Promises and challenges of human papillomavirus vaccines for cervical cancer

Cervical cancer is the leading cause of death in women living in developing countries, mostly in Asia, South America, Africa and Europe. Owing to improved personal hygiene and regular Pap smear screening programs it is becoming less of a threat to women’s health in developed countries. However, it is still the second most common cancer in women worldwide. In terms of etiology, infection in high-risk groups of human papillomavirus (HPV) has been known to be a major cause of cervical cancer incidence.

The burden of this disease includes high medical, financial and psychological costs. Fortunately, present advances in the understanding of HPV biology and molecular technology have allowed us to develop prophylactic vaccines against HPV infection. For instance, Gardasil™ (Merck, NJ, USA) composed of L1-based virus-like particles (VLPs) of HPV6, 11, 16 and 18, and Cervarix™ (VLP of HPV16 and 18; GlaxoSmithKline, Hertfordshire, UK) have been licensed as a prophylactic vaccine. These vaccines are estimated to reduce the incidence of cervical cancer and its precancerous diseases, thus improving the wellbeing of women. While these prophylactic vaccines appear very promising, some issues still remain unclear, for example, the duration of protective immunity after vaccination, availability of the vaccines owing to their high costs in developing countries, the extent of crossprotection against infection with nonvaccine HPV types, who and when to vaccinate, ideal combination of HPV subtypes, how vaccination affects screening programs and safety issues.

It has been well known that less than 5% of women infected with HPV develop dysplasia, cervical intraepithelial neoplasia (CIN) I, II and III, and carcinoma in situ in sequence prior to cervical cancer. What determines the malignancy versus clearance of HPV infection is believed to mainly be the host’s immune status. Despite current successes in preventing HPV infection through the vaccination protocols, their therapeutic efficacy is thought to be lacking in women with persistent HPV infection. This is due to the nature of HPV biology. For example, HPV expresses its regulatory proteins but not L1 coat proteins in the basal epithelial cells. Under this circumstance, L1 coat protein-specific antibodies generated by prophylactic vaccination are irrelevant to control of viral replication. Among the viral regulatory proteins, E6 and E7 are expressed constantly in cervical cancer cells and are also required for tumorigenesis and maintenance of cervical cancer [1–4], therefore making them a potential therapeutic target for immunotherapy against cervical cancer and its precancerous diseases.

Cytotoxic T-lymphocytes as a positive factor for cervical cancer treatment

Concurrent chemoradiation therapy has been the current standard therapy in locally advanced cervical cancer. A large benefit of concurrent chemoradiation for survival, progression-free survival and local/distant control rates in cervical cancer patients has been suggested [5,6]. In some of these patients, however, recurrence of the disease has been problematic. Both tumor regression and tumor response to conventional therapy (i.e., radical surgery, chemotherapy and radiotherapy) seem to be greatly influenced by the host immune status. For example, E6- and E7-specific cytotoxic T-lymphocyte (CTL) responses were detected more commonly in HPV16-positive women without CIN, as opposed to HPV16-positive women with CIN [7]. CTLs also exerted a suppressive effect on development of squamous intraepithelial lesions in patients [8]. Furthermore, the magnitude of HPV E6- and E7-specific T-cell responses correlates inversely with recurrence of diseases in CIN patients following ablative or excisional treatment [9]. In addition, a high number of CD8⁺ tumor-infiltrating T cells were associated with a lack of lymph node metastasis in cervical cancer patients [10]. These clinical findings correlate well with those of numerous animal studies, demonstrating a major effector role of CTL in either protecting or removing tumors in the experimental model systems [11–19]. These animal and clinical studies collectively suggest that both induction and enhancement of E6- or E7-specific CTL responses, as well as an increase in CTL infiltration to tumor tissues, are critically important for obtaining a better response in patients with cervical cancer and its precancerous disease. Thus, immune therapy against cervical cancer should focus on how to augment tumor-specific CTL responses and increase CTL infiltration to cancer tissues.

Strategies for enhancing CTL responses

There have been numerous approaches to augmenting CTL responses to various vaccine types. In particular, an inverse correlation was observed between tumor size and the therapeutic efficacy of therapeutic vaccine-driven CTL responses in an HPV E6/E7-expressing tumor model [18], suggesting that the curability of tumors of a larger size needs more vigorous tumor antigen-specific CTL activity. In this regard, there are two major strategies, direct and indirect, to obtain better CTL activity for tumor regression. These have been tested mostly in an animal model system.

Direct enhancement of CTL responses

Novel strategies have been tested to directly enhance tumor antigen-specific CTL responses in an HPV E6/E7-expressing tumor animal model. The use of substances (such as CpG-oligodeoxynucleotides [ODN], Quil-A, liposome-polycation DNA particle, dsRNA [poly(I:C)]) as an adjuvant for HPV16 E7 protein and peptide vaccines is one example [12,20–23]. Conjugation of a lipid tail or helper T-cell epitope to E7 CTL peptides was also shown to improve an antigen-specific CTL response [24,25]. In this regard, potent helper T-cell functions are required for the high frequency and long duration of CTL activity in vivo[26]. Moreover, direct targeting of E7 antigens to dendritic cells by utilizing the E7/adenylate cyclase fusion proteins was shown to enhance anti-tumor CTL responses [27].

In E7 DNA vaccines, modification of intracellular targeting patterns of E7 antigens, as well as use of molecular adjuvants (encoding anti-apoptotic proteins and serine protease inhibitors) [15,17,28–33], is a typical example of modulating CTL responses. In the viral and bacterial vector delivery system, conjugation of E7 antigens to a costimulatory molecule CD40 ligand [14,34] or addition of E7 antigens with IL-12 [13,35,36] were both effective for antigen-specific CTL augmentation. In adoptive CTL therapy, infusion of cytokines is probably more effective for improving CTL activity in cervical cancer patients, This is based on the data of other tumor model studies showing that adoptive T-cell therapy was more successful with infusion with IL-2 or IL-15 [37,38]. These direct CTL-enhancing approaches are applicable for treating patients with cervical cancer. Presently, more research efforts focusing on direct augmentation of CTL responses are ongoing.

Indirect enhancement of CTL activity

Indirect strategies for enhancing CTL activity include the methods of increasing antigen-specific CTL activity by removing any negative factors responsible for suppression of CTL induction and maintenance. The feasibility of using antibodies specific for Treg cells (known to inhibit CTL responses) for enhancing E7 vaccine-induced CTL responses for anti-tumor resistance has been demonstrated in an HPV animal tumor model (Sin J-I, Unpublished Data). These data correlate well with that of other tumor model studies in which either depletion or blockade of CD4⁺CD25⁺ Treg cells positively influenced vaccine-specific CTL responses and anti-tumor resistance [39–43]. Similarly, the anti-tumor effectiveness of CTLA-4 (known to suppress CTL induction) blockade by delivering anti-CTLA-4 antibodies was demonstrated in melanoma cell vaccine models [42,43]. Thus, antibody treatments for either removing Tregs or blocking production of inhibitory signals for CTL activation appear promising for indirectly influencing the anti-tumor CTL responses of therapeutic vaccines.

Clinical aspects of HPV vaccines in cervical cancer & its precancerous disease

Precancerous cervical cancer lesions are divided into several stages, such as dysplasia, CIN I, II and III, and carcinoma in situ, depending on histologic observation (disturbed epithelial architecture and progressive cellular atypia). Therapeutic vaccines (E7 peptide and heat-shock protein-conjugated E7 protein vaccines, viral vector delivery of E2 proteins and E7-fused VLP) have been tested for their effectiveness in treating the precancerous lesions [44–47]. In these studies, the vaccines were successful for inducing antigen-specific CTL responses, as well as displaying objective clinical benefits, such as regression of HPV-associated precancerous lesions. These findings offer some optimism for the utility of therapeutic vaccines in clearing precancerous HPV lesions.

However, in cervical cancer patients, the efficacy of therapeutic vaccines was found to be lacking. It has been thought that 5-year survival rates decrease with the increasing cancer stages. In addition, primary tumor size, lymph node involvement and metastasis to other body parts are a limiting factor in treating cervical cancer. In these patients, therapeutic vaccines (e.g., peptide vaccines, viral vector delivery and dendritic cell vaccines) were well tolerated without any clinical benefits, even in the presence of HPV-specific CTL in some patients [48–52]. In particular, some of these patients were immune-compromised (e.g., a lack or loss of MHC expression in cancer cells and an induction of immune tolerance). Thus, downregulation of the host immune system is a definite negative factor for treating cervical cancer patients by therapeutic vaccination. Presently, treating patients with cervical cancer by immunotherapy alone appears less promising.

Implications of HPV vaccines for cervical cancer treatment

After clinical practice of radical surgery, chemotherapy and radiation, cervical cancer cells tend to be metastasized to other parts of the body and/or form resistant cancer cells against the previous practice. These are the main causes of tumor recurrence. Tumor recurrence might be prevented by the addition of therapeutic vaccines to conventional therapy protocols. In this regard, therapeutic vaccines can induce systemic tumor-specific immune surveillance, which can control the occurrence of any resistant and metastatic tumor cell clones. This possibility was tested in an animal HPV16 E6/E7-expressing tumor model system. For instance, a dramatic therapeutic synergy between a cytotoxic drug, cisplatin, and HPV16 E7 subunit vaccines was observed [19]. The combination strategy increased tumor cure and response rates, and decreased tumor recurrence through induction of long-term anti-tumor memory responses. A similar anti-tumor therapeutic effect was also observed when radiation and HPV16 E7 subunit vaccines were combined [53]. In these two studies, therapeutic synergy was mediated through increased tumor cell sensitivity of drug-treated or radiated tumor cells to CTL-mediated killing. In particular, both chemotherapy and radiation increased the expression of tumor cell surface markers, which have been known to be important for facilitating CTL-tumor cell contacts (Sin J-I, Unpublished Data; [53]). Chemotherapy and radiation also kill tumor cells directly through induction of apoptosis, allowing antigen-presenting cells to crosspresent and then generate tumor-specific CTL responses. This possibility was demonstrated in an HPV animal model [54]. Finally, chemotherapy and radiation probably make cancer tissues more inflammatory, thereby inducing a greater infiltration of lymphocytes to the tissues, as previously shown by our group [19,53]. Thus, it is reasonable to integrate therapeutic vaccines to conventional therapy modalities for a better cervical cancer intervention. However, it is still unclear how these animal study results translate to clinical cases. When used in combination with chemotherapeutic drugs (usually suppressing the host’s immune system), the timing and schedule of delivering therapeutic vaccines (as an active immune therapy) need to be carefully designed for adequate induction of an antigen-specific CTL response. However, in the case of adoptive transfer of CTLs, which are already activated in vitro, immune suppression by chemotherapeutic drugs might be beneficial, as the drugs tend to inhibit the host immune cells, including Tregs (known to counteract CTL activity). In this case, the fully activated and then adoptively transferred CTLs are unlikely to be affected by the drugs. This possibility was tested by our group using an HPV E6/E7-expressing tumor model [Sin J-I, Unpublished Data]. Therefore, a careful evaluation of the effects of drugs or radiation on the tumor, tumor microenvironment and the immune system is essential for obtaining optimal therapeutic benefits in cervical cancer patients.

In summary, presently available prophylactic vaccines are expected to reduce HPV infection rates dramatically, thus, protecting women from cervical cancer incidence. Therapeutic vaccines, however, present us with far more challenges than prophylactic vaccines. These challenges include the immune-compromised state of cervical cancer patients and the immune-evasive nature of cancer cells themselves. In spite of these challenges, there have been many promising preclinical data demonstrating an improved anti-tumor efficacy of therapeutic vaccines in combination with chemotherapy and radiotherapy. Although the ultimate outcome of these preclinical results in clinical situations is not yet known, there is no doubt that HPV vaccines can contribute to the eradication of cervical cancer and, eventually, the wellbeing of women worldwide.

Financial & competing interests disclosure

This work was supported by the Korea Research Foundation Grant (KRF-2004–041-E00094). The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.