Despite multiple approaches to therapy and prevention, cancer remains a major cause of death worldwide.
Most nonsurgical approaches targeting rapidly dividing cells, using radiotherapy or chemotherapy, also affect normal cells and result in side effects that limit treatment.
In principle, the exquisite specificity of the immune system could be marshaled to precisely target cancer cells without harming normal cells.
This hope has motivated much research over several decades but has met with only limited success to date.
However, the rapid increases in knowledge of the immune system and its regulation have led to a resurgence of interest in immunologic approaches to target and eliminate cancer.
A major difference between microbial pathogens and tumors as potential vaccine targets is that cancer cells are derived from the host, and most of their macromolecules are normal self-antigens present in normal cells.
To take advantage of the immune system's specificity, one must find antigens that clearly mark the cancer cells as different from host cells, limiting the number of antigens available.
Additionally, many potential tumor antigens are not expressed on the surface of tumor cells and thus are inaccessible to antibodies.
The immune system has evolved a solution to this problem: the MHC antigens (HLA molecules in humans) that act as an internal surveillance system to detect foreign or abnormal proteins made inside the cell (Figure 1).
The class I MHC antigen processing pathway acting as an internal surveillance mechanism to detect any abnormal or foreign protein synthesized in the cell.
Tumor antigens encoded in the endogenous DNA of the tumor cell, or encoded in a DNA plasmid or viral vector vaccine taken up by an Antigen Presenting Cell (APC), are synthesized and cleaved by the 26S proteasome into fragments that are transported by TAP, the transporter associated with antigen processing, into the endoplasmic reticulum, where they are loaded onto newly synthesized class I MHC molecules that transport them to the cell surface for recognition by the T cell receptor.
A sampling of all proteins synthesized in the cell is cleaved by proteasomes into short fragments (peptides) that are transported into the endoplasmic reticulum.
There, the peptides are loaded onto newly synthesized class I MHC molecules.
The peptide-MHC complexes are transported to the cell surface for recognition by the T cell receptors (TCRs) of CD8+ T lymphocytes, such as CTLs.
Thus, CTLs recognize short peptides, 8-10 amino acid residues in length, arising from the proteasomal degradation of intracellular proteins and able to bind to class I HLA molecules.
For this reason, CTLs are not limited to tumor antigens expressed intact on the cell surface but can detect any abnormal protein synthesized in the cell, greatly expanding the range of tumor antigens detectable by the immune system.
Furthermore, CTLs play an important role in the rejection of transplanted organs and tissues, analogous to tumors as foreign or abnormal human cells invading the host.
Thus, although monoclonal antibodies have clearly shown therapeutic efficacy in certain cancers (e.g., trastuzumab, rituximab, alemtuzumab), most cancer vaccine strategies have focused on induction of CTLs that lyse tumor cells.
Recent understanding of the mechanisms of activation and regulation of CD8+ T cells has given new life to tumor immunology.
Notwithstanding the critical role of CD8+ T cells, induction of tumor-specific CD4+ T cells is also important not only to help CD8+ responses, but also to mediate antitumor effector functions through induction of eosinophils and macrophages to produce superoxide and nitric oxide.
For naive CD8+ T lymphocytes to be activated initially, or "primed," they generally require presentation of antigens by professional Antigen Presenting Cells (APCs), such as Dendritic Cells (DCs).
DCs express high levels of costimulatory molecules, such as CD80 and CD86, which can make the difference between turning off the CTL precursor and activating it.
DCs also secrete critical cytokines such as IL-12 and IL-15 that contribute to CTL activation and memory.
In addition, a number of regulatory mechanisms that dampen the immune response are exploited by tumors to escape immunosurveillance.
These mechanisms include the inhibitory receptor CTLA-4 on the T cells themselves and negative regulatory cells such as the CD25+CD4+ regulatory T cell and certain types of CD4+ natural killer T (NKT) cells that inhibit tumor immunosurveillance.
Major hurdles in developing cancer vaccines include: identification of antigens that focus the exquisite specificity of the immune system on cancer cells without harming normal cells; development of methods to induce an immune response sufficient to eradicate the tumor, in the face of self-tolerance to many tumor antigens; and overcoming mechanisms by which tumors evade the host immune response.
An extensive listing of the known tumor-associated antigens is available, and more are being discovered.
Tumor antigens can be categorized into four groups: (a) antigens unique to an individual patient's tumor; (b) antigens common to a histological similar group of tumors; (c) tissue-differentiation antigens; and (d) ubiquitous antigens expressed by normal and malignant cells.
To specifically target tumors, antigens must be expressed only in tumor cells.
Cancer vaccine modalities
The vaccine strategies used against cancer depend on how well defined the target antigens are and whether there are conserved antigens that are shared among tumors of the same type in many individuals.
We will discuss the rationale for, and experience with, some of the most widely studied approaches (Table 1).
Modified tumor cell vaccines
The richest source of rejection antigens is the tumor itself.
However, use of autologous tumor cell vaccines is cumbersome and not amenable to large-scale vaccine production, and tumor samples are often unavailable.
Approaches using allogeneic or generic cell lines, as vaccines are more widely applicable.
Tumor cells engineered to secrete a number of different cytokines have been shown to protect mice from challenge with the same tumor type.
Of the cytokines studied, GM-CSF appeared most effective.
Local expression of GM-CSF increases DCs and other APCs at the site of injection of vaccination.
These acquire, process, and present antigen to T cells.
Elucidation of the crystal structure of the MHC and of the peptides bound to it and discovery of anchor-residue sequence motifs accounting for binding specificity of peptides to MHC molecules has provided the visual and mechanistic answer to how T cells recognize antigens in the form of short peptides.
The observation that short peptide segments (8-10 amino acids) fit into a groove in the MHC molecule, combined with knowledge of the amino acid sequences of tumor epitopes, prompted the use of peptides as therapeutic agents in the treatment of cancer.
Recombinant viral vectors
A number of trials utilizing recombinant viruses expressing tumor antigens such as carcinoembryonic antigen (CEA) or prostate-specific-antigen (PSA), some with immunostimulatory cytokines, have been reported or are in progress.
Adenovirus, vaccinia, and avipox vectors have been used.
Intramuscular injections of naked DNA expression plasmids have been shown to generate immune responses.
Such DNA vaccines introduce tumor antigen genes into DCs for endogenous processing and presentation to CTLs in draining lymph nodes or into other cells for cross-presentation by DCs, without the need for a viral vector.
Thus, problems of competition from viral vector epitopes, reduced efficacy due to prior immunity to the viral vector, and potential dangers associated with a live virus are avoided.
Dendritic cell vaccines
DCs are professional APCs and are the most powerful stimulators of naive T cells.
Immature DCs sample the antigenic environment through phagocytosis, micropinocytosis, receptor- and lectin-mediated endocytosis and are more effective at processing antigen.
When DCs encounter inflammatory mediators, they mature.
Helper T cells also induce DC maturation via CD40 ligand interaction with CD40.
As they mature, DCs downregulate their antigen uptake and processing machinery; express CD83; upregulate MHC, costimulatory molecules (CD80 and CD86), and the chemokine receptor CCR7; and travel to lymph nodes where they activate antigen-reactive T cells.
Demonstration of acquired defects in DC maturation and function in tumor-bearing animals and cancer patients suggests a rationale for using ex vivo-generated DCs as antitumor vaccines.
DCs pulsed with tumor lysates, tumor protein extracts, synthetic peptide tumor epitopes, or DCs fused with irradiated tumor cells could generate protective immunity to subsequent tumor challenge.
Transfer of nucleic acids encoding tumor antigens into DCs using plasmid transfection, retroviral vectors, recombinant adenoviruses, lentiviruses, or electroporation of tumor RNA has been effective.
Antigens can also be targeted to DCs by coupling to DC-specific antibodies.
Transfer of genes encoding costimulatory molecules (B7) and cytokines (IL-12) into DCs has also enhanced antitumor vaccine efficacy.
Antitumor vaccines in clinical trials
(BCG - bacilli Calmette-Guerin; KLH - keyhole limpet hemocyanin.)
|Whole tumor cell
||1. Studied extensively
2. Can be processed to enhance antigen presentation (e.g., irradiated tumor cells or tumor lysates);
3. Can be administered with adjuvants (e.g., BCG, KLH, viruses, etc.);
4. Likely to express the relevant tumor antigens;
5. Antigens need not be defined
|1. Requires availability of autologous tumor or an allogeneic cell line sharing the relevant tumor antigens;
2. Poor ability to stimulate immune responses;
3. Few responses and little benefit reported when used adjuvantly in randomized clinical trials
|Gene-modified tumor cells
||1. Likely to express the relevant tumor antigens;
2. Antigens need not be defined;
3. Often engineered to coexpress immunostimulatory molecules and cytokines (e.g., GM-CSF, IL-2);
4. Use of allogeneic tumor cell lines and fibroblasts are under investigation as an approach to accelerate vaccine production;
5. Some immunological and clinical responses reported
|1. Requires availability of autologous tumor or an allogeneic cell line expressing the relevant tumor antigens;
2. Weak antigen presentation by many tumors;
3. Long manufacturing time;
4. Need for ex vivo cell culture;
5. Cost, time, and labor intensive
|Plasmid (naked) DNA
||1. Constructed to express the relevant tumor antigen;
2. Easy to produce and stable;
3. Can be administered as a direct injection or biolistically ("gene gun")
|1. Requires detailed knowledge of the antigen DNA sequence;
2. Low immunological potency for self (tumor) antigens;
3. Response may be Th2 skewed;
4. High doses of plasmid DNA are required to generate immune responses
||1. Can limit immune response to epitopes distinct from the wild type (e.g., point mutations or breakpoint-fusion genes);
2. Epitopes can be enhanced;
3. Easy to produce and stable;
4. Can be combined as cocktails of peptides;
5. Some immunological and clinical responses reported
|1. Requires knowledge of the specific epitope;
2. Immunogenicity restricted to a limited number of MHC molecules;
3. Usually requires the addition of an adjuvant for immunogenicity
|Viral gene transfer vectors
||1. Engineered to express the relevant tumor antigen;
2. Can be engineered to coexpress immunostimulatory molecules and cytokines;
3. Wide variety of available vectors (e.g., adenovirus, pox viruses, lentiviruses, etc.);
4. Some cellular immune responses reported
|1. Immunodominance of viral antigens over tumor antigens;
2. Weak antitumor responses seen with most viral vectors;
3. Preexisting immunity against viral vectors may attenuate the antitumor response;
4. Risk of toxicity with "live" viruses
||1. Use of powerful APCs;
2. Techniques available to generate large numbers of clinical grade DCs;
3. Target antigens may be defined or uncharacterized;
4. Multiple antigen loading techniques (e.g., peptide, lysates, whole protein, RNA transfection, viral vectors, etc.) are available;
5. Some immunological and clinical responses reported
|1. Need for ex vivo cell culture;
2. Cost, time, and labor intensive;
3. Optimal technique for antigen loading remains undefined;
4. Possibility of tolerization by immature DCs;
5. Lack of criteria for standardization of final product
As prophylaxis against acute infectious diseases, vaccines have been among the most cost-effective agents, saving many millions of lives.
However, for treatment of chronic infections and cancer, vaccines have yet to achieve widespread success.
Increased understanding of the immune system has raised new hope of harnessing the exquisite specificity of the immune system to attack cancer and has led to novel second-generation vaccine approaches that hold promise to control or cure cancer.
The pace of identification of new tumor antigens has accelerated.
New strategies are being developed to make more potent vaccines against inherently weak tumor antigens, to selectively induce high avidity CTLs more effective at clearing tumors, and to overcome negative regulatory mechanisms that inhibit tumor immunosurveillance and immune responses to antitumor vaccines.
A number of promising new cancer vaccine strategies have entered clinical trials, and we eagerly await their findings.