Peptide epitopes for melanoma-reactive cytotoxic T-cells were first identified in 1991, epitopes for helper T cells have been identified in recent years. These agents have been employed in experimental melanoma vaccines over the past 10 years or less. In some of those vaccines, it has been possible to generate antigen-specific T cells at high frequencies. However, clinical tumor regressions have been rare. It is increasingly clear that successful immune therapy of cancer depends on combining multiple approaches, including at least these three: (1) augmentation of T-cell responses, (2) activation of tumor-associated T cells, and (3) reversal of tumor-mediated immune suppression and tolerance. Low clinical response rates with peptide vaccines should not be surprising, when single peptides are used as immunogens. Antigenic heterogeneity is the rule in tumor deposits. Furthermore, T cells infiltrating tumor deposits are commonly found to be anergic or poorly responsive to antigenic stimulation, leading to the perception that the tumor microenvironment is hostile to the T-cell response. Effective immune therapy will require induction of T-cell responses to large numbers of antigens simultaneously, and to maintenance of T-cell activation in the tumor deposits. Though additional approaches to block immunoregulatory mechanisms may well be needed as well, it is necessary that a strong and persistent effector function be induced. Our current studies focus on the development of complex multipeptide vaccines, the validation of the vaccine preparations, the evaluation of the ability to induce multivalent immune responses, and on adding melanoma-reactive helper T-cell responses to cytotoxic T-cell responses.

We have prepared a mixture of 12 peptides from cancer-testis antigens and melanocyte differentiation proteins. To assess the immunogenicity of this preparation, and to assess the feasibility of vaccinating with complex mixtures, 51 patients were enrolled in a randomized trial, with arm 1 receiving 4 peptides and arm 2 receiving 12 peptides. Vaccines were administered as an emulsion of 100 µg of each peptide, 190 µg of a tetanus helper peptide, 1 ml Montanide ISA-51 adjuvant, and 110 µg GM-CSF. T-cell responses were evaluated in peripheral blood lymphocytes and in a sentinel immunized node by IFN-gamma ELIspot assay after one in vitro sensitization. All patients have completed treatment, and immunologic assessment has been completed on 31 patients. Preliminary immunologic results from these 31 patients revealed immune responses in 11/14 patients (79%) on arm 1 and 17/17 patients (100%) on arm 2. Responses to 9 of the peptides are listed in the Table that follows.

Patients on arm 2 had immune responses to an average of 2.8 different peptides, whereas those on arm 1 had responses to an average of only 0.9 peptides. By comparing T cell responses to three index peptides in both peptide mixtures, it will be possible to assess whether co-administration of 4 peptides binding the same MHC molecule interferes with immunogenicity of the index peptides. Current results indicate that the median of the best immune responses for index peptides (fold-increase over background) was 9.7 in arm 1 and 11.1 in arm 2. The mean number of responses to index peptides per patient was 0.7 for Group 1, and 1.1 for Group 2. These findings support that (1) the most immunogenic peptides in this study were from tyrosinase, gp100, MAGE-A1, and MAGE-A10; (2) multiple peptides can be administered in combined aqueous mixture with safety and immunogenicity, and (3) immunogenicity is maintained despite peptide competition for binding to MHC. Further investigation of multiple-peptide vaccine mixtures is warranted.

Despite the ability to generate T cell responses to multiple antigens, this approach is limited by the absence of melanoma-specific helper T cell responses. Now that multiple helper epitopes from cancer-testis antigens and melanocytic differentiation proteins have been defined, we have prepared a mixture of 6 such peptides restricted by HLA-DR1, 4, 11, 13, and/or 15. An ongoing phase I trial (UVA-Mel41) of 200-800 µg of each peptide provides evidence of safety thus far at all doses, and preliminary data reveal immune responses to an average of approximately 3 different peptides per patient.

Studies are planned for the incorporation of the 6 helper peptides with the 12 CTL epitope peptides both in the adjuvant setting (UVA-Mel44) and in the setting of advanced stage IV melanoma (ECOG 1602). By combination of these peptide mixtures, it will be possible to assess whether this 18 peptide combination induces greater CTL responses, and whether CTLs interacting with tumor cells in vivo maintain activation by the effects of melanoma reactive helper T cells that colocalize to the tumor.

One of the advantages of peptide vaccines is that immunologic monitoring can permit the evaluation of the frequency of specific antigen-reactive T cells, and their functional phenotype in multiple compartments, and these effects can be quantified in response to other immune modulators. The addition of immunomodulating chemotherapy agents (e.g.: cyclophosphamide) and of novel adjuvants (TLR agonists) will also be evaluated in future studies, with the intent of optimizing the therapeutic potential of peptide vaccines and for the purpose of evaluating the effects of those immunomodulatory agents in patients with melanoma.

This abstract was published in Cancer Immunity, a Cancer Research Institute journal that ceased publication in 2013 and is now provided online in association with Cancer Immunology Research.

T-cell responses detected in Mel39 trial.

Source protein Residues HLA restriction N Response 
Tyrosinase 240-251 A1 10 100% 
MAGE-A3 168-176 A1 50% 
Tyrosinase 368-376D A2 17 47% 
gp100 209-217-2M A2 100% 
gp100 280-288 A2 17 12% 
MAGE-A10 254-262 A2 78% 
gp100 17-25 A3 13 77% 
gp100 614-622 A3 44% 
MAGE-A1 96-104 A3 67% 
Source protein Residues HLA restriction N Response 
Tyrosinase 240-251 A1 10 100% 
MAGE-A3 168-176 A1 50% 
Tyrosinase 368-376D A2 17 47% 
gp100 209-217-2M A2 100% 
gp100 280-288 A2 17 12% 
MAGE-A10 254-262 A2 78% 
gp100 17-25 A3 13 77% 
gp100 614-622 A3 44% 
MAGE-A1 96-104 A3 67%