Immunogenicity Assessment of Tbo-Filgrastim in Cancer Patients Receiving Chemotherapy

Introduction: Neutropenia is the primary cause of chemotherapy (CT) delays, interruptions, and dose reductions, potentially compromising patient outcomes, including survival and complete response rates. Granulocyte colony-stimulating factors (G-CSFs), which support the proliferation, differentiation, and activation of hematopoietic cells, reduce the incidence, severity, and duration of febrile neutropenia in CT patients. Tbo-filgrastim (Granix®) with a non-glycosylated methionyl group is a short-acting version of G-CSF approved since 2012 under a biologic license application in the United States. Recombinant proteins may induce an immune reaction. Generation of antidrug antibodies (ADAs) may have serious consequences, including anaphylaxis resulting from hypersensitivity, neutralization of endogenous counterparts by cross-reactive neutralizing antibodies (NAbs), and reduction of drug efficacy. The objective of this analysis is to evaluate the immunogenicity of tbo-filgrastim, including the presence of the binding antibodies and neutralizing antibodies against tbo-filgrastim and its endogenous counterpart (native G-CSF), in patients receiving CT for solid and hematologic malignancies.

Methods: Blood samples were collected during three independent phase 3 clinical studies that compared the efficacy and safety of tbo-filgrastim with filgrastim and/or placebo, administered 1 day after CT for 5 to 14 days. A 3-tier approach (screening, confirmation, and titer assay) was implemented to evaluate binding ADAs for tbo-filgrastim using a validated homogeneous enzyme-linked immunosorbent assay. Neutralizing activity was assessed for the confirmed-positive ADA test samples using a validated cell-based proliferation assay based on the ability of NAbs from the test samples to inhibit tbo-filgrastim or glycosylated G-CSF-stimulated proliferation of NFS-60 cells in vitro . Clinical measures were examined for all patients with confirmed-positive samples of ADAs for a possible correlation between the presence of ADAs and the potential clinical impact of immunogenicity.

Results: A total of 436 patients with cancer who received the drug were assessed for immunogenicity of tbo-filgrastim, including 213 patients with breast cancer, 160 patients with lung cancer, and 63 patients with non-Hodgkin lymphoma (NHL). From breast cancer patients, only 3 patients, each having one positive sample, were confirmed positive for ADAs with low titer; none of these samples showed cross-reactivity to endogenous (glycosylated) G-CSF or neutralizing antibody activity. From lung cancer patients, only 3 patients, each having one positive sample, were confirmed positive for ADAs with low titer; none of these ADA-positive samples cross-reacted with endogenous G-CSF, and one pre-dose sample showed NAb activity in cycle 1 before tbo-filgrastim treatment. From NHL patients, only 1 patient was confirmed positive for ADAs on cycle 3. This patient had only one positive post-dose sample. The ADA titer was too low to be determined; however, no cross-reactivity or NAb activity was associated with this NHL sample. Overall, the immunogenicity incidence was 1.6% (7/436) in all cancer patients tested. None of the patients with a confirmed-positive ADA sample had evidence of hypersensitive or anaphylactic reactions or loss of efficacy.

Conclusions: Incidence of immunogenicity or treatment-emergent ADAs in patients receiving CT and tbo-filgrastim was low across all three cancer populations, and all positive samples had low-titer ADAs. None of positive samples had cross-reactive ADAs to the endogenous G-CSF. Except for one pre-dose sample that showed neutralizing activity towards tbo-filgrastim, no neutralizing activities were found in other positive samples towards tbo-filgrastim or endogenous G-CSF. None of the patients with a positive ADA sample had clinically relevant immunogenicity-related symptoms during the conduct of the study.