Author: Ribas A, Butterfield LH, Glaspy JA, Economou JS
Source: Journal of Clinical Oncology, 21(12), 2003, pg 2415-2432
This article reviews the immunologic basis of clinical trials that test the means of tumor antigen presentation, recognition and activation. An antigen is a protein segment specific to a cell. Therefore, tumor antigens are contained on tumor cells. This should provide for the mechanistic understanding of ongoing cancer vaccine and cellular immunotherapy clinical trials.
The immune system recognizes other cells as foreign by recognizing protein fragments presented on the cell surface by molecules called major histocompatibility complexes (MHCs). T-lymphocytes ("T-cells") engage these protein-MHC complexes via their T-cell receptors. This allows the immune system to differentiate something foreign (by their foreign antigens) from "self", which should obviously not be targeted for attack.
There are two types of MHC molecules, aptly named MHC I and MHC II. MHC class I molecules are recognized by CD8 T-cells, which are the principal effector cells of the adaptive immune response. They mediate cytotoxic effects against those cells expressing these foreign antigens. Also, once a foreign antigen is recognized, there is a clonal expansion of T-cells specifically designed to attack and kill those cells expressing that specific antigen. MHC class II molecules are mainly expressed on the surface of so-called antigen presenting cells (APCs), the most important of which seems to be the dendritic cells. These APCs stimulate naïve T-cells, as well as other cells in the immune system. They stimulate both CD8 T-cells and CD4 T-cells, which help activate and maintain CD8 T-cell cytotoxic responses. APCs also stimulate natural killer (NK) cells, which normally serve as the first line of defense against foreign pathogens, cancer cells, etc. They can eliminate cells seemingly at will (hence the name), but their activity is tightly regulated. In a complex manner, the majority of signals they receive are actually negative, keeping them in check. This is obviously important, so that the NK cells do not attack native tissues, which would result in autoimmune diseases. However, they can be activated by the APCs to initiate immune responses when needed.
The majority of tumor cells, however, are ignored by the immune system. It was thought for a long time that tumor antigens did not exist. Recently, however, the number of known tumor antigens has increased rapidly, in many different cancers. It has also been discovered that only very short peptide sequences of the entire tumor antigen protein are immunogenic (able to elicit an immune response). These sequences are presented to the MHC molecules, just as a foreign protein from a virus or bacteria is presented. As more antigens are found, the number of targets that can be used to develop vaccines increase. Now, using gene analysis techniques, many more tumor antigens may be discovered, making these vaccine strategies much more broadly applicable in the years to come.
MHC presentation of antigens is described above, but what happens after the T-cell recognizes the MHC-antigen complex as foreign? Once the complex is recognized, the T-cell lyses the target cell. This is a very effective means of eliminating foreign pathogens. However, as with NK cells, there must be a tight control to ensure against attack of "self" tissues by strongly autoreactive T-cells. This happens by selective killing of T-cells during our development as well as some T-cells existing in an "ignorant" state by the lack of having a recognizable antigen. Therefore, the immune system has the ability to recognize self and nonself antigens and must be properly stimulated to be activated. This is very important for many reasons. As far as cancer is concerned, this is a difficulty that must be overcome in the development of cancer vaccines. Very few cancer antigens are truly foreign, but rather they are slight mutations of our own innate antigens. Therefore, the activation of an immune response to cancer cells must be a very delicate system.
Two signals, the antigen itself and a costimulatory molecule, are required for the initial activation of T-cells. These two signals generally will only happen if the antigen is presented by an APC, such as a dendritic cell. If only the antigen is presented, which is the case for most tumor cells, an immune response will not be activated because of the lack of the second signal. Therefore, it seems as if APCs are required to process the tumor antigen and provide the costimulatory molecule in order for an immune response against a tumor to be initiated.
A balance exists between the immune response to a tumor and the tolerance of tumor antigens. This exists in the recognition of truly foreign antigens, such as infection, but in a less tightly controlled manner. The balance is important, because uncontrolled expansion of T-cells after recognition of a tumor antigen could quickly overwhelm the entire immune system and unchecked cytokine (chemical signals from cell to cell) production and activity could easily lead to autoimmunity. This is achieved by many different mechanisms.
Dendritic cells are reported to have both a stimulatory and inhibitory effect on the immune system. The maturation status of the dendritic cell appears to be a major determinant of which this will be. Immature dendritic cells inhibit the immune response while mature dendritic cells activate it. There are many cytokines that function in roles to either stimulate or help control the immune response. They stimulate immune reactions by attracting and activating T-cells, NK cells, and other inflammatory reactants. They control the immune response by assisting in activation-induced cell death, which is a process where T-cells cause their own death when repeatedly exposed to the same antigen, and by assisting in recruiting regulatory T-cells which have a suppressor role on immune responses. Although these mechanisms are needed to guard against autoimmunity, they tend to mute the response to cancer cells.
Tumor cells have also developed mechanisms aimed at avoiding the immune system. These include the downregulation of tumor antigen expression and presentation, interfering with dendritic cell antigen presentation, direct inhibition of activated lymphocyte function, and resistance to T-cell attempts to induce programmed cell death. Therefore, the tumor microenvironment has developed it own means to protect itself from immune attack.
There are many targets for exploration that could use the knowledge gained to use immunotherapy in the fight against cancer. Intratumoral injection of bacillus Calmette-Guerin (BCG) may be the earliest forms of cellular immunotherapy tested in cancer. BCG generates an inflammatory process ideal for the attraction of APCs, which can then be used to present tumor antigens. BCG is currently used in the treatment of superficial bladder cancer. The administration of the antigen HLA-B7, which is actually an antigen found in a minority of humans, acts in much the same way.
Whole cell tumor vaccines have undergone decades of clinical investigation. They allow for the presentation of many tumor antigens to the immune system, and when administered with an immunologic adjuvant (BCG, other toxins, viruses), designed to activate the immune system, the hope is that it can be a useful immunotherapy. However, these vaccines are personalized, and take a significant amount of time to prepare. Also, there have been problems stimulating an active immune response. Gene-modified tumor vaccines, composed of whole cells transfected with genes encoding the costimulatory molecules could allow the vaccine to produce all of the signals required for T-cell activation, bypassing the need for APCs. However, these also take quite a bit of time, effort and money for their manufacturing.
Vaccines with just the tumor specific antigen have also been attempted. These can be synthetically manufactured and administered (again, with an immunologic adjuvant). Hence, their main advantage is their ease of manufacture and storage. However, these too have had problems initiating an effective immune response. Also, the tumor cells can escape immune attack simply by stopping the expression of the targeted antigen. In the same light, the DNA that codes for tumor antigens themselves have been tried as vaccines, as have viral vectors, which use viral machinery to produce the tumor antigens. These too suffer from an inability to generate an effective immune response. Administering DNA followed by viral vectors (called the "prime-boost" strategy) has resulted in more of an immune activation and this exciting method is being tested now in the clinic.
APCs have been the target of many vaccination strategies. These include using cytokines such as GM-CSF to increase the number of APCs. There have also been methods attempting to generate large numbers of dendritic cells from precursor cells. These can be specifically loaded with antigens to be a powerful immunologic adjuvant. These strategies have been tested in phase I and II trials, with promising results, though there is a continuing concern about the stimulation of autoimmune diseases.
Stimulating the immune system by the administration of cytokines or by producing cytokines by genetic recombination techniques has also been attempted. Though these can be effective, early clinical trials that have used them had to be closed early due to increased toxicity. Therefore, other, non-stimulatory cytokines have been administered. The rationale behind this method is that, if the regulatory pathway of the immune system can be blocked, there can be a more effective activation and maintenance of an immune reaction against tumor cells. Animal models have shown this to be true. Clinical trials are ongoing.
In summary, cancer immunotherapy attempts to shift the balance of the immune system toward ejection of cancer, using the many methods illustrated. In order for these to be successful and for more novel ideas to be devised, scientists and clinicians must make use of the various aspects of the immune system that have just recently been discovered and which this paper does a good job of describing.