First Potential "Universal" Cancer Vaccine Shows Promise In Lab

The scientists found that the active part of telomerase, a protein expressed in all major human cancers, can stimulate development of immune cells when used as the basis for a cancer vaccine. Furthermore, these immune cells can kill multiple, unrelated mouse and human cancer cells in the test tube and can also slow tumor growth in mice.

While not as effective as vaccines based on a tumor's entire genetic material, the promising results of these early studies indicate the potential of this telomerase-based vaccine as one component of a "universal" cancer vaccine, the researchers said in reporting their findings in the Sept. 1 issue of the journal Nature Medicine released Tuesday.

Cancer vaccines are being tested to help the body fight off certain existing cancers or to keep a cancer from returning, rather than to prevent the disease from developing initially. Like vaccines that prevent other diseases, cancer vaccines work by stimulating the immune system to build an army of cells that recognize, attack and kill the target cells -- in this case, cancer cells.

"The question posed in the study was whether the telomerase-based vaccine can stimulate an immune response from cancer patients, and whether those cells can attack and kill the patient's tumor cells," explained Eli Gilboa, principal investigator of the study and director of the Center for Genetic and Cellular Therapies at Duke. "The results of this study are the first indication that a more broadly applicable cancer vaccine might be possible."

Most known tumor-specific antigens -- proteins displayed on tumor cells but not normal cells -- are primarily associated with a single cancer type, such as prostate specific antigen (PSA). Even CEA, or carcinoembryonic antigen, which is found in 90 percent of colon cancers, 40 percent ofbreast cancers, and to lesser degrees in other cancers, can't match the broad expression of telomerase in human cancer, scientists said.

"The thinking has been that because every cancer is different -- melanoma, breast, etc. -- that each cancer has its own specific set of antigens that must be used for a vaccine," said Gilboa, who is also professor of experimental surgery and member of the Duke Comprehensive Cancer Center. "We're looking for a universal antigen -- one antigen to try to treat every cancer patient." Because telomerase is found in some types of normal cells, the scientists weren't sure an immune response against it would even be possible, Gilboa said. Telomerase rebuilds the repetitive ends of chromosomes, called telomeres, one step in many that allow cancer cells to divide unchecked. In normal cells that lack telomerase, the telomeres become shorter each time the cell's DNA is copied. Geron has many patents and pending applications related to telomerase, including its use as an antigen for cancer therapy.

"In spite of the fact that telomerase is a self-antigen, the body has not developed a complete tolerance to it and we were able to stimulate an immune response against it," Gilboa explained. "However, by itself, telomerase is not a strong antigen, so to make an effective, broadly applicable cancer vaccine we will need to optimize and possibly combine it with other universal antigens." Using techniques developed in Gilboa's lab, the researchers used RNA, the instructions for building proteins, to teach so-called dendritic cells to make a certain part of the telomerase protein, called TERT, and display it on their surface. Dendritic cells act as most-wanted posters for the immune system, showing immune cells how to recognize foreign cells and stimulating development of immune cells called cytotoxic T lymphocytes -- killer T cells -- that look specifically for cells matching the dendritic cells' display.

The scientists also made tumor-specific vaccines by using all of a tumor cell's RNA and compared its effectiveness to the TERT vaccine. While the tumor-RNA vaccines stimulated immune responses that more efficiently killed tumor cells, the telomerase-based vaccine stimulated an immune response that recognized and killed a much wider variety of cancer cells. The TERT vaccine stimulated an immune response that slowed tumor growth of melanoma, breast and bladder cancers implanted into genetically unrelated mice, and an effective response against two different mouse cancer cell lines in lab studies. No other vaccine was able to induce such a broad immune response.

Besides testing the TERT vaccine in mice, researchers led by Dr. Johannes Vieweg, a co-principal investigator of the study and assistant professor of urology at Duke, obtained cells from human cancer patients to see if an immune response against TERT could be developed. They used tumor cell RNA to create a tumor-specific vaccine from each patient, and each vaccine stimulated formation of tumor-specific killer T cells when incubated with blood cells from the patient, they reported. In addition, the TERT vaccine stimulated development of TERT-specific immune cells from each patient's cells.

Because it is difficult to obtain enough tumor cells from most patients to test immune cells' activity directly, the researchers used the tumor-RNA vaccine cells in place of tumor cells. Theoretically, the dendritic cells in the vaccine display all of the tumor's proteins, Vieweg noted.

In laboratory experiments, the tumor vaccine-stimulated immune cells and TERT-stimulated immune cells from each patient were able to recognize and kill the patient's tumor-RNA vaccine cells. Furthermore, for the one patient whose kidney cancer was established as a cell line, the immune cells were able to recognize and kill the patient's tumor cells in test tubes in addition to the patient's vaccine cells, validating the use of the vaccine as a surrogate for tumor cells. No longer is it necessary to have a large supply of tumor cells to verify immune activity, Vieweg said.

"By using dendritic cells transfected with tumor RNA -- the tumor-specific vaccine itself -- as a surrogate for human tumor cells in laboratory studies, we can expand the scope of antigen discovery and validation," Gilboa said. "It's not perfect, of course, because the vaccine cells may display the antigens differently than the cancer cell would, but it is a very useful screening method for identifying broadly expressed human cancer antigens needed for a universal cancer vaccine."

The study was funded by an anonymous gift to the Duke Comprehensive Cancer Center. Co-authors on the study are Smita Nair, Axel Heiser, David Boczkowski and Michio Naoe of Duke, and Anish Majumdar and Jane Lebkowski of Geron Corp., Menlo Park, Calif.