A monoclonal cytolytic T-lymphocyte response observed in a melanoma patient vaccinated with a tumor-specific antigenic peptide encoded by gene MAGE-3

Abstract

Vaccination of melanoma patients with tumor-specific antigens recognized by cytolytic T lymphocytes (CTL) produces significant tumor regressions in a minority of patients. These regressions appear to occur in the absence of massive CTL responses. To detect low-level responses, we resorted to antigenic stimulation of blood lymphocyte cultures in limiting dilution conditions, followed by tetramer analysis, cloning of the tetramer-positive cells, and T-cell receptor (TCR) sequence analysis of the CTL clones that showed strict specificity for the tumor antigen. A monoclonal CTL response against a MAGE-3 antigen was observed in a melanoma patient, who showed partial rejection of a large metastasis after treatment with a vaccine containing only the tumor-specific antigenic peptide. Tetramer analysis after in vitro restimulation indicated that about 1/40,000 postimmunization CD8(+) blood lymphocytes were directed against the antigen. The same TCR was present in all of the positive microcultures. TCR evaluation carried out directly on blood lymphocytes by PCR amplification led to a similar frequency estimate after immunization, whereas the TCR was not found among 2.5 x 10(6) CD8(+) lymphocytes collected before immunization. Our results prove unambiguously that vaccines containing only a tumor-specific antigenic peptide can elicit a CTL response. Even though they provide no information about the effector mechanisms responsible for the observed reduction in tumor mass in this patient, they would suggest that low-level CTL responses can initiate tumor rejection.

Figures

Figure 1

Detection of anti-MAGE-3.A1 CTLp. (A) Clinical evolution of patient CP64. (B) Detection of anti-MAGE-3.A1 CTL in postimmunization PBMC. PBMC were thawed, incubated with peptide MAGE-3.A1, washed, and distributed at 3.106 cells/well in the presence of IL-2, -4, and -7. On day 7, the cultures were restimulated with peptide and cytokines. On days 7 and 13, the cells were incubated with the A1/MAGE-3 tetramer coupled to PE and with a control HLA-A1 tetramer containing an influenza peptide and coupled to APC. Anti-CD3 antibodies coupled to FITC and anti-CD8 antibodies coupled to peridinin chlorophyll protein were then added. Cells were washed, fixed, and analyzed by flow cytometry. One million events were acquired, but the plots include only the CD3+CD8+ lymphocytes. The lytic activity of CTL clones, derived from two populations of cells labeled with tetramer, was tested on HLA-A1 EBV-B cells incubated or not with the MAGE-3.A1 peptide (2 μM), natural killer target cells K562, and allogenic melanoma cells MZ2-MEL, which naturally express the MAGE-3.A1 antigen. The pattern of lysis of two representative clones is shown. (C) Estimation of the frequency of anti-MAGE-3.A1 CTLp in postimmunization PBMC. Thirty-six cultures were set up with 160,000 Post II PBMC stimulated with the MAGE-3.A1 peptide and IL-2, -4, and -7. The lymphocytes were restimulated on day 7 by the addition of peptide and cytokines and labeled on day 15 with tetramers and anti-CD3 and anti-CD8 antibodies. Only CD3+CD8+ lymphocytes are included in the plots. Clusters of lymphocytes specifically labeled with the A1/MAGE-3 tetramer are boxed, and their proportion among the CD3+CD8+ cells is indicated. (D) Analysis of prevaccination PBMC. Five cultures of 3 million PBMC were set up and analyzed as in C.

Figure 2

Phenotype of PBMC expressing TCR 48. Postimmunization (Post II) PBMC were thawed and labeled either with anti-CD45RO antibodies coupled to FITC and anti-CD45RA antibodies coupled to PE, or with an anti-CCR7 IgM antibody followed by biotinylated anti-IgM antibodies and streptavidin coupled to PE. The cells were immediately sorted by flow cytometry. Sorted populations were divided in groups and cDNA obtained from each group was tested for the TCRα and β clonotypic PCR. Numbers of CTL 48 were calculated with the Poisson distribution.