Factors Affecting Tumor Immunogenicity

FACTORS AFFECTING TUMOR IMMUNOGENICITY

Tumor cells present antigenic epitopes bound to MHC molecules expressed on their cell surface (signal 1), but they do not express co-stimulatory molecules (signal 2), a preroga-tive of APC. It is now well established that expression of signal 1 in the absence of signal 2 may lead to tolerance and anergy of potential effector cells; therefore, one would expect that the size of the tumor burden may correlate with the degree of immunosuppression ob-served in a cancer patient. Clinical experience has confirmed this supposition, leading to the general consensus that immunotherapeutic modalities may be more appropriate for pa-tients exhibiting a small tumor burden, as well as for patients in remission (e.g., following curative surgery) in whom immunotherapy is administered as an adjuvant to prevent re-currence of the cancer.

It is also well established that tumor cells release a number of factors/cytokines that can be immunosuppressive like TGF-β , IL-10, several prostaglandins, etc. Obviously, an increasing tumor burden would worsen the immunosuppression mediated by these factors.

It is generally assumed that tumor heterogeneity, both phenotypic and genotypic, is the result not only of intrinsic genomic instability associated with the cancer phenotype, but also the result of selective pressures that favor anaplastic variants with a more ag-gressive growth pattern. It is entirely plausible that some of these variants may result from immune selection: immune escape refers to the ability of progressing tumors to avoid recognition and killing by the host immune response. For instance, the level of ex-pression of MHC-I molecules by a large variety of tumors is generally lower than in the normal tissue counterpart. Since tumor-associated antigenic epitopes are recognized by CTL only when bound to the extracellular portion of MHC molecules, downregulated ex-pression of MHC-I would render recognition and killing of tumor cells by CTL more problematic, if not impossible. Consistent with this concept is the well-documented observation that, in many cases, sections of metastatic human tumors exhibit selective loss of individual HLA class I antigens (e.g., HLA-A2), which are prevalent in the general population, and for which it has been demonstrated the binding of a large num-ber of immunogenic peptides derived from tumor-associated antigens. Decreased im-munogenicity of human tumors may also result from the loss of antigenic expression it-self, because the gene products corresponding to most human tumor–associated antigens do not play any direct role in the development of the malignant phenotype. Antigen loss variants have been demonstrated in biopsy specimens from metastatic human tumors.

Immunosuppression of cancer patients can be specific for the autologous tumor or extend to other aspects of the immune response, as in the case of a decreased reactivity to standard viral or bacterial recall antigens. This more generalized type of immunosup-pression is often the consequence of treating patients with anticancer agents such as chemotherapeutic drugs or ionizing radiation, since neutrophils and lymphocytes are highly susceptible to these agents. These are important considerations when trying to as-sess the potential efficacy of immunotherapeutic regimens administered to patients who have previously undergone extensive chemotherapy or radiation therapy. Another nega-tive factor to consider is age, which is also associated with a generalized decline of some immune responses. Since cancer most often strikes older people, a weaker immune re-sponse may make them somewhat more vulnerable to the disease and less likely to ben-efit from immunotherapy.