Cancer Vaccine Program:

A required step in the design of an immunotherapeutic vaccine for treatment of metastic cancer is to identify one or more molecules and their peptide fragments expressed on the surface of tumor cells. These tumor associated molecules and peptide fragments can then be used as immunogens to induce systemic immune responses, involving the activation and expansion of specific cytotoxic T lymphocytes ("CTLs") that recognize these tumor antigen "epitopes" and as a consequence, eliminate tumor cells spread throughout the body. Identification of these tumor antigens using conventional technologies is difficult, time-consuming, and expensive. To be effective as a potential immunogen in a cancer vaccine, the tumor antigen peptide must contain an appropriate sequence of amino acids that provide for it to be bound by peptide receptors, molecules of the major histocompatibility complex ("MHC"), present on antigen presenting cells of the immune system. These peptide/MHC complexes must then be capable of interacting with immune receptors on precursor CTL that cause their activation and differentiation into effective killer T cells. Thus ideal candidate tumor peptide epitopes must interact with MHC molecules and T cell receptor molecules to be immunogenic. Many strategies have been used to identify unknown CTL peptide epitopes.

• If the sequence of the native protein is known, then overlapping 8- to 10-mer peptide sequences can be synthesized and screened as potential CTL targets. This has been an effective approach in identifying epitopes from proteins of viral origin.
• For unknown tumor antigens, isolation of tumor-associated peptides from MHC class I molecules followed by purification and Edman sequencing or mass spectrometry has led to identification of a few CTL epitopes.
• Genomic approaches involving the screening of DNA expression libraries provides another alternative; this has led to the discovery of several human and mouse tumor antigens recognized by specific CTL.

But, because of the complexity and low abundance of these epitopes expressed on tumor cells, these procedures have proven to be difficult and time-consuming, often requiring large amounts of biological material for extraction and identification of natural epitopes. Especially in situations involving human cancers, the amount of material available for analysis is limited. Only occasionally is enough material available from a biopsy.

Mixtures Sciences' approach bypasses this difficulty by screening Positional Scanning Libraries in order to identify individual peptide sequences that will trigger activation of cancer-specific killer T cells in humans. Mixture Sciences' peptide combinatorial libraries contain virtually all of the possible MHC-presenting peptides, therefore enabling a rapid and complete search. These scans result in the identifiction of peptide sequences that have the combined ability to bind to MHC molecules and to interact with immune receptors on T lymphocytes. Sequences that mimic naturally occurring tumor antigen peptides will be identified by their ability to stimulate proliferation, cytokine production and release, and cytolytic activity by CTL clones having lytic specificity for selected tumor systems, but having no such lytic activity against normal cells. Mixture Sciences uses these mimics for tumor antigen peptides to construct tumor-specific vaccines. In model systems these mimics are orders of magnitude more effective than the natural epitopes in stimulating T cell responses. In preparation for human trials, these peptides will then be tested in animal models as vaccines in various constructs to determine their ability to produce protective and therapeutic immune responses against metastatic disease.

Mixture Sciences has the ability to conduct a comprehensive search through trillions of compounds with a speed and effectiveness that we believe has previously not been possible, and which we believe no competitor can achieve.