Validation of Nivestym compared to Neupogen: An NMDP analysis

The Coronavirus Disease 2019 (COVID-19) pandemic in early 2020 resulted in multiple challenges for the NMDP. At the time, the Investigational New Drug (IND) protocol for peripheral blood stem cell (PBSC) mobilization was limited to the use of Neupogen (NP) (Amgen, CA), which created logistical problems on the supply side as it was difficult to rely on only one medication for mobilization. In fact, there were times when NP was unable to be secured for donors and this put undue stress on the operations teams and required IND exemptions. We realized early on in 2021 that allowing filgrastim biosimilars as mobilization agents would eliminate these issues without compromising donor safety. Moreover, the World Marrow Donor Association (WMDA) had recommended that Donor Registries could use filgrastim biosimilars for PBSC mobilization initially in 2017 and then a publication in 2019.¹ Thus, in February of 2022, the IND for PBSC mobilization was amended to allow all filgrastim biosimilars, including filgrastim-aafi (Nivestym, NV), filgrastim-sndz (Zarxio) and the recombinant human granulocyte-stimulating factor (G-CSF) TBO-filgrastim (Granix) as, well as the original NP. In March of 2022, NV was the preferred PBSC mobilization agent instead of NP for stem cell mobilization for all NMDP donors. We herein present a comparative analysis of the two drugs.

The NMDP routinely collects data about donors and grafts under an IRB approved protocol. In February of 2022, most donors received NV for stem cell mobilization due to availability and cost. The primary objectives were to compare donor safety, efficacy, and adverse events between March and July of 2021 where donors received NP and March-July 2022 where donors received NV. Of note, donors who received other biosimilars or similars after February 2022 were excluded from this study. Donor toxicities were assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 4.0). All statistical analyses were performed using R statistical software version 4.2.0 (Boston, MA). Either 2 sample t-test or 2 sample test for proportions were used for 2 group comparisons, depending on whether the end point was an average or proportion. A P < .05 was considered statistically significant. Average dose values used a 95% trimmed mean to avoid skew from outliers.

There were 801 and 913 healthy donors who underwent stem cell mobilization with NP and NV, respectively. The main outcomes are summarized in Table 1. Average donor weight was similar between the groups (85 kg for NP vs 83 kg for NV). Day +5 pre-apheresis peripheral blood CD34 count and percentage did not differ between groups (95.4 cells/μL for NP vs 96.0 cells/μL for NV, P = .56; 0.22% vs 0.22%, P = .73). Median liters of blood processed was similar (16.2 for NP vs 15.8 for NV; P = .92) as was the requirement for a 2-day collection (5.0% for NP vs 5.4% for NV; P = .40). Adverse events (AE; grades 3 and higher) were comparable and numerically less in the biosimilar group with 28 in the NP group and 23 in the NV group. The most common AEs were headache, fatigue, and bone pain. There were two grade 4 AEs in the NP group (fatigue and headache) and one in the NV group (subdural hemorrhage); all have been reported previously and are expected.

In this analysis, NV performed as well as NP as a mobilization agent in healthy individuals. This is in line with other studies comparing other filgrastim biosimilars.² The first biosimilar to be approved, Zarxio was tested in a phase III trial of patients receiving chemotherapy for breast cancer.³ Based on the equivalence shown in that study, approval was granted across all indications of filgrastim. Real world data confirmed the equivalency of this biosimilar in other indications.^{4, 5} Another study examined the biochemical, pharmacodynamic, pharmacokinetic, and reviewed the clinical properties of filgrastim compared to lenograstim and concluded equivalency of that biosimilar.⁶ A recent retrospective analysis of healthy pediatric and adult donors showed that NV was as effective as NP for stem cell mobilization.⁷ The data shown here provide additional support that in a large number of volunteer unrelated allogeneic stem cell donors, NV is effective as a mobilization agent.

Donor safety remains paramount for the NMDP. Mobilization and apheresis of an adequate cell dose with a normal graft composition while maintaining donor safety is crucial for the success of transplant programs across the world. These data provide reassurance to transplant teams and apheresis centers that mobilization of volunteer unrelated donors with NV results in comparable safety compared to NP. AEs reported in this study were expected and had all been previously reported. The data presented herein give confidence to practitioners seeking standardization and cost-reduction by modifying their formulary to include filgrastim biosimilars while preserving efficacy and ensuring donor safety.

Dr Maakaron declares that the University of Minnesota receives research funding on his behalf from Gilead, CRISPR, Precision Biosciences, Scripps Calibr, FATE Therapeutics, ADC. He declares no conflict of interest. Dr Devine declares conflict of interest related to NMDP support of Orca and Vor and Sanofi for consulting. All other authors declare no conflict of interest.