 0.5 to 3.4 kb, p<0.02). In contrast, no relevant telomere elongation was observed in patients who did not undergo leukemic progression (9.3 kb at study entry and 8.8 kb after a median of 31 months (range 11 to 52 months) following lenalidomide treatment). Furthermore, telomere length was independent of the presence of additional aberrations. Two patients with additional aberrations and subsequent progression had shorter telomeres than another patient with additional aberrations who did not undergo progression (5.7 kb and 7.9 kb versus 10.2 kb). Two patients with disease progression showed a mosaic of normal and aberrant metaphases in the bone marrow. In both cases, telomeres of the normal metaphases were longer than the telomeres of aberrant metaphases (9.1 kb versus 6.9 kb; 9.7 kb versus 5.5 kb, respectively). Three of the patients who underwent leukemic progression acquired additional chromosome aberrations and developed complex karyotypes. They had a median telomere length of 6.4 kb (range: 4.8 to 7.9 kb), which was not significantly different from the median telomere length of 6.0 kb (range: 4.8 to 7.5 kb) in the group of patients who showed an isolated del5q during the leukemic transformation. Thus, telomere shortening not only preferentially occurred in cells with del(5q), it was also independent of clonal evolution. Although the underlying mechanisms leading to excessive telomere shortening in MDS pathogenesis remain unclear, telomere shortening seems to predispose to leukemic transformation in patients with MDS and del5q. Currently, factors predictive for increased risk of leukemic transformation are lacking and only clinical, morphological and cytogenetic follow-up investigations provide meaningful information regarding treatment success and disease stage. Identification of a prognostic marker to identify patients at increased risk for leukemic progression at an early time point before lenalidomide treatment would greatly improve patient care. Telomere length measurement in these patients may therefore provide a diagnostic tool for predicting treatment response and disease outcome following lenalidomide treatment. Clearly, these data need confirmation in larger patient cohorts, before telomere length measurement can be used for risk assessment prior to starting lenalidomide treatment.