Pretreatment with granulocyte colony stimulating factor (G-CSF) failed to improve dendritic cell (DC) collection for an autologous DC-based vaccine for advanced, resectable melanoma, according to results of a phase 2b study presented at the virtual Society for Immunotherapy of Cancer 35th Anniversary Annual Meeting & Preconference Programs (SITC 2020).
G-CSF pretreatment causes a smaller blood draw when harvesting DCs, but may suppress the antigen-presenting activity of DCs, thereby reducing response to a DC-based vaccine. This study aimed to evaluate the effect of G-CSF pretreatment on the efficacy of a tumor lysate, particle-loaded, DC (TLPLDC) vaccine.
The double-blind phase 2b study randomly assigned 144 patients with stage III/IV melanoma in a 2:1 ratio to receive TLPLDC vaccine or placebo. G-CSF pretreatment was administered to patients based on decisions by their health care providers. The primary endpoint was 24-month disease-free survival (DFS) and secondary endpoints included 36-month DFS and overall survival (OS).
Autologous DCs were obtained from peripheral blood and then incubated with autologous tumor cell lysate. DCs were then injected back into the patient. Vaccination occurred 6 times during the study.
Patient demographics were similar between groups, except that the median patient age was 69.5, 61.9, and 58.7 years among patients in the TLPLDC alone, TLPLDC plus G-CSF pretreatment, and placebo arms, respectively.
DFS was shorter with G-CSF pretreatment, with a rate of 27.1% at 36 months compared with 51.8% among patients who did not receive G-CSF and 23.4% in the placebo group (P =.027).
Patients with stage IV disease demonstrated a more pronounced difference in 36-month PFS, with a rate of 68.6% in the no G-CSF group compared with 18.8% and 0% among patients in the G-CSF pretreatment and placebo arms, respectively.
The association with G-CSF pretreatment was also strong among patients who had previously received immunotherapy prior to receiving the vaccine. The 36-month DFS was 61.9% among patients who did not receive G-CSF compared with 35.7% and 11.5% among patients who did receive placebo or G-CSF, respectively.
The 36-month OS was also shorter in the G-CSF and placebo arms, at 62.8% and 70.3%, respectively, compared with 92.9% among patients who did not receive G-CSF (P =.022).
The authors concluded that “further studies will examine the possibility that G-CSF mobilization leads to collection of a phenotypically different subset of DCs.” They added that a phase 3 trial of the vaccine plus G-CSF is ongoing.
O’Shea AE, Chick RC, Clifton GT, et al. The effect of pretreatment with G-CSF prior to dendritic cell collection during the phase IIb trial of an autologous DC-based vaccine for advanced, resectable melanoma. Presented at: Society for Immunotherapy of Cancer 35th Anniversary Annual Meeting & Preconference Programs (SITC 2020); November 11-14, 2020. Abstract 310. J Immunother Cancer. 2020;8(Suppl 3):A656–A959.