Results of a Phase I/II Study of Tagraxofusp in Combination with Decitabine for Patients with Myelodysplastic/Myeloproliferative Neoplasms and Higher Risk Myelodysplastic Syndromes

INTRODUCTION: Clinical outcomes of patients with myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemia (CMML) after failure to hypomethylating agent (HMA) therapy remain dismal. Distinct hematopoietic populations which contribute to disease progression after HMA, such as myeloid progenitors, myeloid-derived suppressor cells or plasmacytoid dendritic cells, exhibit cell-surface expression of the IL-3 receptor alpha chain (IL3RA or CD123). Tagraxofusp (TAG), a recombinant fusion protein consisting of human IL-3 conjugated to a truncated diphtheria toxin, has demonstrated preliminary activity in myeloid malignancies. Here, we report the safety and activity of TAG in combination with decitabine in patients with higher-risk MDS and CMML previously treated with HMA.

METHODS: We designed a Phase I/II dose escalation study of TAG in combination with decitabine for patients with higher-risk MDS, defined as MDS with IPSS-R score >3.5 or with TP53 mutation, with >5% bone marrow blasts, and CMML-1/CMML-2 who had no response after 6 cycles of HMA or relapse or progression after any number of cycles of HMA. Four dose levels (DL) of TAG administered days 1-3 were planned: DL1: 5µg/kg/day, DL2: 7µg/kg/day, DL3: 9µg/kg/day, DL4: 12µg/kg/day. Decitabine was administered at 20mg/m²/day on days 1-5. A maximum of 7 cycles of therapy with TAG could be administered. In patients who achieved response to therapy, treatment with decitabine could be continued after cycle 7 until loss of response or disease progression. The primary endpoint was to evaluate the safety, efficacy, and tolerability of the combination. Responses were evaluated by the 2006 IWG criteria.

RESULTS: Nine patients diagnosed with high-risk MDS and CMML were enrolled between August 2022 and July 2024. The median age was 75 years (range 59-83). Five patients had MDS, and four had CMML. Among MDS patients 4 (80%) and 1 (20%) had very high and high risk by IPSS-R, respectively. All MDS patients had very high-risk by Molecular IPSS including 3 (60%) patients with TP53 mutations. Among CMML patients, 1 (20%) had intermediate-1, 3 (60%) had intermediate-2, and 1 (20%) had high risk by CPSS-Mol score. The median number of prior cycles of HMA was 18 (range 6-57). One patient (11%) had received prior allogeneic stem-cell transplantation and 1 (11%) had received prior HMA-venetoclax therapy.

Six patients (67%) were treated at DL1 and 3 (33%) at DL 2. Dose reduction of decitabine was required in 1 (11%) patient treated at DL1. The most common grade 1-2 treatment emergent adverse effects (TEAEs) included fluid retention, diarrhea, constipation and cough all in 3 (33%) patients. Febrile neutropenia, atrial fibrillation and pneumonia were each observed in 1 patient (11%). Notably, grade 3 capillary leak syndrome with concurrent cytokine release syndrome was observed in 2 patients (22%): 1 with myeloproliferative CMML (DL1), and 1 in a patient with high risk MDS (DL2). Both events occurred in patients >75 years of age during cycle 1 of therapy and reverted to baseline (< grade 1) with appropriate management including steroids, albumin and furosemide. One patient received a single dose of tocilizumab. The median number of days to cycle 2 was 34 (range 31-70). The median number of administered cycles was 1 (range 1-6). The median number of cycles to best response was 2 (range 1-3). At the time of analysis, 7 patients are evaluable for response (78%). Three patients (43% of evaluable, 33% overall) achieved response to therapy including CR in 2 (29%) patients with CMML treated at DL1 and mCR in 1 (14%) patient with MDS treated at DL2.

The median follow-up was 5.3 months (range 0-22, 95% CI 2.1-8.6 months). The median OS was 8.4 months (95% CI 5.4-11.4 months) with no differences in survival between MDS or CMML patients (8.4 vs 6.6 months, p=0.757). The median EFS in the entire population was 4.1 months (95% CI 0.5-7.7 months) and there were no significant differences in EFS between MDS and CMML patients (2.7 vs 4.1 months, p=0.472).

CONCLUSIONS: Initial results of this phase I/II study suggest that the combination of TAG with decitabine has early signs of clinical activity in both MDS and CMML. Although TAG at low doses is associated with acceptable safety profile with no new safety signals, CLS and CRS can be observed in older patients. Further follow up and enrollment is ongoing in younger (<75 years of age) patients to confirm the long-term efficacy and safety of this combination.