A Study of Tbo-Filgrastim (Granix) to Disrupt the Bone Marrow Microenvironment in Patients with Multiple Myeloma Undergoing Autologous Stem Cell Transplantation

Multiple myeloma is the second most common hematologic malignancy in the US. The current standard of care for transplant eligible patients is therapy with high-dose melphalan (HDM) followed by autologous stem cell transplant (ASCT). Although ASCT improves progression-free (PFS) and overall survival (OS), it is not curative and virtually all patients will relapse. Attempts to improve upon HDM by adding other agents to transplant protocols have largely resulted in unacceptable increases in toxicity.

Preclinical studies performed by our group suggest that granulocyte-stimulating factor (G-CSF) disrupts the bone marrow microenvironment, resulting in a striking loss of plasmablasts, plasma cells, and decreased expression of chemokine/cytokines contributing to plasma cell maintenance. Tbo-filgrastim (Granix, Teva Pharmaceuticals) is a recombinant methionyl human granulocyte colony-stimulating factor. We hypothesized that tbo-filgrastim treatment may provide a potent and well-tolerated method to disrupt the 'myeloma cell niche', rendering patients more sensitive to HDM.

Methods: Here, we report results from an open label, single center, phase II randomized study to test the efficacy and safety of tbo-filgrastim plus HDM (tbo-filgrastim arm) versus HDM alone (SOC arm) prior to ASCT (NCT02112045). Patients were randomized 1:1 to tbo-filgrastim on Day -7 through Day -2 (480 or 960 mcg/day, based on weight) and melphalan on Day -2 prior to ASCT (140 or 200 mg/m2, based on age) or melphalan alone. The primary objective was to compare CR rate at day 100. Secondary objectives included comparison of the toxicity, overall response rate, PFS, OS, and rate of neutrophil and platelet engraftment between the two arms. Eligible patients were ≥18 years with symptomatic multiple myeloma enrolled within 12 months of receiving at least 2 cycles of any systemic therapy, were undergoing their first ASCT, and had an adequate ASCT collection product (at least 2 million CD34+ cells/kg). Target enrollment was 176 patients, with an interim analysis of efficacy and futility planned after 88 patients reached Day 100 post-ASCT. Early stopping rules for unacceptable toxicity were in place. Responses were evaluated by IMWG criteria.

Results: Ninety patients were enrolled (median age 59.5, range 33 to 77) and 89 were evaluable for response. The early stopping rules for toxicity were not met. The planned interim analysis showed that the proportion of patients in CR at Day 100 was similar between the arms and the study was halted for futility (39.5% on the tbo-filgrastim arm vs. 37.8% on SOC arm). The overall response rate (CR + VGPR + PR) between the tbo-filgrastim and the SOC arm was 95% vs 93%, respectively. At the interim analysis, with median follow-up time for the study of 21.7 months, (range 8.8 to 25.8), the median PFS and OS had not been reached for either arm. There was no difference in PFS between the tbo-filgrastim and the SOC arm (84% vs 80%, respectively, p=0.60). There was no difference in OS between the tbo-filgrastim and the SOC arm (90.9% vs 95.6%, respectively, p=0.43). All patients in the study achieved neutrophil (ANC > 0.5 K/cumm) and platelet (> 20 K/cumm) engraftment. The median time to neutrophil engraftment for the tbo-filgrastim arm was 5 days (range, 3-9) vs 4 days (range, 3-7) in the SOC arm, p<0.001. There was no difference in the median time to platelet engraftment between the arms (11 days in the tbo-filgrastim arm, range 2-23, and 10 days in the SOC arm, range 1-24, p=0.67). Adverse events for both arms were typical of those observed in the ASCT population.

Conclusions: The administration of tbo-filgrastim in the setting of HDM prior to ASCT is feasible, without excess toxicity or loss of engraftment. There was no difference in Day +100 CR or ORR rates, PFS, or OS in patients treated with tbo-filgrastim plus HDM versus HDM alone with a median follow-up of 21.7 months. The lack of efficacy may be secondary to the high pre-ASCT response rates seen with modern agents.