A phase II study evaluating the efficacy and safety of imetelstat in patients with advanced myelodysplastic neoplasms or AML failing HMA-based therapy – interim analysis results of the IMpress study

Abstract Platzbecker U, Lane S and Adès L contributed equally. A study by the European Myelodysplastic Neoplasms Cooperative Group (EMSCO) Introduction: Hypomethylating agents (HMAs), either as monotherapy or combined with venetoclax (VEN), are standard of care for patients with higher-risk MDS (HR-MDS) or AML ineligible for intensive chemotherapy. However, most patients fail to respond or relapse. Outcomes for progress after HMA-based therapy remain poor, with limited effective options. Imetelstat is a potent selective telomerase inhibitor recently approved by FDA and EMA for transfusion-dependent lower-risk (LR) MDS patients ineligible or relapsed/refractory to ESA. Its clinical efficacy in HR-MDS or AML has not yet been established. Methods: The multicenter phase 2 IMpress trial (NCT05583552), led by EMSCO, evaluates safety and efficacy of imetelstat sodium in HR-MDS or AML patients refractory, relapsing or intolerant following at least 6 or 4 cycles of either azacitidine (AZA) or decitabine (DAC), respectively, or 3 cycles of VEN/AZA. In cohort 1 (ASH 2024), patients received 7.5 mg/kg i.v. once every 4 weeks for 4 cycles. Due to a lack of response and no new safety signals, the protocol was amended to biweekly dosing. The primary endpoint (PE) was overall response rate after 4 months (CR, CRi, PR, HI). All patients achieving CR, CRi, PR or HI after 4 months of imetelstat were considered responders and allowed to continue treatment until loss of response/disease progression. Non-responding patients stopped treatment after 4 months. Results: Between Aug-Dec 2024, 24 patients from 8 centers in Germany, France and Australia were screened for the second cohort. Of these, 23 (MDS=6, AML=17) received ≥1 dose of imetelstat. The median number of doses was 5/patient, compared to 3 in the first cohort. Median age was 77 years in both first (68–87) and second cohort (66–88). More male than fem. patients were enrolled, with 15M and 8F in the first cohort, and 14M and 9F in the second. The median ECOG performance status was 1 (0–2) in both cohorts. Complex karyotype was reported in 6/23 patients across both cohorts. Median percentage of bone marrow blasts at screening was 30% (8–91) in the first cohort and 27% (5.5–86) in the second. Prior treatment with VEN was reported in 47.8% (11/23) of the first cohort and 39.1% (9/23) of the second. In the second cohort, 6/23 reached the primary endpoint visit, scheduled after 4 cycles of treatment, compared to none in the first cohort: 3 had stable disease (SD) and 3 showed progressive disease (PD). One patient with SD achieved neutrophil response (HI-N) and received 3 additional doses in the extension phase. 13 of the 23 patients reached the first (preliminary) disease assessment after 2 cycles. At this point, only 1 patient showed a response (SD with HI-N), 9 patients had SD and 2 had PD. As of 17th June 2025, the median OS of the first and second cohort was 119 (95% CI: 83-144) and 102 days (95% CI: 67-TBD), respectively. The median PFS of the first and second cohort was 69 (95% CI: 57-113) and 92 days (95% CI: 57-TBD), respectively. In the second cohort, 24 SAEs occurred in 19 patients from August 2024 until 8th May 2025, of which none were deemed related to imetelstat. The most common SAEs were febrile neutropenia (n=6), disease progression/transformation to AML (n=4), sepsis (n=2) and fever (n=2). Overall, 11/24 SAEs resulted in death (n=4 disease progression or AML, n=2 cardiac/cardiorespiratory arrest, n=1 febrile neutropenia, n=1 sepsis, general health alteration and death of unknown reason). 8 of the SAEs resolved without sequelae and 2 remained unresolved. In the first cohort, 30 SAEs occurred in 18 patients, with 3 deemed possibly related to imetelstat (pneumonia [n=2], febrile neutropenia [n=1]). The most common SAEs were disease progression/transformation to AML (n=9), pneumonia (n=3), febrile neutropenia (n=3), and sepsis (n=2). 10/30 SAEs resulted in death. Despite higher drug exposure in the second cohort, no increase in toxicity was observed. Conclusion: Increased exposure to imetelstat in refractory, relapsing or intolerant AML/HR-MDS patients did not appear to be associated with additional toxicity. PE visit completion rose from 0/23 patients to 6/23 (26%) in the second cohort, despite only one achieving HI, and no marrow responses were observed. One patient remains on treatment, and 2 are in follow-up. Imetelstat has limited single agent activity in this very adverse prognosis group.