A randomized pilot trial evaluating safety and immunogenicity of recMAGE-A3 + AS15 immunotherapeutic administered by intramuscular versus intradermal/subcutaneous routes

Introduction

The recMAGE-A3 protein has been administered intramuscularly (IM) with immunostimulant AS15 as an experimental immunotherapeutic. AS15 contains 3-O-desacyl-4'-monophosphoryl lipid A (MPL), QS-21, CpG 7909 and liposome. This MAGE-A3/AS15 immunotherapeutic has not been studied for intradermal (ID) or subcutaneous (SC) use. A clinical trial (NCT01425749) was initiated to test the hypotheses that ID/SQ administration is safe and may induce CD4⁺ and CD8⁺ T cell responses to MAGE-A3.

Patients and methods

Twenty-five eligible patients with resected stage IIB-IV MAGE-A3⁺ melanoma were randomized to 2 arms, treated with MAGE-A3/AS15 Immunotherapeutic IM (Arm A, n = 13) or ID/SC (Arm B, n = 12). Adverse events (CTCAE 4) were recorded. Antibody (Ab) responses to MAGE-A3 protein were assessed by ELISA assay. T cell responses were assessed by flow cytometry after intracellular cytokine staining (ICS) for multifunctional CD4⁺ and CD8⁺ responses to overlapping MAGE-A3 peptides, assaying lymphocytes from peripheral blood (PBMC) and sentinel immunized node (SIN), after one in vitro stimulation.

Results

In both arms, the recMAGE-A3/AS15 immunotherapeutic was well-tolerated, with only one grade 3 treatment-related adverse event (hyperglycemia, Arm B), and no grade 4 or 5 events. Grade 2 injection site reactions were observed in 10 patients in Arm A and 7 in Arm B (P > 0.3). Ab responses were detected in all patients, most with high titers persisting at least 6 months, without difference between arms. Preliminary T cell data are that multifunctional (IFNg and TNFα) CD4⁺ T cell responses to MAGE-A3 were detected in 64% of patients (54% A; 75% B; Table 1). Multifunctional CD8⁺ T cell responses were evident in 20% of patients (8% A, 33% B). CD4⁺ responses were higher magnitude in SIN than in PBMC.

Table 1 Multifunctional (IFNg and TNFα) T cell responses to MAGE-A3

Conclusion

Safety profiles were comparable for ID/SC and IM administration of the MAGE-A3/AS15 immunotherapeutic, which induced high-titer Ab, multifunctional CD4⁺ Th1 responses, and CD8⁺ responses when administered by either route. Immune responses were more readily detected in the SIN than in PBMC. These pilot data support further investigation of ID/SC immunization with antigen plus AS15 to support Th1 CD4⁺ responses and CD8⁺ responses. Production of Th1 cytokines IFNg and TNFα suggests the induced CD4⁺ responses may support CD8⁺ T cells. Other forms of antigen (e.g.: long peptides) may further support induction of CD8⁺ T cell responses in combination with AS15.

Funding source: GlaxoSmithKline Biologicals SA