Cancer Antigen Discovery Program

Vaccine Composition

To create the safest and most efficacious cancer vaccine, LICR is trialing different vaccine compositions that incorporate different forms of antigen, various components that increase the immune response to the antigen, and the method of vaccine delivery.

Introduction

LICR development of cancer vaccines is focused around three antigen systems; the melanocyte differentiation antigen Melan-A/MART-1, and the cancer/testis (CT) antigens MAGE-3 and NY-ESO-1. To determine the safest and most efficacious cancer vaccines containing these antigens, LICR is comparing vaccines containing different forms of antigen; primarily peptide (short protein fragments), protein, viral vectors, and ‘naked’ DNA. The various antigenic forms provide the means to compare vaccines that incorporate additional components that increase immune responses (adjuvants and immunostimulatory compounds), or alternative methods of vaccine delivery (direct injection, viral delivery, particle-mediated epidermal delivery).

Peptides

Studies conducted by LICR investigators and others indicate that vaccine-induced CD8+ T cells can be induced using peptide-based vaccines. These peptide-based vaccines use antigenic sequences previously identified as the minimal CD8+ or CD4+ T cells ‘epitopes’ (short protein sequences recognized by the T cells) (Cancer Antigen Characterization). These are combined with adjuvants or immunostimulatory compounds, such as granulocyte macrophage-colony stimulating factor (GM-CSF, a cytokine discovered by LICR investigators), the drug Montanide, or CpG (a specific nucleotide sequence), that stimulate the immune response to an antigen.

Since the technology to manufacture peptides is relatively simple, a number of different antigens may be tested reasonably quickly and inexpensively to confirm an antigen’s immunogenicity. However, T cell responses to such vaccines are limited to the selected peptides and are restricted to patients that have a defined histocompatibility leukocyte antigen (HLA) genotype; these genes encode the major histocompatibility complex (MHC) molecules, which present particular antigenic peptides for recognition by T cells. Peptide-based vaccines are therefore limited in their application, as they will only be efficacious for patients that have the appropriate MHC molecules required for presenting that particular peptide. Additionally, this approach requires the extra step of HLA typing before vaccination. LICR is designing vaccines that combine peptides for a number of frequently expressed HLA’s to induce broad immune responses applicable to most cancer patients with tumors expressing the appropriate antigen(s).

Proteins

LICR is also testing cancer vaccines based upon the antigen’s complete protein. The protein antigens are being combined with adjuvants or immunostimulatory compounds, such as saponin-formulated adjuvants, ISCOM™ and QS21, cholesteryl hydrophobized polysaccharide (CHP), Montanide, and CpG. Since a protein contains the complete repertoire of antigenic peptides, it is predicted that the vaccine will be applicable to patients of any HLA type.

An early-phase clinical study, conducted by LICR, with NY-ESO-1 protein complexed with ISCOM™ has demonstrated the desired induction of an integrated (antibody plus CD8+ and CD4+ T cell) immune response to this vaccine. Now that the vaccine has been confirmed as safe for use in humans, further clinical trials will be conducted to determine clinical efficacy and to optimize immunization with this vaccine.

Viral Vectors

Vaccine compositions that utilize microbial (bacteria, virus, or yeast) nucleic acid sequences (vectors) to package and transport the human antigen nucleic acid sequence are also being studied. An advantage of using viral delivery is the theory that the presence of the (harmless) virus acts as a natural adjuvant; the immune response against the virus enhances the immune response against the antigen that the virus is delivering.

Preliminary clinical evaluation of recombinant viral (human sequence plus viral sequence) vectors that encode the NY-ESO-1 or MAGE antigens indicate that specific T cell immunity can be induced following vaccination with these vectors. Preclinical development of additional microbial vectors, particularly adenovirus, salmonella and listeria (bacteria), and yeast, for the delivery of LICR antigens is ongoing.

Plasmid DNA

The fourth antigen composition being trialed by LICR is plasmid DNA (small, circular fragments of DNA) containing antigen sequences. One method of plasmid DNA-based vaccination is particle mediated epidermal delivery, in which minute gold particles coated with the DNA are ‘shot’ (painlessly) into the skin using pressurized air. This mode of delivery, using the NY-ESO-1 antigen, is in preclinical development pending final approval for clinical investigation.

Multi-Antigen Vaccines

LICR is developing multi-antigen cancer vaccines to increase vaccine efficacy, and ensure that the vaccine can be of use for the greatest proportion of patients suffering from a particular cancer. The expression of antigens can be heterogeneous, meaning that not all cells in the tumor will express the antigen. Consider a scenario in which Individual A has only 20% expression of Antigen I (i.e. only 20% of the cells in the tumor express Antigen I) and 60% expression of Antigen II, whilst Individual B has 80% expression of Antigen I and 5% expression of Antigen II. Delivering a vaccine for Antigen I will induce an immune response against 80% of Individual B's tumor cells, but only 20% of Individual A's tumor cells. Delivering a vaccine for Antigen II will induce an immune response against 60% of Individual A's tumor cells, but will barely help Individual B at all. Delivering both antigens in a single vaccination will ensure that the tumors in both Individual A and B are being targeted, and potentially more of the cells in each tumor may be targeted via the multi-pronged attack against both Antigens I and II.

Multi-antigen vaccines are also expected to reduce the likelihood of cancer cells escaping recognition by the immune system. There is a possibility that the tumor might respond to immune recognition by decreasing antigen expression through the process of immunoediting (Cancer Immunosurveillance). Additionally, tumor cells that are destroyed by the immune system may simply be replaced by other cells that do not express the antigen that is being targeted. This theoretical evasion of immune recognition by tumors may be overcome by the use of multi-antigen vaccines, as it is less likely that the cancer cells will be able to down-regulate the expression of multiple antigens.

Key Publications

Centers Involved in this Research