To be precise, the exact capacity of biochemist/immunologist Charles Nicolette's SPHERE-screening library amounts to 47,045,881 peptides. Here's how he reaches that complexity count.

"The combinatorial library has its 9-amino acid form. Every peptide in the library has three fixed residues. They all begin with phenylalanine, leucine, alanine amino acid residues. Now we have six residues in the middle that are randomized. It could be any one of 19 of the 20 natural amino acids occupying those positions. So if we do the math," Nicolette observed, "it essentially comes out as 19 to the sixth power (19⁶), which is 47 million and change."

He is associate director of antigen discovery at Genzyme Molecular Oncology (GMO), of Framingham, Mass., and sole inventor and patentee of GMO's proprietary SPHERE technology. "The acronym SPHERE,'" Nicolette explained, "stands for Solid-Phase Epitope Recovery.' It's solid phase because the peptides are synthesized on a solid-phase support, namely plastic beads. Epitope recovery means that at the end of the screen we actually walk away with one of 10,000 plastic beads that carries the active peptide - the antigenic determinant."

Nicolette is senior author of a paper in the Sept. 1, 2002, issue of the bimonthly Journal of Immunology. Its title: "Solid-phase epitope recovery: A high throughput method for antigen identification and epitope optimization."

"Our principal finding," he told BioWorld Today, "was that we could take a tumor-rejection antigen and make it more immunogenic for the purpose of vaccinating individuals. We did this by screening approximately 50 million different peptides, or protein fragments, and looked for the ones that gave the greatest immune response.

"What's unique about it," Nicolette continued, "is that the library is made on plastic beads, each containing a different peptide. It sounds like magic to be able to make 50 million of those, all different from one another, but the technique we've been using takes only about two weeks. We made them synthetically. They represent the structure of a minimal epitope that's recognized by a cytotoxic T cell."

T Lymphocytes Call The Shots

"T lymphocytes," he went on, "are the foot soldiers of the immune system. They are the cells that go out and actually kill the tumor, by lysing its cells. What they're recognizing on the tumor cells are short, 9-amino-acid fragments that are presented on the surface of the tumor cell. So our finding has been that when you try and vaccinate with a 9-amino-acid peptide that's derived from a normal non-mutated protein, your body doesn't really respond to it. That's why we're not all dying of autoimmune disease. You're tuned not to respond.

"We have developed the ability to take out of these 50 million peptides those that specifically mimic an epitope that occurs naturally, but is sufficiently different that the immune system responds by seeing it as foreign. We've done all of these immunization studies in vitro with human T cells," Nicolette recounted, "and we frequently find that the SPHERE-modified peptides will generate an antitumor response in nearly everybody that we test. So we thought this was a pretty big advance. Although peptides with these properties have been reported by other groups, there hasn't been a systematic way to rapidly generate these modified variants that have defined properties. So we feel that this approach can really allow us to attain a sustainable competitive advantage in making more immunogenic vaccines.

"One particular epitope," Nicolette pointed out, "is encoded by a melanoma antigen known as gp100 - a tumor protein frequently overexpressed in melanomas. In fact, many clinical melanoma trials have been designed using the gp100 protein or the entire gene in the recombinant virus, or even purified protein. None of those has really led to a high degree of success. Despite anecdotal success stories, it's impossible to predict whether one in 50 is going to respond dramatically.

"The experiments we report," he added, "indicated that by looking at human immune cells in vitro - the wild-type peptide that occurs in nature - we could not raise an immune response in more than a quarter of the people we tested. Yet 90 percent of those individuals would respond to the SPHERE-modified peptides.

"We are now in the process of moving forward into clinical trials," Nicolette said, "beginning with melanoma. We expect next year to be testing SPHERE peptide variants that mimic two melanoma antigens. They should encompass more than 80 percent of melanomas. We expect to pull the trigger on a trial by next year.

"As far as non-melanoma indications," he went on, "we're in the planning stages, trying to select from our catalogue of antigens and modified epitopes which ones make the most sense to go forward with. So we're doing a sort of market analysis, and assessing the need for medical interventions at various indications."

Planned Melanoma Trials Only For Openers

" We're very active in infectious diseases, too," Nicolette recounted, "applying this to HIV, with T-cell clones that are important for rejecting the AIDS virus, and clones that tend to be present in long-term nonresponders. We are applying SPHERE to those to generate more immunogenic variants of what they recognize in order to formulate a more potent vaccine. We have also done this for cytomegalovirus. So in infectious disease the same approach can be applied that we're taking to cancer.

"In terms of autoimmune disease," Nicolette continued, "the screen can be modified so that we're not looking for the most potent stimulators of the immune response but those that are the most potent inhibitors. For example, in multiple sclerosis certain antigens trigger an erroneous immune response, and that sets off onset of the disease. We may be able to find peptides that block the ability of those immune effector cells to induce pathology associated with MS.

"We have identified and optimized antigens spanning a number of indications," Nicolette noted, "including lung, colon, ovarian and breast cancers. We've been doing this for a couple of years now, and we have numerous patent applications filed on these malignancies. Right now we have enough antigens to move into almost any solid-tumor indication. I suspect moving forward we would probably focus on something like lung or colon cancer. But a decision," he concluded, "hasn't been made on that yet."