Evaluation of Romyelocel-L Myeloid Progenitor Cells to Decrease Infections in De Novo AML Patients Receiving High-Dose Ara-C-Based Induction Therapy

Background: Induction chemotherapy (7+3) or high-dose ara-C-based (HIDAC) for AML results in prolonged neutropenia with a high risk of serious infection and attendant morbidity, and prolonged hospitalization. Romyelocel-L is a human off-the-shelf allogeneic myeloid progenitor cell (MPC) preparation manufactured by ex vivo culture expansion of CD34+ cells. Following infusion, MPCs are expected to home to bone marrow (BM) and produce neutrophils. In a randomized, open-label, controlled Phase 2 trial, the effect of romyelocel-L was studied in de novo patients receiving HIDAC or 7+3 induction therapy for reduction of fever and infection. The results from pooled and 7+3 cohorts, were previously presented, showing decrease in infections and days in hospital. The results from cohorts receiving HIDAC chemotherapy are presented here.

Methods: 53 subjects with de novo AML (age ≥55) and receiving HIDAC induction were randomized on Day 0 (first day of induction) to receive either romyelocel-L (7.5x10⁶ cells/kg) on Day 9 + GCSF qd starting on Day 14 (treatment) or GCSF alone starting on Day 14 (control) qd until ANC recovery to 500/µL. Of the 53 subjects, 42 were considered evaluable (20 in treatment and 22 in control) which meant that they received romyelocel-L + GCSF or GCSF alone, were in study ≥ 28 days and did not receive additional chemotherapy before Day 28. Subjects routinely received corticosteroids concurrently with HIDAC induction. Most patients received prophylactic antibacterial (74%) and antifungal (86%) agents. Endpoints assessed from Day 9 to Day 28 included days in a Febrile Episode (DFE, primary endpoint) and microbiologically defined bacterial or fungal infections (MDI, defined as an infection when a specific pathogen is identified)/clinically diagnosed infections (CDI, defined when there is supporting clinical, laboratory, and/or radiologic data that is sufficient to indicate an infection and no specific pathogen is identified). MDIs and CDIs were adjudicated by a blinded independent committee. The number of days in hospital was assessed through Day 42. Data are reported for 2 periods: from infusion of romyelocel-L (Day 9-28) and from 6 days after infusion of romyelocel-L (1 day after start of GCSF) - Day 28 (denoted as Day 15-28; the period when romyelocel-L derived neutrophils are likely to be circulating.). One-sided p-values are reported.

Results: The baseline characteristics, including median age, mean WBC, and ECOG status were balanced between the treatment and control groups. The mean DFE during Day 9-28 in the treatment and control was 6.00 and 4.04 days respectively. Absence of fever was seen in 0/20 and 9/22 (40.9%) in the treatment and control group, respectively. The mean DFE for Day 15-28 was 1.37 and 2.98 days in the treatment vs. control respectively. The incidence of MDI or CDI during Day 9-28 was 40% vs 41% in the treatment vs. control, however a significant decrease was observed during Day 15-28 with incidence of 5% vs 27.3% in the treatment vs control (p=0.027). The use of prophylactic antimicrobial agents was similar between study arms. Mean number of days in hospital were 22.6 vs 25.7 days in treatment vs. control groups (p=0.027). The median number of days to ANC recovery of 500/µl was 19.5 vs 21.5 days in treatment vs control (p=0.103). Remission rates were similar in the two groups. All seven subjects assessed for chimerism had romyelocel-L cells detected in peripheral blood (PB) on the day of infusion. Post infusion, chimerism was observed in 6/7 subjects in one or more of BM, peripheral blood and gingival rinse samples. No deaths were observed in either of the two arms within the first 42 days. Romyelocel-L was generally tolerated well.

Summary/Conclusion: The incidence of infections was decreased during the day 15-28 period and a decrease of three days in hospital stay was observed in de novo HIDAC AML subjects receiving romyelocel-L. Romyelocel-L may provide a new option to reduce infections in AML patients undergoing HIDAC induction therapy.