The great escape: Is immune evasion required for tumor ... - Nature

© 1999 Nature America Inc. • http://medicine.nature.com

NEWS & VIEWS


2. Comstock, G.W., Livesay, V.T., and Woolpert, S.F. The prognosis of a positive tuberculin reaction in childhood and adolescence. Am. J. Epidemiol. 99, 131–138 (1974). 3. McKinney, J.D., Jacobs, W.R. & Bloom, B.R. in Emerging Infections (eds. Krause, R., Gallin, J.I. & Fauci, A.S.) 51–146 (Academic, New York, 1998). 4. Grigg, E.R.N. The arcana of tuberculosis, with a brief epidemiologic history of the disease in the USA. Am. Rev. Tuberc. Pulm. Dis. 78, 151–172; 426–453; 583–603 (1958). 5. Blower, S.M. et al. The intrinsic transmission dynamics of tuberculosis epidemics. Nature Med. 1, 815–821 (1995). 6. Tascon, R.E. et al. Vaccination against tuberculosis by DNA injection. Nature Med. 2, 888–892 (1996). 7. Huygen, K. et al. Immunogenicity and protective efficacy of a tuberculosis DNA vaccine. Nature

Med. 2, 893–898 (1996). 8. Lowrie, D.B. et al. Therapy of tuberculosis in mice by DNA vaccination. Nature 400, 269–271 (1999). 9. Rees, R.J.W. & Hart, P.D. Analysis of the host-parasite equilibrium in chronic murine tuberculosis by total and viable bacillary counts. Br. J. Exp. Pathol. 42, 83–88 (1961). 10. McCune, R.M. & Tompsett, R. Fate of Mycobacterium tuberculosis in mouse tissues as determined by the microbial enumeration technique. I. The persistence of drug-susceptible tubercle bacilli in the tissues despite prolonged antimicrobial therapy. II. The conversion of tuberculous infection to the latent state by the administration of pyrazinamide and a companion drug. J. Exp. Med. 104, 737–801 (1956). 11. Barclay, W.R., Ebert, R.H., Le Roy, G.V., Manthei, R.W. & Roth, L.J. Distribution and excretion of ra-

dioactive isoniazid in tuberculosis patients. J. Am. Med. Assn. 151, 1384–1388 (1953). 12. Edwards, W.M., Cox, R.S., Jr., Cooney, J.P. & Crone, R.I. Active pulmonary tuberculosis with cavitation of forty-one years’ duration. Am. Rev. Respir. Dis. 102, 448–455 (1970).

1

Harvard School of Public Health 667 Huntington Avenue Boston, MA 02115-6096 email: [email protected] 2 The Rockefeller University 1230 York Avenue, Box 21 New York, NY 10021-6399 email: [email protected]

The great escape: Is immune evasion required for tumor progression? A study on page 938 reports the identification of an ovarian and uterine tumor-associated ligand, RCAS1, which inhibits growth of activated T lymphocytes.

V

IRUSES HAVE DEVELOPED many well-studied mechanisms to ‘hijack’ their hosts’ metabolic pathways for their own benefit and to evade immune attack. Similarly, tumor cells are under selective pressure to manipulate their environment and to escape the growth-restricting mechanisms that act on the rest of the organism. For example, stromal cells can provide cancer cells with nutrients and soluble or membrane-anchored growth-promoting factors such as cytokines or integrins (Fig. 1). Tumor cells can also produce factors that modify the environment to their advantage, for ex-

Immune system

Cancer cells

Immune escape Clonal expansion V ascularization

Proliferation Cell death FasL

874

Tumor cell

Immune escape RCAS1 receptor

Growth arrest

ample, by promoting blood vessel growth (neovascularization) to meet their increasing demand for oxygen and nutrients. In normal conditions, tissue growth is limited by inhibitory signals from neighboring cells. Mutations in oncogenes or tumor suppressor genes, however, allow the tumor to overcome the normal growth constraints. Tumors develop independence from external growth signals and become refractory to cellular senescence and growth-inStromal cells hibitory or apoptosis-inducing stimuli. Certain types of

Integrins Growth factors Cytokines

Myeloid cells, NK cells, Activated lymphocytes

Activated T cell

ANDREAS VILLUNGER & ANDREAS STRASSER

Cytokines

RCAS1

CTLA4

Cytokine receptors B7-1/2 MHC I

CD28 TCR Fas

FasL DcR3

DF3/MUC1

cancer cells have also developed strategies to evade or counterattack the immune system (Fig. 1). Macrophages and granulocytes can kill cancer cells in an antigen- and major histocompatibility complex (MHC)-independent manner. In contrast, cytotoxic T lymphocytes (CTLs) must recognize antigen-derived peptides presented by MHC class I molecules with their T-cell receptor (TCR) to become activated, whereas natural killer (NK) cells are triggered by cells lacking MHC class I molecules. Mutation-induced resistance to immune attack may therefore contribute to neoplastic transformation. Reduced MHC class I expression or impaired antigen processing prevent recognition of tumor cells by CTLs, but may target them for NK cell attack1. Certain viruses overcome this

Fig. 1 Upper panel, complex three-way interactions between tumor cells, their microenvironment and the immune system. Cancer cells can receive nutrients and growth-stimulatory signals from neighboring stromal cells and stimulate neovascularization. Cells of both the innate and adaptive immune systems may attack cancer cells. Myeloid cells can attack tumor cells in an antigen- and MHC-independent way. Cytotoxic T lymphocytes need to be activated by antigen-derived peptides presented by MHC molecules, whereas NK cells are triggered by cells lacking MHC class I molecules. Lower panel, different mechanisms may lead to evasion or counter-attack of cancer cells towards the immune system: impaired antigen presentation with limited CTL activation; expression of decoy receptor DcR3 on cancer cells and neutralization of Fas ligand produced by CTL and NK cells; expression of Fas ligand may kill tumor-infiltrating CTL, NK cells, granulocytes or macrophages; and finally mucins, such as DF3/MUC1, or the new ligand RCAS1, described here, block T-cell proliferation, adding to the arsenal of weapons that tumors may use to evade control by the immune system. NATURE MEDICINE • VOLUME 5 • NUMBER 8 • AUGUST 1999



NEWS & VIEWS obstacle by expressing class I MHC surrogates that divert NK cells but do not present antigens to CTLs. Tumor cells have other mechanisms to avoid the multiple effector arms of the immune system (Fig 1). Some colon carcinomas overexpress DcR3, a decoy receptor that neutralizes Fas ligand, which is used as a weapon by CTLs and NK cells2. In other cancers (such as melanoma and hepatocellular carcinoma), tumor cells express Fas ligand on their surface and are thought to thereby kill infiltrating CTLs, NK cells and possibly also neutrophils and macrophages, which all express Fas (ref. 3). Mammary carcinomas can express mucins (such as DF3/MUC1), which inhibit growth of activated T cells4. Immunomodulatory substances, such as interferons, interleukin-10 or prostaglandins, produced either by tumor cells or their environment, may also affect the complex three-way interaction between cancer cells, their microenvironment and the immune system5. On page 938 of this issue, Nakashima et al. describe a growth-inhibitory molecule, RCAS1, which is expressed in many ovarian and uterine carcinomas6. RCAS1 is synthesized as a type II membrane protein that is thought to oligomerize through homotypic interaction of C-terminal coiled-coil structures. RCAS1 can also be secreted, and both secreted and transmembrane forms induce cell cycle arrest and apoptosis in RCAS1 receptor-expressing cells. Sensitive targets include T and NK cells, consistent with the idea that RCAS1 might be involved in tumor immune escape. Surprisingly, several tumor-derived cell lines expressed a 25kDa RCAS1 binding protein on their surface and were sensitive to RCAS1, suggesting that such tumors may have originated in an environment in which RCAS1 was not expressed. In RCAS1 receptor-positive cells, treatment with RCAS1 induced tyrosine phosphorylation of several intracellular proteins, but it is unclear how this relates to induction of cell cycle arrest and apoptosis. In a manner similar to DF3/MUC1 mucin, RCAS1 did not kill receptor-bearing quiescent T cells, but inhibited their activation by mitogens. Thus, induction of apoptosis is probably a consequence of cell cycle arrest rather than a direct effect of RCAS1 and DF3/MUC1. Consistent with this idea, cell death occurred late after RCAS1

treatment, and blocking apoptosis by inhibiting its central executioners, the caspases, enhanced survival of RCAS1stimulated K562 cells but did not prevent their growth arrest. In contrast, caspase inhibition during Fas ligand treatment generally results in sustained cell proliferation and enhanced viability. It is therefore unlikely that RACS1 functions through indirect activation of Fas or related ‘death receptors’. It is still not clear whether RCAS1 receptors promote growth inhibition after engagement of RCAS1 or RCAS1 prevents interaction between RCAS1 receptors and a growth-promoting ligand. What might be the physiological roles of RCAS1? It is unlikely that this gene has evolved for the purpose of tumor immune escape. The failure to detect RCAS1 protein in normal tissues may indicate that its expression is normally very low or is restricted to a rare cell type. The growth-inhibitory activity of RCAS1 could be involved in delineating tissue boundaries during embryogenesis. As activated T cells are RCAS1-sensitive, RCAS1 receptor signaling might also be involved in attenuating immune responses, as with CTLA-4 (ref. 7) (Fig 1). It will be interesting to see if spontaneous or experimentally induced mutations in RCAS1 or its receptor cause developmental abnormalities or defects in the immune system, such as autoimmunity or lymphadenopathy. How essential is immune evasion for tumor initiation and propagation? Are cancer cells recognized by the immune system as self or as a foreign intruder? It is probably fair to say that we do not know the answers to these questions. Cancer research has taught us that antigens that are strictly tumor-specific are rare, the exceptions being those created by mutations in oncogenes or tumor suppressor genes such as bcr-abl, ras or p53. It is not clear whether peptides from these onco-proteins could even be presented in a sufficiently immunogenic form by growing cancer cells to risk being deleted by immune attack8. Also, mutant mice lacking the adaptive immune system (such as rag-deficient mice) are not prone to tumor development. Immunocompromised humans have abnormally high incidence of cancers in which viral oncogenes contribute to cell transformation (such as Epstein-Barr virus-linked lymphomas and Kaposi’s sarcoma). However, in-

NATURE MEDICINE • VOLUME 5 • NUMBER 8 • AUGUST 1999

creased cancer risk in such cases may be a direct consequence of a failure to clear infections rather than defective immune surveillance of emerging tumors. It is also not clear whether expression of Fas ligand on tumors serves to attack the immune system or is used to destroy surrounding normal tissues to provide space for a growing cancer. The issue of a role for FasL expression by cancer cells in tumor immune evasion is further complicated by the finding that enforced expression of FasL on certain tumors causes their destruction by neutrophils9. Nevertheless, evidence for immunemediated regression of some cancers does exist and tumor immunotherapy has met with some success. Work on Her2/neu has shown that an antigen does not have to be absolutely restricted to tumors to represent a suitable target for immunotherapy10. Treatment with antibodies to RCAS1 to prevent destruction of activated T cells, in conjunction with cytotoxic drugs and perhaps also immunization to boost RCAS1-specific CTLs, may lead to promising results in the treatment of ovarian and uterine carcinomas. 1. Seliger, B., Maeurer, M.J. & Ferrone, S. TAP off— tumors on. Immunol. Today 18, 292–299 (1997). 2. Pitti, R.M, et al. Genomic amplification of a decoy receptor for Fas ligand in lung and colon cancer. Nature 396, 699–703 (1998). 3. Walker, P.R., Saas, P. & Dietrich, P.-Y. Tumor expression of Fas ligand (CD95L) and the consequences. Curr. Opin. Immunol. 10, 564–572 (1998). 4. Gimmi, C.D. et al. Breast cancer-associated antigen, DF3/MUC1, induces apoptosis of activated human T cells. Nature Med. 2, 1367–1370 (1996). 5. Levy, L.S. & Bost, K.L. Mechanisms that contribute to the development of lymphoid malignancies: roles for genetic alterations and cytokine production. Crit. Rev. Immunol. 16, 31–57 (1996). 6. Nakashima M., Sonoda K. & Watanabe K. Inhibition of cell growth and induction of apoptotic cell death by the human tumor-associated antigen, RCAS1. Nature Med. 5, 938–942 (1999). 7. Greenfield, E.A., Nguyen, K.A. & Kuchroo, V.K. CD28/B7 costimulation: a review. Crit. Rev. Immunol. 18, 389–418 (1998). 8. Van den Eynde, B.J. & van der Bruggen, P. T cell defined tumor antigens. Curr. Opin. Immunol. 9, 684–693 (1997). 9. Seino, K.-I., Kayagaki, N., Okumura, K. & Yagita, H. Antitumor effect of locally produced CD95 ligand. Nature Med. 3, 165–170 (1997). 10. Disis, M.L. & Cheever, M.A. HER-2/neu oncogenic protein: issues in vaccine development. Crit. Rev. Immunol. 18, 37–45 (1998).

Walter & Eliza Hall Institute for Medical Research P.O. Royal Melbourne Hospital Victoria 3050 Australia email: [email protected] 875