Harnessing the Vitiligo Effect. The onset of one disease may play a role in treating another.

A mouse in which vitiligo has been induced as part of a recent NCI study. When the subjects of experimental melanoma vaccine studies lose pigment in their skin after receiving treatment, few of them seem to mind. They overlook the cosmetically undesirable side effect because it's an excellent prognostic sign. For about 20 years, cancer research immunologists have known that some patients with melanoma develop vitiligo, an autoimmune disease--or, as some investigators contend, a complex of diseases with immune and nonimmune biologic causes--that causes pigmentation loss in the skin. As researchers develop better immunotherapies for melanoma--often using peptides or cytokines such as interleukin-2 (IL-2) to recruit the immune system's help in fighting off the cancer--they've not only seen patients' melanoma subside but have seen an increase in the frequency of vitiligo-associated discolorations. The onset of vitiligo represents far more than a cosmetic nuisance or an inconsequential side effect. The autoimmune disease has, in fact, helped point researchers in the direction of so-called tissue-specific cancer vaccines that have greater specificity and fewer toxic side effects (unlike, for instance, the frequently used and somewhat effective IL-2, a nonspecific cytokine therapy that stimulates T-cell growth). Researchers discussed the latest progress of such prospects as part of an April 10 educational session, "Immunological Approaches to Cancer Treatment," at the American Association for Cancer Research (AACR) annual meeting in Philadelphia. The apparent association between melanoma and vitiligo suggested that there was some biologic event that attacked not only the cancer, but also the normal cell counterpart of the pigment cell melanocyte. In theory, if this attack is actually due to an immune response, and if that response could somehow be harnessed, new so-called tissue-specific cancer vaccines might be developed for use against melanoma and other cancers. Key to conceiving of such treatments has been the finding in recent years that antigens recognized by the immune system are largely "self" rather than "foreign" or tumor-specific. Several melanoma models have shown that the body requires 10-fold less immunity to acquire tumor immunity than it needs to acquire autoimmunity, suggesting that immunotherapies could achieve enough immunity to defend against tumors without conferring so much that autoimmune reinforcements kick in. Also crucial, explains Alan Houghton of the Memorial Sloan-Kettering Cancer Center in New York, was the demonstration by his lab and others that autoimmunity and tumor immunity mechanisms can differ. Uncoupling these mechanisms may, according to Houghton, eventually create strategies for triggering potent immunity against self-antigens that only attack the unwanted tumor. "The idea is that one is trying to break tolerance to tissue-specific antigens and induce essentially tissue-specific autoimmune disease to treat tumors," asserts Drew Pardoll, an oncologist at Johns Hopkins University. The first indication that such treatments might be feasible came from Houghton's lab about four years ago.1 They showed that if mice were given back antibodies against an antigen called tyrosinase related protein-one (TRP-1), or gp75, the animals would develop autoimmunity, depigment, and reject tumors. About a year later, researchers at the National Cancer Institute's surgery branch, with an eye toward treatment, attempted to get a better grasp of the vitiligo-melanoma relationship in humans. They noticed that many of those persons who responded well to an experimental IL-2 therapy also developed vitiligo. In a subsequent prospective study, investigators formally confirmed this observation, noting that all of the subjects who developed vitiligo responded well to the therapy, though some patients did respond favorably without getting vitiligo.2 So vitiligo was a good, but not a necessary, prognostic sign. "It appears that the only autoimmune manifestation of this problem is in the skin," remarks Nicholas Restifo, NCI senior investigator and leader of the NCI group. "And that's acceptable in most patients." Using mechanisms not yet well understood, the body prevents the loss of similar pigments in the inner ear, the eye, and the brain's substantia nigra, a phenomenon referred to as "immune privilege." To identify which antigens were recognized by which tumor-specific T cells, the NCI group probed a library composed of all the genes expressed by melanoma cells. They found that the antigens of interest were enzymes involved in the pathway that leads to the generation of melanin pigment--cells that are, in essence, the enzymes responsible for making the melanoma black. "It's kind of a wild coincidence," says Restifo. In follow-up studies reported in the March 16, 1999 issue of Proceedings of the National Academy of Sciences,3 the NCI investigators then asked which of the enzymes involved in melanin biosynthesis induced vitiligo in mice, and whether those same enzymes could induce tumor protection. They sought answers using a panel of recombinant vaccinia viruses that expressed the mouse versions of the melanosome-associated enzymes. Restifo's group was able to identify the specific antigen that induced vitiligo: TRP-1, the same antigen recognized by patients' T cells. TRP-1 both mediated the induction of vitiligo and conferred tumor protection. "This is the first demonstration that a genetically altered virus can induce these changes in experimental animals," notes Restifo, senior author of the PNAS paper. A group led by Houghton has achieved similar results using naked plasmid DNA to deliver human TRP-1 into mice.4 "The really scientifically interesting thing," Pardoll says of the Restifo study and a couple of others like it, "is that only one antigen--namely TRP-1--worked, which says that not all antigens are alike." Pardoll also notes that several studies including the Restifo group's, one of his own,5 and several who recently published results,6 have suggested that CD4, a CD marker on T-helper cells, rather than the oft-studied CD8, a CD marker on cytotoxic T cells, are the critical players in these immunotherapies. That is, immunity is intact with CD8 cells depleted, but it wanes with a decrease in CD4 cells. As Suzanne L. Topalian of NCI suggested at the April 10 AACR session, the field of cancer immunology is starting to accept that more than just cytotoxic T cells are important. Clinical trials, meanwhile, continue. "Both strategies, the vaccinia DNA and the naked plasma DNA ... need to be pursued very aggressively to try to get them into clinical trials," suggests Houghton. He adds that several labs are testing different types of antigens at different stages, ranging from mouse models, to early clinical trials, to final stage clinical trials where there's the potential for usable vaccine within a couple of years.7,8 Above and to the left are examples of the effects of vitiligo in melanoma patients treated with experimental IL-2 therapies. Findings like Restifo's, Houghton's, and Pardoll's are indicative of a growing trend in the area of cancer vaccines. Restifo's March 16 paper "represents a good example of something that is being found in a number of cases," says Pardoll, chairman of the April 10 AACR educational session, where Restifo also spoke. "If you can break tolerance with a particular tissue, that correlates to better antitumor immunity." In an upcoming commentary,9 Pardoll suggests that investigators are becoming more and more optimistic that inducing tissue-specific rather than tumor-specific autoimmunity could translate into effective cancer vaccines. And melanoma models could serve as springboards for immunotherapies in other cancers. "Melanoma's a wonderful model to look at this because it's so visual. If you attack the normal tissue, you depigment," Houghton remarks. "But you can imagine this in, for instance, breast cancer or prostate cancer, where the cancer arises from a tissue that is not necessary to remain alive." "Our own idea has been to specifically look at other tumors [in addition to] melanoma," says Pardoll. "Because if you can get an effect of a vaccine or an immunotherapy in a tumor other than melanoma, then you've really done something important." His group's approach has been to vaccinate with tumor cells transduced with the GM-CSF gene. They've done trials in kidney cancer, prostate cancer, and pancreatic cancer, and they plan to do trials in cervical cancer, colon cancer, leukemia lymphoma, and multiple myloma. "The way we've been going, [these therapies] haven't been that successful," Restifo says of the cancer immunotherapy field. "There's some excitement now that we might be able to actually dent this problem and the cancer problem in general." Pardoll cautions, though, that no immunotherapy, even in melanoma, has been proven to really work better than nonspecific therapies such as IL-2. Houghton declares, "Vaccines in cancer will be useful as part of treatment. What I can't tell you is when." I. Hara et al., "Implicating a role for immune recognition of self in tumor rejection--passive-immunization against the brown locus protein," Journal of Experimental Medicine, 182:1609-14, 1995. S.A. Rosenberg, D.E. White, "Vitiligo in patients with melanoma: normal tissue antigens can be targets for cancer immunotherapy," Journal of Immunotherapy, 19:81-4,1996. W.W. Overwijk et al., "Vaccination with a recombinant vaccinia virus encoding a "self" antigen induces autoimmune vitiligo and tumor cell destruction in mice: requirement for CD4+ T lymphocytes," Proceedings of the National Academy of Sciences (PNAS), 96:2982-7, March 16, 1999. L.W. Weber et al., "Tumor immunity and autoimmunity induced by immunization with homologous DNA," Journal of Clinical Investigation, 102:1258-64, Sept. 15, 1998. K. Hung, "The central role of CD4(+) T cells in the antitumor immune response," Journal of Experimental Medicine, 188:2357-68, Dec. 21, 1998. Journal of Experimental Medicine, 89:803-10, March 1, 1999. R. Lewis, "End of century marks dawn of clinical trial era for cancer vaccines," The Scientist, 9[7]:15, April 3, 1995. S. Bunk, "Human clinical trials begin for cervical cancer vaccines," The Scientist, 11[21]: 5, Oct. 27, 1997. D.M. Pardoll, "Inducing autoimmune disease to treat cancer," PNAS, 96. In press (May 11, 1999).

Harnessing the Vitiligo Effect. The onset of one disease may play a role in treating another.

Interested in reading more?

Become a Member of