Ryan P. Smith, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 24, 2004
Much improved understanding of tumor cell biology in the recent years has led to the discovery of tumor-associated antigens (protein molecules found specifically on tumor cells) that distinguish cancer cells form normal cells. However, the clinical efficacy of cancer vaccines has lagged behind the science due to various reasons. This article reviews strategies for inducing tumor immunogenicity and the efforts to translate results from the laboratory into the clinic.
Tumor cells accumulate DNA mutations during transformation into a cancer. These mutations leads to the production of altered antigens not found in normal tissues. These represent targets for selective recognition by the host immune system. These antigens are expressed with a molecule called major histocompatibility complex (MHC) for potential recognition by CD8 T-cells, which induce the immune response. In addition, antigen-presenting cells, such as dendritic cells, present antigens to CD4 helper cells, inducing a response including antibodies generated by the immune system. These dendritic cells are being found to be more and important in the immune response. T cell activation is also greatly enhanced by co-stimulatory molecules, which leads to the release of several cytokines. These, in turn, lead to further T- cell activation and the positive feedback system required for an aggressive immune response. Tumor-specific responses are muted, however, because cancer cells present antigen without the co-stimulatory molecules required for this immunity. Hence, potentially reactive T-cells are not induced, causing tolerance to antigens. In addition, tumor cells have been shown to express factors that inhibit antigen-presenting cell function and migration, also suppressing immunity. Several strategies are being pursued to enhance efficacy of cancer vaccines.
The premise of this strategy is that the introduction of a tumor antigen will be taken up by an antigen presenting cell and presented to T-cells. The resultant recognition of the foreign antigen by the T-cells causes an expansion of the population of these specific T-cells with the resultant immune response. Peptide vaccination has most commonly been studied in melanoma. Clinical trials have shown evidence of T-cell induction and response, though minimal. Hence, ongoing work in peptide vaccines is in improving the capacity to stimulate the immune response more and to understand the mechanism of immune suppression by the tumor. The concurrent administration of cytokines has been used as has altering one or more amino acids in the antigen to improve the immunogenicity of the antigen. These methods have both shown to enhance the response to peptide antigens.
Other problems also exist with peptide antigens. They depend on intact antigen presenting and the development of a population of T cells in order to be effective. If the antigens, which remember are just small proteins, are rapidly cleared, a response obviously will not be generated. Also, tumor cells can avoid immune recognition by simply downregulating the antigen that the immune system is responding toward. Especially if the antigen is not essential, this can be very easy for the tumor cells to accomplish.
Another approach in vaccine therapy is using genetic material, introduced into the antigen presenting cells as either naked DNA or as a viral vector, to encode the desired targeted tumor antigen. By introducing the DNA directly into the antigen presenting cells, the hope is that there would be a rapid and intense immune response. In preclinical investigations, using naked DNA for this purpose has elicited only weak responses. Although the concurrent introduction of cytokines (to enhance the immune response) has had some success, overall this method is not being used in the clinic. The other way DNA for gene therapy vaccines is introduced is through viral vectors. This is using a modified virus, with the desired DNA contained within its genome, to introduce the DNA. This approach is potentially limited by the generation of a host-immune response to the viral vector itself. This could lead to a rapid clearance of the vector without much introduction of antigen. Ongoing work is focusing mainly on introducing cytokines or other co-stimulatory molecules along with the DNA to attempt to enhance the immune response.
Another method of augmenting antigen presentation and immunogenicity is using a vaccine of modified tumor cells. This has the advantage of presenting the immune system with many antigens at once (since the entire cell could be foreign). However, this method has also had problems generating an aggressive immune response because of the lack of co-stimulatory molecules. Hence, a variety of cytokines and other co-stimulatory molecules have been concurrently administered. However, this has shown only modest benefits. Recently, a study investigated the use of a whole cell vaccine in conjunction with chemotherapy. This seemed to add to the efficacy of the vaccine, though chemotherapy could turn out to attenuate the immune response rather than assist it since chemotherapy, by its action, tends to decrease white blood cell populations.
As stated above, dendritic cells are an important antigen presenting cell that seems to be essential in the propagation of the immune response, through the many co-stimulatory molecules it releases. Dendritic cells are strong inducers of T-cell mediated immune responses. Therefore, another strategy in the development of cancer vaccines involves the manipulation of dendritic cells. Pulsing dendritic cells with tumor-derived peptides has been attempted to induce tumor-specific immunity. In animal models, vaccination with peptide-pulsed dendritic cells was protective against later challenges with tumors bearing that antigen. In clinical trials, vaccination with peptide-pulsed dendritic cells has been shown to induce antigen-specific immunity in patients with melanoma. Vaccination resulted in the generation of peptide-specific cellular immune responses in most patients.
Another strategy involves pulsing the dendritic cells with whole tumor proteins. Using this method in low-grade lymphoma patients, there were complete or partial remissions observed in several patients. Gene transfer methods using virus vectors or even retrovirus vectors have also been attempted. Again, there are limitations. With viral vectors, there is a risk of the dendritic cell expressing not only the desired antigen, but also the viral antigens, hence targeting itself for immune attack. With retroviral vectors, the retrovirus requires dividing cells for incorporation. Though this can be done with dendritic cells, the process is cumbersome and expensive. Another method involving dendritic cells includes creating hybrids of dendritic cells with tumor cells. This requires electrical or chemical fusion, though this remains a hopeful route of improving on cancer vaccines.
Though vaccine studies have demonstrated minimal toxicity, evidence of immunologic response and clinical responses in small patient populations, correlation between immunologic response and clinical response has not been consistently demonstrated. Different vaccine strategies exist, with little clarity as to which of these methods will prove to be most efficacious. Many of the methods have considerable technical challenges and thus will not be available in many centers. Also, most clinical trials using vaccines are being conducted in patients with advanced disease, although vaccines would likely be most efficacious in patients with low volume disease who have not had their immune systems compromised by multiple chemotherapy regimens. All of these hurdles must be overcome in order to bring the potential successes of cancer vaccines into the clinic.