NEW YORK -- Some 900 patients with vitiligo, being treated atHoward University College of Medicine in Washington, don't yetknow that their discomfiting but benign skin disorder may help todesign vaccines against melanoma.On Wednesday of this week, the chief surgeon of the NationalCancer Institute (NCI), Steven Rosenberg, described vitiligo --spreading patches of milk-white skin and snow-white hair onotherwise normal bodies -- as a dysfunction of melanocytes, theskin cells that make melanin, the body's pigment. He told the firstinternational symposium on cancer vaccines here that melanocytesare the only normal tissues in the body that express a novel tumor-antigen gene, MART-1, identified at NCI."The presence of this gene in both melanomas and melanocytes,"Rosenberg said, "explains a phenomenon that we've seen quitefrequently over the years: That is, vitiligo appears in patients whomwe treat with immunotherapy."Demonstrating its tumor specificity, he added, "this depigmentationof the skin occurred in 12 of 73 melanoma patients who respondedfavorably to the therapy, but in zero of 104 subjects treated for adifferent malignancy, renal cell carcinoma.Rosenberg told BioWorld Today that he has begun informalconversations with the chief of dermatology at Howard, RebatHalder. "We're going to study their vitiligo patients," he said, "tosee what they respond to, because it appears to be the same kind ofantigen, perhaps, that cancer patients are responding to, leading toregression of melanoma."Vitiligo affects about 2 percent of the population, Halder toldBioWorld Today. It affects all races equally, and he noted thatabout two-thirds of his cases are African-American. While thedisorder entails no physical disability, Halder added, "vitiligo isextremely devastating psychologically," because of its violation ofnormal skin color.Halder's own research dovetails to some extent with NCI's, in that,he said, "when patients with metastatic melanoma develop vitiligospontaneously, that indicates a good prognosis in their disease."The specific correlation between turning off of melanin-encodinggenes in vitiligo and in melanotic tumor cells, Rosenberg went on,"suggests the importance of this MART-1 antigen for therapy."MART stands for "Melanoma Antigen Recognized by T-Cells."That acronym encapsulates Rosenberg's "TIL-cell" cancer-vaccinestrategy, which NCI has been using and improving since 1987.Biotechnology Upgrades NCI's Vaccine FormulaT-lymphocytes, of which there are several subsets, command thecell-destroying division of the immune system. (B-leukocytesmuster the antibodies of the humoral defenses.)Leading the T-cells' search-and-obliterate operation againstinvading cells are the cytotoxic T lymphocytes -- CTLs. At thisweek's Cancer Vaccine 1994 symposium in New York (seeBioWorld Today, Oct. 5, p. 1), CTLs were the principal anti-tumoragents that most researchers are recruiting and training."This conference has been dedicated largely to the induction of Tcells against cancer," said Alan Houghton of the Memorial Sloan-Kettering Cancer Center in this city. "Certainly, this is the area ofgreatest interest, where the greatest advances have been made overthe last year."Before a CTL can infiltrate and kill a cancer cell, it must first seekout and overpower that cell's sentries _ the myriad antigenmolecules that stick out all over a tumor's surface. Each type oftumor custom-makes its own type-tailored antigens. MART-1 is themost immunodominant peptide that NCI has thus far discoveredagainst melanoma, Rosenberg told his cancer-vaccine audience.The killer T cells are designed -- a lot like antibodies -- torecognize and attack those tumor antigens.NCI's original clinical strategy was to remove tumor-infiltratinglymphocytes (TIL cells) from a patient's blood, and sensitize theseCTLs in culture, to home in on selected antigens. Then, togetherwith interleukin-2 as an adjuvant or booster, they would be returnedto the patient's bloodstream to hunt down the targeted cancer cells.This general approach has achieved regression of tumors in about35 percent of patients treated.As Rosenberg told BioWorld Today, "We don't understand whysome patients respond, and others do not. But by generating cellsthat are more potent than the natural TIL cells, we hope that thepool of patients that respond will be greater."Much of his hope is now pinned on wrapping these antigen-directedTIL cells in a package with recombinant viral vectors, such asfowlpox, vaccinia or adenovirus, carrying the gene sequencesencoding an antigen "that we know can cause cancer regression.""To identify the molecular structure of the antigens we haveidentified," Rosenberg observed, "is particularly exciting. For thefirst time we have genes in hand that -- we have evidence --encode antigens actually involved in tumor regression."He adds soberly, "Whether they will work or not, we don't know."To find out, NCI has approved a clinical trial of the approach,"which will begin very soon," focused on the MART-1 gene.A truly tumor-specific antigen (rather than merely "tumor-associated") remains an immunotherapeutic will-o'-the-wisp --though proteins such as MART-1 encodes may break the spell.Another such melanoma antigen gene is MAGE-1, isolated byThierry Boon of the Ludwig Institute for Cancer Research inBrussels. Boon told a press briefing at the end of the three-dayCancer Vaccine 1994 symposium, "Besides this first gene, MAGE-1, which I would like to call quote tumor-specific unquote, we havediscovered a number of additional genes, which are also tumor-specific. And it is my impression that there are 100 or 200 yet to bediscovered."Initially, Boon christened MAGE to stand for "Melanoma Antigen,"but as additional genes have come to light, he now defines MAGEas "My Antigen."Anent the vitiligo side effect of immunotherapy, Boon observed,"Losing a number of melanocytes is not too large a price to pay tocure a cancer that can be treated this way." n

-- David N. Leff Science Editor

(c) 1997 American Health Consultants. All rights reserved.