Self-Recognition and Tumor Response to Immunotherapy

Abstract

Cancer cells avoid immune destruction by downregulating surface molecules, such as major histocompatibility complex molecules, costimulatory molecules, and tumorspecific antigens that the immune system recognizes as foreign, and by inducing dysfunction or destruction of immune effector cells—T lymphocytes, natural-killer cells, and dendritic cells. Effector cell function can also be impeded by self-regulatory pathways. Effector cell regulation is a dynamic interplay between constitutively expressed tolerance (inhibition), which maintains the immune system in an “off” position normally, and proinflammatory signals (stimulation) induced by exposure to pathogens, including cancer cells. A primary goal of immunotherapy is to upregulate the proinflammatory response and downregulate immune tolerance to cancer cells. Initiation of an inflammatory response to hostile pathogens is generated by natural-killer cells, neutrophils, monocytes, dendritic cells (innate immune system), and other components of the innate immune system through recognition of shared classes of danger signals called pathogen-associated molecular patterns. Activation of the innate immune system enhances subsequent recognition of pathogen-specific antigens by the adaptive or acquired immune system (T cells). Proinflammatory cytokines or tumor vaccines can mobilize immune effector cells that recognize and destroy cancer, occasionally producing dramatic clinical results, but only in a minority of cancer patients. The absence of benefit in the majority of treated patients may be due, in part, to immune inhibition, involving tolerogenic pathways. Central tolerance, the body’s ability to eliminate self-reactive T cells, is not an all-or-nothing process. The surviving selfreactive T cells have low avidity for self-antigens and, under the right conditions, can become activated, causing autoimmunity. Peripheral tolerance, a network of inhibitory signals and cells, plays a dominant role in downregulating immune response to foreign pathogens, and is primarily responsible for tumor-specific immune inhibition in cancer patients. In particular, peripheral tolerance inhibits helper T cells (CD4 ) and CD8 cytolytic T lymphocytes (CTLs), which are specific to tumor antigens and can be effective in tumor destruction. Thus, immune approaches involving CTLs must block peripheral tolerance to be successful. Peripheral tolerance is mediated through a number of signals including the CTL antigen-4 (CTLA-4) molecule, which competes with CD28 (a T-cell stimulatory receptor) for a mutual ligand on antigen-presenting cells (B7.1/B7.2). CTLA-4 is also constitutively expressed on a subset of CD4 cells (T regulatory cells [TREG] CD4 CD25 ) involved in peripheral tolerance. When CTLA-4 is blocked, CTLs are released from inhibition and the regulatory effects of TREG cells are diminished. Although this creates a permissive environment for tumor-specific CTLs, it also releases the naturally occurring inhibition of self-specific T-cells, thus initiating autoimmunity. In the study by Attia et al, patients with metastatic melanoma were vaccinated with two melanoma peptides [gp100:209-217(210M) and gp100:280 to 288(288V)] and subsequently given one of two different dosages of antagonistic anti–CTLA-4 antibody (MDX-010), resulting in two complete remissions and five partial responses in 56 treated patients. Consistent with the notion that blocking CTLA-4 inhibits peripheral tolerance, severe autoimmune-like toxicity was seen in five of seven responding patients, whereas five nonresponding patients had minimal or mild autoimmune toxicity. Studies using stimulatory immune treatments, such as interleukin-2, have also shown an association between autoimmune-like toxicity and clinical outcome, suggesting the involvement of alternate pathways. In addition, the broader, nonspecific, inflammatory response observed in the study by Attia et al is compatible with the possibility that blocking CTLA-4 stimulates the innate immune system, with subsequent activation of adaptive tumor-specific immunity. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 25 SEPTEMBER 1 2005