Association of homogeneous inflamed gene signature with a better outcome in patients with metastatic melanoma treated with MAGE-A3 immunotherapeutic

Abstract

Purpose: This study assessed clinical activity, safety and immunogenicity of MAGE-A3 immunotherapeutic in patients with MAGE-A3-positive metastatic melanoma.

Patients and methods: In this open-label, multicentre, uncontrolled, Phase II study (ClinicalTrials.gov NCT00896480), patients received ≤24 doses of MAGE-A3 immunotherapeutic (4-cycle schedule). At screening, two skin lesions were biopsied for MAGE-A3 expression analysis and presence/absence of a previously identified gene signature (GS) associated with favourable clinical outcome. Clinical activity was assessed in terms of clinical response, time-to-treatment failure (TTF) and progression-free survival (PFS). Adverse events (AEs) and serious AEs (SAEs) were recorded. MAGE-A3-specific immune responses were assessed. Clinical activity and immunogenicity were analysed overall and separately in patients with 2/2 (GS+/+), 1/2 (GS+/-) or 0/2 (GS-/-) biopsies presenting GS.

Results: Of 49 screened patients, 32 had MAGE-A3-positive tumours; 24 (8 GS+/+, 8 GS+/-, 8 GS-/-) were treated. Two complete (GS+/+ patients) and two partial responses (one GS+/+, one GS+/-) were reported; of note, one of the two complete responses was unlikely to be related to the study treatment. Median TTF and PFS were 14.8 and 7.2 months for GS+/+, 2.3 and 2.8 months for GS+/- and 2.4 and 2.9 months for GS-/- patients. Three grade 3 AEs and two SAEs unrelated to treatment were reported. All patients were seropositive for MAGE-A3 antibodies on vaccination with no differences between the different GS profiles. MAGE-A3-specific CD4+ and CD8+ T cell immunogenicity was detected; 12/16 (75.0%) of patients presented CD4+ T cell responses.

Conclusion: Treatment with MAGE-A3 immunotherapeutic showed signs of clinical activity in GS+/+ patients. Treatment was well tolerated and immunogenic. No differences in immune responses according to GS status were observed.

Trial registration number: NCT00896480 (Results).

Keywords: clinical activity; gene signature; mage-a3 immunotherapeutic; melanoma; safety.

Conflict of interest statement

Competing interests: MG, JL, FU-M and SJ are employees of the GSK group of companies. FFL and PMDSA were employees of the GSK group of companies during the conduct of the study. FFL, JL, PMDSA, FU-M and SJ hold shares in the GSK group of companies as part of their employee remuneration.

Figures

Figure 1

Participant flow. Of the 24 patients included in the study, 3 patients completed the study. Patients may have more than one reason for elimination from ATP population. ATP, according-to-protocol; GS+/+, patients presenting the gene signature on both biopsies; GS+/-, patients presenting the gene signature on one biopsy and not on the other one; GS-/-, patients without the gene signature; N, number of patients.

Figure 2

TTF and PFS (total treated population). TTF was longer in patients for whom, out of the two biopsied lesions, both were GS-positive (GS+/+) (14.8 months) than in patients with only one GS-positive biopsy (GS+/-) (2.3 months) and in patients with both biopsies being GS-negative (GS-/-) (2.4 months) (A). PFS was longer for GS+/+ patients (7.2 months) than for GS+/- patients (2.8 months) and GS-/- patients (2.9 months) (B). PFS, progression-free survival; TTF, time-to-treatment failure.

Figure 3

MAGE-A3-specific geometric mean titres (GMCs) (ATP population for immunogenicity). In the overall population, 2 weeks postdose 2, all patients were seropositive for MAGE-A3-specific antibodies (GMC 1865.7 EU/mL). The antibody levels further increased up to postdose 6 timepoint (GMC 9080.5 EU/mL); thereafter, a plateau was observed (A). There were no marked differences in MAGE-A3-specific antibody profiles between patients presenting the different gene signature on the two biopsies (B). The error bars represent 95% CI. ATP, according-to-protocol; EU, ELISA units; GMC, geometric mean concentration; GS+/+, patients presenting the gene signature on both biopsies; GS+/-, patients presenting the gene signature on one biopsy and not on the other one; GS-/-, patients without the gene signature; Pre, baseline; Post-, postdose number indicated by Arabic numerals.

Figure 4

MAGE-A3-specific cellular responses (ATP population for immunogenicity). Prior to the first MAGE-A3 immunotherapeutic administration, MAGE-A3-specific double-positive TNF-α/IFN-γ-producing (TNF-α/IFN-γ++) CD4+ T cells with immunogenicity score above the assay cut-off value (1.24) were found in one patient; the highest proportion of patients with these cells was observed postdose 12 (9/10 patients (90.0%)) (A). The highest proportion of CD4+ T cell responders was observed postdose 6 at the end of cycle 1 (8/13 patients (61.5%)) (B). CMI response was defined as GMR is above the cut-off value (1.24) and at least a 4-fold increase after immunisation as compared with the patient’s baseline value. The numbers in brackets indicate numbers of patients with GMR ≥cut-off value (A) or number of patients with CMI response (B). ATP, according-to-protocol; CMI, cell-mediated immune; GM, geometric mean of MAGE-A3-specific immunogenicity score (GMR) calculated on all patients; GMR, geometric mean ratio; IFN-γ, interferon gamma; Pre, baseline; Post-, postdose number indicated by Arabic numerals; TNF-α, tumour necrosis factor-alpha.