Open Forum on Cell Signaling
Presidential Address: The American Association of Immunologists
30 April 2008
Nancy R. Gough
The Presidential Address was given by Olivera J. Finn from the University of Pittsburgh and was entitled "Immunologic Weapons Acquired Early in Life Win Battles with Cancer Late in Life".
Dr. Finn is an excellent and lively speaker. She started her career in Yugoslavia and at 16 addressed the annual meeting of the communist youth organization of Yugoslavia (now Serbia). She was then--and still is--self-described as revolutionary, controversial, and provocative.
Her lab developed a technique for generating specific T cells from kidney allograft needle biopsies and applied this technique in the late 80's and early 90's to assay T cells that recognize biopsies of tumors and to identify human tumor antigens. They expected to find tumor-specific antigens (mutated genes or viral oncogenic proteins), but instead they found "self" antigens, such as oncofetal proteins, lineage-specific antigens, and overexpressed transformation-related proteins that were not mutated at the peptide level.
Why were these recognized? She used MUC1, which is present on epithelial cells and lines the luminal surface, as an example. Antibodies against normal nontumor MUC1 did not recognize MUC1 from tumors, whereas MUC1 overexpressed in tumor tissue was recognized by a tumor-specific antibody. So, she argues that the tumor-specific antigens are not really "self" and should be renamed from "self/tumor antigens" to "abnormal self/tumor antigens". "Abnormal self" is antigenic; the immune response against "abnormal self" is important for tumor immunosurveillance; she proposes that "abnormal self" represents something that has been encountered prior to tumorigenesis, such as during infection or inflammatory events. Thus, she proposes that the immune system carries a "memory" of these "abnormal self" antigens, which is activated later in response to their reappearance in cancerous cells.
Going back to MUC1 as an example, MUC1 in tumor cells is hypoglycosylated and much more abundant than in normal cells. It is present in 80% of human tumors and is also present on premalignant lesions and putative tumor stem cells (Engelmann et al. 2008). In vitro, the addition of abnormal MUC1 primes CD8+ and CD4+ T cells; in animal models, there is a cellular and humoral immune response and tumor rejection, but no autoimmunity against "self" MUC1. A cancer vaccine based on abnormal MUC1 was effective in mice and has been used in clinical trials in humans in advanced cancer setting (metastatic prostate cancer) and is beginning to be tested for prophylactic use (resected adenomas).
At present, treatments based on antibodies against endogenous antigens can only be applied in the most extreme setting, because of the "self" classification, which generates concerns about autoimmunity. Thus, she argues that this concept of "abnormal self" is very important.
In collaboration with an epidemiologist, her lab found that certain life events that reduce the risk of ovarian cancer promote MUC1 immunity (Terry et al. 2007). These events all could lead to the presentation of abnormal MUC1 prior to cancer formation. There is a good correlation between cancer risk, these life events, and the appearance of antibodies for abnormal MUC1. The chance of recurrence of breast cancer was also negatively correlated with presence of antibodies against abnormal MUC1.
Cyclin B1 is another example of a protein that may be overexpressed in tumors and somehow "abnormal". Cyclin B1 from tumors is antigenic in vitro and in animal models (Kao et al. 2001), eliciting rejection of transplantable or spontaneous tumors. Anti-cyclinB1 IgG is common in the population. Healthy people, both those at high risk for cancer and those not at high risk, have antibodies to cyclin B1. There is immune memory for cyclin B1, which may arise from the induction of "abnormal" cyclin B1 during viral infection. What is unknown is if there a correlation between the abundance of the anti-cyclin B response and tumor protection.
In her model, viral infection, bacterial infection, inflammation, and tumors all have many nonoverlapping antigens, but a subset are common to all (imagine a Venn diagram of the antigens associated with each condition) and these will be the tumor antigens that can participate in immune surveillance of cancer. Identifying this subset may allow the development of "cancer vaccines". She also described how vaccination against viral infections may increase susceptibility to cancer by preventing the immune memory of abnormal self. When mice were vaccinated with a Pox vaccine, then challenged with a tumor 60 days later, the mice succumbed to the tumor rapidly. In contrast, if mice were vaccinated and then infected with the virus to allow them to respond to the viral infection, which is now not lethal, and then challenged with a tumor 60 days later, they either succumb much more slowly to the tumor or are resistant to the tumor (unpublished).
Her model of immune memory raises an interesting health issue. If immune memory for abnormal self-antigens is part of the cancer immune surveillance armamentarium, but these abnormal self-antigens are not occurring naturally and so the immune memory is not acquired naturally, then is it possible to vaccinate to prevent cancer in the future? (If you have been vaccinated against the major viruses, then you won't get a chance to develop these immune memories against abnormal self-antigens). If the overlap in the antigenic profile of infected cells and tumors can be determined, there may be an opportunity for a "universal" vaccine that would allow immune memories to be formed, which may provide a jumpstart to defending against pathogens, chronic disease, and cancer.
K. Engelmann, H. Shen, O. J. Finn, MCF7 side population cells with characteristics of cancer stem/progenitor cells express the tumor antigen MUC1. Cancer Res. 68, 2419-2426 (2008).
H. Kao, J. A. Marto, T. K. Hoffmann, J. Shabanowitz, S. D. Finkelstein, T. L. Whiteside, D. F. Hunt, O. J. Finn, Identification of cyclin B1 as a shared human epithelial tumor-associated antigen recognized by T cells. J. Exp. Med. 194, 1313-1323 (2001).
A. W. Silk, O. J. Finn, cancer vaccines: A promising cancer therapy against all odds. Future Oncol. 3, 299-306 (2007).
K. L. Terry, L. Titus-Ernstoff, J. R. McKolanis, W. R. Welch, O. J. Finn, D. W. Cramer, Incessant ovulation, mucin 1 immunity, and risk for ovarian cancer. Cancer Epidemiol. Biomarkers Rev. 16, 30-35 (2007).
Science Signaling. ISSN 1937-9145 (online), 1945-0877 (print). Pre-2008: Science's STKE. ISSN 1525-8882