S164: disease modifying activity of imetelstat in patients with heavily transfused non-del(5q) lower-risk myelodysplastic syndromes relapsed/refractory to erythropoiesis stimulating agents in imerge phase 3

Topic: 10. Myelodysplastic syndromes - Clinical Background: A main therapeutic goal in lower-risk myelodysplastic syndromes (LR-MDS) is to alter disease biology by eradicating malignant clones, but not at the expense of toxicity. MDS-initiating cells carrying cytogenetic abnormalities, mutant alleles, or both arise from malignant stem and progenitor cells. SF3B1, involved in RNA splicing, and TET2, involved in DNA methylation, are recurrently mutated genes in LR-MDS, which can be quantified by measuring change in variant allele frequency (VAF) to denote disease burden. In IMerge phase 3 (NCT02598661), treatment with imetelstat, a direct and competitive inhibitor of telomerase, demonstrated statistically significant and clinically meaningful efficacy over placebo with an ≥8-week transfusion independence (TI) rate of 40%. Further understanding imetelstat’s potential disease-modifying activity could provide additional clinical benefits for patients. Aims: Evaluate cytogenetic response, VAF changes, and clinical correlates in patients in IMerge phase 3. Methods: Patients with heavily red blood cell (RBC) transfusion-dependent (TD) erythropoiesis stimulating agent (ESA) relapsed/refractory (R/R) or ineligible non-del(5q) LR-MDS naive to lenalidomide and hypomethylating agents (len/HMA) received 2-hr infusions of imetelstat 7.5 mg/kg or placebo every 4 weeks. An independent review committee assessed abnormal cytogenetic profiles of bone marrow samples taken pre- and every 24 weeks post-treatment to assess cytogenetic response of complete or partial remission. VAF changes were assessed from blood taken pre- and every 12 weeks post-treatment in between bone marrow sampling in patients with ≥5% variant allele at baseline and ≥1 post-baseline assessment. T-tests evaluated group comparison; Fisher’s exact test evaluated association, and linear regression evaluated correlation. Results: Of the 178 randomized patients, 22.0% in the imetelstat group and 21.7% in the placebo group had baseline cytogenetic abnormalities. Cytogenetic response was observed in 9/26 patients (34.6%, 95% CI=17.2–55.7) in the imetelstat group and 2/13 patients (15.4%, 95% CI=1.9–45.5) in the placebo group. Imetelstat-treated patients demonstrated a higher rate of ≥50% VAF decreases in SF3B1, TET2, DNMT3A, and ASXL1 mutations as compared with placebo. The patients achieving ≥8-week, ≥24-week, and ≥1-year TI in the imetelstat group were enriched with patients who had reductions in SF3B1 and TET2 VAF compared with placebo (Figure). Additionally, both 8- and 24-week TI responders in the imetelstat arm had significantly greater reductions in SF3B1 VAF vs non responders (P<0.001, for both). Importantly, greater reductions in SF3B1 VAF in the imetelstat arm correlated significantly with hemoglobin increase; r=−0.626, P<0.001, and longer TI duration; r=−0.549, P<0.001. Summary/Conclusion: In the phase 3 IMerge study, heavily RBC TD ESA R/R/ineligible non-del(5q) LR-MDS patients naive to len/HMA treated with imetelstat experienced more cytogenetic response and reduction in SF3B1, TET2, DNMT3A, and ASXL1 mutational burden compared with placebo. Furthermore, SF3B1 VAF reduction correlated with clinically meaningful endpoints of increased hemoglobin and TI duration. These data, taken together with robust rates of TI that are continuous and durable, may indicate improvement of the ineffective erythropoiesis characteristic of LR-MDS and suggest imetelstat may alter the underlying biology of disease in these patients. Amer M. Zeidan and Rami S. Komrokji contributed equally.Keywords: Telomerase activity, Therapy, Mutation analysis, Myelodysplastic syndrome