Reviewer: Ryan P. Smith, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 27, 2004
Author: Finn, OL
Source: Nature Reviews-Immunology, August 2003, pg 630
This is an excellent review article illustrating the current state of cancer vaccines and the hurdles that must be overcome if vaccines are to make their way into the clinic. Though written for an immunologist or someone familiar with the immune response and vaccine potential in detail, much of the information can be deciphered to gain a better understanding of the complexity of cancer vaccines and the obstacles that stand in the way of success.
Remember that vaccines, in their simplest terms, are designed to prepare the immune system for an encounter with either an infectious agent, or with tumor cells. The hope is that, with the immune system sensitized to the "non-host", it will attack and kill the bacteria, virus, etc, or cancer cell(s). However, all vaccines face certain challenges that are needed to overcome to elicit an effective immune response.
Just as successful vaccines to infectious agents have consisted of live but attenuated pathogens, the first cancer vaccines consisted of whole, inactivated tumor cells. These original vaccines, though, were found to be weakly immunogenic (eliciting a minor or no immune response). It was discovered that, in addition to receiving a signal of a foreign antigen (protein noted by the immune system), na?ve T-cells required additional co-stimulatory signals to elicit a vigorous response. Hence, new vaccines were designed that contained various co-stimulatory molecules or cytokines, which made them much more immunogenic. However, another problem with using whole cancer cells as the vaccine deals with autoimmunity. Though many of the antigens on the cancer cells are foreign, they also contain antigens that are similar to that of a patient's normal tissues. Therefore, cross reactivity could occur, where the immune system begins attacking the patient's normal cells as well. This undermines perhaps the most advantageous aspect of cancer vaccines-that of complete specificity. For these reasons, unique tumor antigens, which are products of random mutations or gene rearrangements (which in many cases are the exact carcinogenic signal) specific to tumor cells have been used as the antigens in vaccines.
Another challenge in a cancer, or any vaccine, is choosing the right "adjuvant", which is used to activate Antigen Presenting Cells (APCs). This assists in stimulating T cells, natural killer (NK) cells, or other cells of the immune system to produce cytokines, promote the survival of the antigen-specific T cells, and hence amplify the immune response. Currently, there are only two adjuvants available for clinical use. However, cytokines, such as interleukin-2, granulocyte-macrophage colony-stimulating factor, and several others have been tested as adjuvants in cancer vaccines. Bacterial products, which are obviously recognized as foreign by the immune system, have also been used as adjuvants. More recently, adjuvants have been developed to trigger a response in a desired processing pathway of the immune system. They either react to stimulate CD8 T-cells, causing T-cell dominated killing, or they stimulate CD4 T-cells, which leads to the production of antibodies. These adjuvants need continued development to advance the immune response induced by cancer vaccines.
Another issue is generating the correct immune response. Although metastatic cancer is a systemic disease, and hence expected to generate an immune response, most cancers start as mucosal diseases. The mucosal immune system has evolved to keep a balance between reacting to pathogens and ignoring more harmless foreign antigens, such as food and non-pathogenic bacterial flora that has colonized the mucosa. It seems as if the mucosal immune system ignores the cancer at first, and by the time it begins to recognize that it deserves attention, the tumor is too large to effectively control. None of the cancer vaccines developed thus far have been able to elicit mucosal immunity. Therefore, therapeutic (in contrast to prophylactic) vaccines are always fighting an uphill battle. By the time they are administered, the tumor bulk is so great that it is difficult to elicit a strong enough response to be efficacious.
Another obstacle in developing an effective vaccine is the difficulty in eliciting a long-term immune system "memory". This is especially true of prophylactic vaccines, such as those commonly administered to infants. The entire goal is to expose the immune system to an antigen from a pathogen (or in cancer vaccines, a carcinogen or mutated gene product) so that an immune response is not only generated but also remembered. This is required so that the next time the antigen is detected, a rapid immune response is again generated, keeping infection (or cancer growth) from occurring. The main problem that has deterred investigation into this field has been the difficulty in separating memory T-cells from other T cells. This continues to require advancement.
There are reasons that many of our vaccines against diseases are given in infancy or childhood. A major one is that our thymus is still functioning in childhood, with the ability to churn out a great number of na?ve T cells that can then react to antigens. Cancer vaccines have been tested in patients aged 65-80 years-decades after the thymus has stopped functioning. Therefore, the generation of a population of cells that respond to a vaccine is dependent on the already existing repertoire of T-cells. This age dependency has been proven in cancer vaccines in mice, as younger mice are able to generate a stronger immune response to vaccine than older mice. Hence, attention should be given to designing vaccines that can overcome the weakness of the aging immune response.
There is also evidence that the tumor itself induces immunosupression and evades the immune system. This is evident by the fact that tumor cells lose various antigens as they grow, indicating that the immune system had attempted to get rid of the tumor but failed. This is called cancer immunoediting. The ways in which tumors achieve this evasion have been described. The maturation and function of dendritic cells, perhaps the most important antigen presenting cells, are inhibited in cancer patients. Defects are also seen in T-cell activation and function. These effects can be mediated by interleukin-10, transforming growth factor- b, and other cytokines produced by tumors. There is also suppression of the activation of innate immunity by the tumor causing a state of oxidative stress, which is not completely understood. Devising treatment regimens to reverse this immunosupression before therapeutic vaccination is attempted may be key.
Cancer vaccines would be most efficacious in scenarios where they are designed to elicit or boost antitumor immunity in patients with minimal disease to attempt to prevent recurrence. However, in most studies completed thus far, the vaccines have been used in late stage disease with a large tumor burden. Hence, the success of the vaccines is dependent on the ability of the immune system to overcome imunosuppression induced by the many therapies the patient has had, age, and the tumor itself and then display efficacy against a large bulk of tumor. This is obviously a very difficult achievement.
The disease in which therapeutic vaccines have had the most testing is melanoma. Phase I and II trials in stage IV patients have shown a 10-20% response rate, with another 10-20% displaying stabilization of disease. In a phase III trial against a four-drug chemotherapy regimen, a vaccine showed similar response rates and survival with less toxicity. However, the key to therapeutic vaccines may come with the advent of those dendritic cell based vaccines that stimulate a vigorous immune response.
Many of the problems that diminish the efficacy of therapeutic vaccines would not need to be considered in the setting of cancer prevention. The idea of prophylactic cancer vaccines is akin to the classical way vaccines are thought of for the prevention of infectious diseases. An immune system that is primed to tumor antigens could destroy the tumor before it becomes clinically evident. The most successful testing of a prophylactic cancer vaccine is that against the human papillomavirus type 16 (HPV 16). HPV is a common infection in the general population, but some serotypes (HPV 16 and 18) are associated with cervical (and other) cancers. Therefore, a prophylactic vaccine against HPV 16 would attempt to protect against the infection that can lead to cancer. In a phase III trial, compared to control women, women who received the HPV 16 vaccine had zero infections with HPV 16. This would almost certainly translate into a huge decrease in cervical cancer. Another example of a prophylactic vaccine is the hepatitis B vaccine, which should guard against liver cancer.
One option is to continue to develop and test vaccines in cancer patients in small phase I and II studies. This should yield results for a limited number of patients at major medical centers in developed countries. Its effect on the global health problems as a whole, however, will likely be negligible. The other option is to shift to more of a preventative mode and develop more prophylactic vaccines. As discussed, therapeutic vaccines pose many problems in efficacy that prophylactic vaccines do not. Though this poses significant financial problems, this may prove to be more worthwhile in the long run.