Infection of A549 cells with adenovirus 5 E6 and E7 (Ad E6/E7) oncoproteins results in functional protein expression. The E6/E7 transgene protein expression was tested after overnight transduction of A549 cells with Ad5 E6/E7 or control cells at a multiplicity of infection of 100. Western blot analysis was used to assess E7 protein expression and function. The top row shows production of E7 protein; the second row of proteins shows the E7-induced loss of Rb (retinoblastoma) (Jurkat cell lysate is shown as a positive control for Rb protein expression); and the third row of proteins shows E6-induced loss of p53.
Coculture of splenocytes with tumor cells demonstrates antigen-specific production of interferon-γ (IFN-γ). For each group, splenocytes were harvested from mice (n = 3 in each group) 70 days after implantation of E6/E7/H-ras tumors in tumor-clearing mice and when euthanized in tumor-bearing mice. Naive mice served as sex- and age-matched controls. To show a specific response to E6/E7, the splenocytes from tumor-clearing mice were cocultured with E6/E7/H-ras mouse tonsil epithelial cells (MTECs) or small hairpin PTPN13/H-ras MTECs for 36 hours, and supernatants were analyzed for IFN-γ production using an enzyme-linked immunosorbent assay. To compare responses between naive, tumor-bearing, and tumor-clearing (spontaneously or after vaccination) mice, the splenocytes from these groups were again incubated with E6/E7/H-ras. Statistical analysis (Mann-Whitney test) showed that tumor-clearing mice splenocytes specifically produced IFN-γ during coculture with human papillomavirus (HPV)-positive cells (P < .05). Similar statistically significant results show that tumor-bearing mice responded with more IFN-γ production than naive mice to the HPV-positive cells. Tumor-clearing mice responded more than tumor-bearing mice, and inoculated mice had the most vigorous response at all doses of adenovirus 5 E6 and E7 (Ad5 E6/E7) vaccine tested. Gray bars indicate HPV-positive (+) E6/E7/H-ras) cells; black bar, HPV-negative (−) (small hairpin PTPN13/H-ras) cells. *5 × 105 plaque-forming units (PFUs); †5 × 106 PFUs; ‡5 × 107 PFUs). All experiments were performed in duplicate.
Time course of antigen-specific immune response after vaccination with adenovirus 5 E6 and E7 (Ad5 E6/E7) oncoproteins. Interferon (IFN)-γ production from the splenocyte–tumor cell cocultures from mice inoculated with 5 × 107 plaque-forming units of Ad5 E6/E7. Splenocytes from these mice were harvested just before and 3, 15, 28, and 70 days after vaccination (n = 3 in each group). The splenocytes were cocultured with E6/E7/H-ras mouse tonsil epithelial cells followed by assessment of IFN-γ production. All experiments were performed in duplicate. The black bar indicates IFN-γ production from the splenocyte–tumor cell cocultures from mice that spontaneously cleared tumors as a positive control (n = 3). Mann-Whitney analysis was used to compare response after vaccination. On days 15, 28, and 70, all responses were significantly increased compared with prevaccination values. *P < .05.
Representative in vivo images (In Vivo Imaging System [IVIS]) in mice that cleared tumors either spontaneously or after vaccination compared with mice that allowed tumor growth. E6/E7/H-ras/Luc mouse tonsil epithelial cells (1 × 106) were implanted into the thighs of wild-type mice with or without inoculation with adenovirus 5 E6 and E7. IVIS images were obtained in live mice on the indicated days.
Adenovirus 5 E6 and E7 (Ad5 E6/E7) oncoprotein vaccine prevents tumor growth and enables 100% survival when tumor is implanted orthotopically. Wild-type mice were inoculated (n = 5 per group) with either Ad5 E6/E7 (5 × 105, 5 × 106, or 5 × 107 plaque-forming units) or a control Ad5 virus (5 × 107 plaque-forming units). Two weeks after intratracheal vaccination, human papillomavirus (HPV)–positive mouse tonsil epithelial cells (MTECs) were implanted into the posterior part of the oropharynx in mice. The log-rank test was used to demonstrate that survival differences between the Ad5 control and Ad5 E6/E7 groups were statistically significant at all doses of vaccine tested. *P < .05.
Lee DW, Anderson ME, Wu S, Lee JH. Development of an Adenoviral Vaccine Against E6 and E7 Oncoproteins to Prevent Growth of Human Papillomavirus–Positive Cancer. Arch Otolaryngol Head Neck Surg. 2008;134(12):1316-1323. doi:10.1001/archoto.2008.507
Copyright 2008 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2008
To test whether an immunization strategy that targets the E6 and E7 oncoproteins (E6/E7) may be an effective means to prevent the development of human papillomavirus (HPV)–positive head and neck squamous cell cancers using an in vivo mouse model and to determine whether such a response would prevent the establishment of viral transformed cells in vivo.
Adenoviral recombinant vaccine expressing HPV-16 E6/E7 (adenovirus 5 E6/E7, or Ad5 E6/E7) was generated. Specificity and timing of the E6/E7-specific cellular immune response was determined in vivo. Adenovirus 5 E6/E7 efficacy and route of administration required for clearance of HPV–positive tumor cells were monitored.
We generated Ad5 E6/E7 oncoproteins. Splenocytes from mice immunized with Ad5 E6/E7 produced interferon (IFN)-γ to cells expressing E6/E7 but not to cells lacking these oncoproteins. A time course of IFN-γ response showed that E6/E7-specific IFN-γ production is significantly increased in the first 2 weeks after administration of the vaccine and is substantially maintained for up to 70 days. Inoculated mice cleared E6/E7-expressing tumor 70 days after implantation. At all dosages of vaccine, mice inoculated with Ad5 E6/E7 completely cleared E6/E7-expressing tumor cells implanted 2 weeks after either intratracheal or submucosal inoculation, with significant E6/E7-specific IFN-γ production.
Immunization with HPV–16 E6/E7 oncoproteins can be an effective method of protecting a host from E6/E7-expressing head and neck squamous cell cancers via generation of a potent immune response. Such a response may be beneficial when combined with traditional treatment such as surgery, chemotherapy, or radiotherapy, thus improving the prognosis and quality of life of patients with HPV-16–associated head and neck squamous cell cancers.
The incidence of head and neck squamous cell cancers (HNSCCs) has increased worldwide.1 Substantial evidence supports a role of human papillomavirus type 16 (HPV-16) in the carcinogenic progression of at least 25% of HNSCCs.2,3Specifically, a high percentage of oropharyngeal HNSCCs (60%) are HPV-16 positive, and these occur with more metastatic and advanced disease than HPV-negative cases in oropharyngeal squamous cell cancers.2,4 In particular, the incidence of HNSCC of the tonsillar region is increasing.2,5,6 Furthermore, HPV-16–related HNSCCs are prevalent in populations with human immunodeficiency virus infection and in patients with immune system compromise from other causes.7,8 With the worldwide increase in the prevalence of human immunodeficiency virus infection, it is likely that there will also be a corresponding increase in HPV-16–related HNSCCs. Clearly, there is a need to better understand the mechanism of this emerging disease.
Unlike HPV-negative cancers, HPV-positive cancers express nonself viral proteins that aid in the malignant transformation of epithelial cells.9- 16 Therefore, because of the vital role these oncoproteins have in transformation, it is likely that persistent expression is required for invasive growth. The 2 major viral oncoproteins are E6 and E7 (E6/E7), both of which are multifunctional oncoproteins. The best-known function of the E6 protein is the targeting of the p53 protein for ubiquitin-mediated degradation,9 and inactivation of the p53 pathway is important for cellular immortalization and the abrogation of normal responses to DNA damage.10,17,18 The E6 protein also has numerous other functions, including degradation of a phosphatase, the loss of which enables invasive growth and activation of telomerase, the ribonucleoprotein complex that adds telomeric repeats to the ends of eukaryotic chromosomes, thus allowing immortal growth.11,12 The best-known function of the E7 protein is the binding and inactivation of the retinoblastoma tumor suppressor protein pRb,13 which has a key role in the transition of cells from the G1 to the S phase. The E7 protein also has many other reported functions,14 including blocking the effects of the cyclin-dependent kinase inhibitors p21 and p27.15,16
Because HPV-positive cancers are related to a viral infection and express viral proteins, it may be possible to develop and use an immunization strategy to prevent infection. The recent capsid protein vaccines prevent HPV infection.19- 21These results are likely related to development of a neutralizing antibody response to the viral coat protein, thus preventing the initial virus entry into the cell. Despite this initial success with use of the viral coat proteins, it will also be valuable to develop other strategies to prevent infection. It is possible that the viral coat protein may change in a manner that enables infection and prevents neutralization. If such a strain were to develop, it is likely that the functional oncoproteins would remain similar because they are crucial to cell transformation and progression to disease. Also, the current vaccine will likely be ineffective in treating established disease because transformed cells often do not express viral coat proteins targeted by the vaccine. Preliminary studies have shown no effect on resolving established disease.22,23 In the present study, we tested whether an immunization strategy that targets the E6 and E7 oncoproteins (E6/E7) may be an effective means to prevent the development of HPV–positive HNSCCs using an in vivo mouse model and to determine whether such a response would prevent the establishment of viral transformed cells in vivo.
Previously, we developed a preclinical mouse model of HPV-positive and HPV-negative tonsillar cancers.24 This model uses reimplantation of syngeneic mouse tonsil keratinocytes that have been transformed with HPV-16 E6/E7 and H-ras, hence HPV positivity. The HPV-negative lines were generated by knocking out a cellular target of E6 (phosphatase PTPN13) using a small hairpin RNA approach and H-ras, thus HPV negativity. In the present study, we created an immunization strategy that uses adenovirus to deliver the viral oncogenes and we tested whether immunization with this vector develops an immune-specific response to HPV-positive tumor cells. In addition, we sought to determine whether such a response would prevent the establishment of viral transformed cells in vivo. The findings from the present study may have clinical implications in preventing both the development of HPV-positive cancer and its recurrence.
Normal mouse tonsil epithelial cells (MTECs) were isolated from the oropharyngeal epithelium in C57BL/6 mice and cultured as previously described.24 In brief, MTECs were transduced with HPV-16 oncogenes E6/E7 plus H-ras, or small hairpin PTPN13 plus H-ras using retroviral vectors, and stable cell lines were developed. To test the adenoviral recombinant vaccine expressing HPV-16 E6/E7 (adenovirus 5 E6/E7, or Ad5 E6/E7) virus, we used A549 cells cultured in Dulbecco modified Eagle medium with 50% of Ham F12 medium supplemented with 10% heat-inactivated calf serum; penicillin, 100 U/mL; streptomycin, 100 μg/mL; and L-glutamine, 1%. All cell lines were cultured in a humidified incubator with 5% carbon dioxide at 37°C. All products required for the cell culture were purchased from Invitrogen Corp, Grand Island, New York, unless otherwise stated.
The E6/E7 open-reading frame was excised from the pLXSN16E6E7 vector using BamHI and EcoRI digestion and ligated into the BamHI and EcoRI sites on the adenoviral shuttle virus plasmid VQ Ad5CMVK-NpA vector. Adenovirus 5 E6/E7 was generated by ViraQuest Inc, North Liberty, Iowa. All vectors were propagated in HEK-293 cell line and purified using cesium chloride gradient centrifugation, and titers were determined using standard plaque assays.25
Exponentially growing A549 cells were infected at a multiplicity of infection of 100 with Ad5 E6/E7 overnight. All infections were performed with the cells at 80% cell confluence.
After 18-hour transduction of Ad5 E6/E7 into the A549 cells, the cells were harvested and total proteins were extracted at 4°C with protein lysis buffer consisting of TRIS hydrochloride, 50mM, pH 7.5; sodium chloride, 150mM; EDTA, 5mM; sodium orthovanadate, 2mM; sodium pyrophosphate, 10mM; sodium fluoride, 100mM; glycerol, 10%; Triton X-100, 1%; pepstatin, 10 μg/mL; leupeptin, 20 μg/mL; and aprotinin, 20 μg/mL. Expression of E7, p53, Rb protein, and actin was measured by immunoblotting total cellular protein with the specific E7 (Invitrogen Corp), p53 (Cell Signaling Technology, Inc, Danvers, Massachusetts), pRb (retinoblastoma protein) (BD Pharmingen, San Jose, California), and actin (Santa Cruz Biotechnology, Inc, Santa Cruz, California) antibodies using a standard Western blot technique. After incubation with primary antibodies, proteins were detected with peroxidase-conjugated donkey antigoat or goat antimouse secondary antibodies (Jackson ImmunoResearch Laboratories, Inc, West Grove, Pennsylvania) and blots were developed using chemiluminescence (Pierce Biotechnology, Inc, Rockford, Illinois). Jurkat cell lysates (BD Transduction Laboratory, San Jose, California) were used as a positive control for Western blot analysis in cases of Rb protein. Nontransduced cells or cells transduced with A549 were used as controls.
Adenovirus 5 E6/E7 oncoproteins were injected into mice intratracheally or submucosally at 5 × 105, 5 × 106, or 5 × 107 plaque-forming units (PFUs) (n = 5 in each group) before challenge with E6/E7/H-ras MTECs. For intratracheal administration, each mouse was anesthetized with halothane and infected by inoculating 20 μL of phosphate-buffered saline solution containing virus particles into a nostril. Mice receiving submucosal inoculation were also anesthetized with halothane and infected by injecting 50 μL of phosphate-buffered saline solution containing virus suspension into the base of the tongue. Adenovirus infection experiments were carried out in biosafety level 2 containment facilities and performed according to the National Institute of Health Guide for the Care and Use of Laboratory Animals. As a control, Ad5 viruses were given at 5 × 107 PFUs intratracheally or submucosally before challenge with E6/E7-expressing tumor cells. None of the immunized mice exhibited any clinically apparent adverse effects after inoculation with the vaccines.
Tumor growth was monitored using 2 methods. In all cases, tumors were measured using calipers after implantation of E6/E7/H-ras MTECs. To aid in visualizing clearance of tumor cells, we also produced a cell line expressing luciferase that can be monitored in vivo using live imaging. To create this cell line, E6/E7/H-ras MTECs were transduced with a retroviral vector that expresses luciferase and hygromycin resistance gene, and a pure population of cells expressing luciferase was obtained by selecting the cells with hygromycin. Growth of E6/E7/H-ras/Luc MTECs and E6/E7/H-ras MTECs are identical in mice (data not shown).
In these experiments, E6/E7/H-ras MTECs were implanted in the mice in 2 locations, either 1 × 106 cells subcutaneously in the flank or 5 × 105 cells submucosally in the oropharynx. Tumor growth was monitored at regular intervals by measuring the tumor volume or using the In Vivo Imaging System (IVIS26; Xenogen Corp, Hopkington, Massachusetts). Detection of tumor cells with the IVIS was accomplished by injecting D-luciferin, 15 mg/mL, in phosphate-buffered saline solution (Xenogen Corp) intraperitoneally at a dose of 100 μL per 10 g of body weight. Approximately 10 minutes after injection with D-luciferin, mice were anesthetized using isoflurane inhalation, then moved into the IVIS imaging cabinet, where they were maintained with isoflurane, 1% to 3%, delivered through nose cones on a manifold. Optimum exposure time to the IVIS camera is determined empirically; however, exposure times were less than 5 minutes. Imaging data were analyzed and signal intensity was quantitated with commercially available software (Living Image; Xenogen Corp).
To test immune response to cells expressing E6/E7, splenocyte suspensions were made by gently mincing spleen into a suspension, and red blood cells were lysed with ammonium chloride. Splenocytes, 1 × 106, were cocultured with 2 × 104 E6/E7/H-ras MTECs or 2 × 104 small hairpin PTPN13/H-ras MTECs (50:1) in RPMI (Roswell Park Memorial Institute) 1640 medium (Invitrogen Corp) plus fetal bovine serum, 10%; penicillin or streptomycin, 1%; and 2-mercaptoethanol, 10 μmol/L (Sigma-Aldrich Corp, St Louis, Missouri) in 12-well plate wells for 36 hours. The supernatants were used to assess spontaneous interferon (IFN)–γ production using a mouse IFN-γ immunoassay kit (Quantikine; R&D Systems, Minneapolis, Minnesota) according to the manufacturer's instructions.
Cytolytic T-cell response experiments were analyzed using the Mann-Whitney test comparing the amounts of IFN-γ generated from splenocytes in mice inoculated with Ad5 E6/E7 with those in mice inoculated with Ad5 control viruses or in naive mice and comparing the IFN-γ production just before and 3, 15, 28, and 70 days after inoculation. Survival differences between Ad5 control and Ad5 E6/E7-inoculated mice were analyzed using the log-rank test.
We chose to use adenovirus to develop an E6/E7 immune response for several reasons. It is easy to produce,25 has been used in multiple clinical trials,27,28 and effectively develops a cell-mediated immune response to transgenes.29,30 Adenovirus can also be used to infect cells of the aerodigestive tract.31,32 To accomplish the goal of adenoviral transgene expression of E6/E7, we inserted the open-reading frame of E6/E7 into a shuttle plasmid vector. After sequence verification, high-titer replication-defective Ad5 E6/E7 was produced commercially (ViraQuest, Inc). Insertion of these viral oncogenes did not affect the ability of the virus-producing cells to produce high-titer viral stocks. Therefore, it was possible to create high-titer adenoviruses that expressed the transgenes.
In this first study, we believed it was important to test whether an immune response could be developed to functional viral oncoproteins. It is possible that functional viral oncogenes may not allow a correct immune response because they alter the antigen presentation of viral proteins.33,34 To make certain that the viral vector produced functional viral proteins, we infected A549 cells with Ad5 E6/E7 (multiplicity of infection of 100) overnight. It is possible to detect E7 using a monoclonal antibody using Western blot analysis; however, a reliable E6 antibody is unavailable. Figure 1 shows that A549 cells transduced with Ad5 E6/E7 express E7 protein. To determine whether this protein is also functional, we tested whether Rb protein was degraded after transduction. Figure 1 shows that, after transduction with the vector producing E7, Rb protein is substantially lost. Because it was impossible to directly detect E6, we examined whether the expression of vector resulted in p53 loss, a known consequence of E6 expression.9,10 Western blot analysis for p53 after transduction showed that, even at 18 hours, E6 induced a substantial loss of p53. Reverse transcriptase–polymerase chain reaction was also used to confirm that E6 messenger RNA was produced after infection (data not shown). Together, the results suggest that the Ad5 E6/E7 vector produced functional HPV-16 E6/E7 oncogenes.
To determine whether Ad5 E6/E7 would allow a mouse to develop an immune response to cells immortalized by the viral oncogenes E6/E7 and also to determine when an immune response develops, we evaluated antigen-specific immune response by examining whether an antigen-specific IFN-γ response developed in inoculated mice. To test these parameters, spleens were removed from intratracheally Ad5 E6/E7 (5 × 107 PFUs) inoculated mice, and the white blood cells were allowed to react with either syngeneic mouse tonsil cells that express E6/E7 or cells that are immortalized by blocking PTPN13 expression (negative control cells that lack E6/E7 antigens). Figure 2 shows that white blood cells from immunized mice react only with cells that express E6/E7 and not with cells lacking the viral antigens, which suggests that mice were able to develop an E6/E7-specific immune response. To determine when this response occurs, splenocytes were harvested from naïve and inoculated mice 3, 15, 28, and 70 days after immunization. As shown in Figure 3, Ad5 E6/E7 inoculation resulted in E6/E7-specific IFN-γ responses from splenocytes, with significant enhancement of IFN-γ production at days 15, 28, and 70 after vaccination compared with no vaccination (P < .05, Mann-Whitney test). In addition, the time course of IFN-γ response showed that E6/E7-specific IFN-γ production is rapidly increased in the first 2 weeks after vaccination and is substantially maintained for up to 70 days.
To confirm intratracheal vaccination of Ad5 E6/E7 as a potential tool for prophylaxis against E6/E7-expressing tumor, E6/E7/H-ras/Luc MTECs (1 × 106) were implanted at 70-days postvaccination, and the mice were monitored for tumor growth at regular intervals by measuring the tumor volume and using the IVIS.26 Representative data from IVIS imaging at 28 days after E6/E7-expressing tumor cell implantation are shown in Figure 4. All of the mice previously administered the vaccine cleared E6/E7/H-ras/Luc MTECs, and no tumor developed up to 60 days after tumor cell implantation.
We wanted to determine whether the specific immune response in vitro would also protect mice against tumor formation in vivo. In addition, we wanted this preclinical model to closely mimic head and neck disease. Therefore, we implanted tumor cells in an orthotopic location in the posterior part of the tongue or oropharynx. We used 2 vaccine delivery methods to determine whether a viral infection of the surface epithelium or a submucosal infection may provide protection. Groups of C57BL/6 mice were immunized intranasally or submucosally in the posterior part of the tongue with 5 × 105, 5 × 106, or 5 × 107 PFU recombinant Ad5 E6/E7. Control mice were immunized with 5 × 107 PFU Ad5 viruses. Two weeks later, mice were challenged with submucosal injection into the posterior part of the oropharynx with 5 × 105 E6/E7 or H-ras/Luc MTECs, and tumor development was recorded. Forty percent of mice receiving adenovirus alone were able to clear the implanted tumor cells. In previous studies, we showed that 25% of mice can spontaneously clear these tumor cells by means of an immune mechanism (J.H.L., unpublished data, June 1, 2008). The response in the adenoviral controls is 15% higher than what we expected and may be due to increased nonspecific immune response stimulated by the adenovirus. As shown in Figure 5, all of the intratracheally Ad5 E6/E7-immunized mice were able to clear the tumor cells. Even a relatively low amount (5 × 105 PFUs) of infectious adenovirus conferred protection. All submucosally inoculated mice cleared the tumor cells at all vaccine doses (data not shown). Therefore, both routes of administration effectively cause the mice to clear tumor cells. To determine whether the effect of immunization was long lasting, tumor cells were implanted in 3 mice 70 days after immunization, and none of these mice formed tumors. However, we were able to detect an increase in IFN-γ production in splenocytes (Figure 3). These data strongly suggest that an immune-specific response to E6/E7 provides clearance of tumor cells that require these viral oncogenes for survival.
From the above experiments, we have 3 groups of mice: (1) mice that are not inoculated with Ad5 E6/E7 that forms tumors, (2) mice that are not inoculated with Ad5 E6/E7 that spontaneously clears tumors, and (3) mice that are vaccinated with Ad5 E6/E7 that clears tumors. We performed the following procedures to better understand whether the physiologic differences could be related to the level of antigen-specific immune response that develops in these animals. To examine the immune response, splenocytes from tumor-clearing or tumor-bearing mice in the previous vaccination experiment were cocultured with E6/E7/H-ras MTECs for 36 hours, and IFN-γ levels in the obtained supernatants were assayed. As shown in Figure 2, the splenocytes from mice immunized to E6/E7/H-ras MTECs only or Ad5 E6/E7 vaccines followed by E6/E7/H-ras MTECs generated significantly increased levels of IFN-γ in vitro on stimulation with E6/E7/H-ras MTECs compared with the splenocytes from age-matched naive mice (P < .05, Mann-Whitney test). Irrespective of their immunization status, the IFN-γ responses of splenocytes from tumor-clearing mice were significantly superior to those from tumor-bearing mice (P < .05, Mann-Whitney test). Furthermore, the IFN-γ productions in tumor-clearing mice inoculated with Ad5 E6/E7 were superior to those in tumor-clearing mice inoculated with control vector. However, only the group inoculated with the dose of 5 × 105 PFUs showed statistical significance (P < .05, Mann-Whitney test), and the amount of IFN-γ generation did not seem to increase as the vaccination dose of Ad5 E6/E7 was increased.
Human papillomavirus–related HNSCC offers the ability to advance treatment strategies based on a known oncogenic mechanism. It is feasible to offer a molecular test, within hours of a patient coming to the clinic, that will determine whether a tumor is in part caused by mechanisms related to HPV infection (thus, express E6 and E7) or to other causes.3,4,35,36 This ability to detect an oncologic cause will enable use of treatment strategies based on a viral mechanism to be formed. In the present study, we tested whether an immunologic vaccination strategy with a second virus that infects the epithelium of the head and neck will prevent formation of tumors with HPV oncogenes. Results of our preclinical studies suggest that such a strategy is feasible to prevent the formation of new HPV-related cancers in immune-competent mice. In the following discussion, we discuss potential clinical applications, compare work with this model with other strategies, and address applicable questions about such a strategy.
The goal of immunotherapeutic interventions is to directly or indirectly develop or augment immunologic responses against tumor-associated antigens (TAAs).37 Thus, successful immunotherapy for HNSCC depends on identification of suitable TAAs, which are expressed as a result of oncogenic cell transformation. However, compared with other human tumors, only a few TAAs, such as members of the c-erb-B family, certain germline antigens, or telomerase, have been defined in HNSCC. These TAAs have been largely used as tumor markers and have not been tested for their potential as immunogens.37- 40 Insofar as suitable TAAs as immunogens in HNSCC, especially in HPV-16–associated cases, HPV-16 E6/E7 proteins show promise. HPV-16 E6/E7 proteins are critically important in the induction and maintenance of cellular transformation in HPV-16–related cancers9- 16 and are consistently expressed as foreign antigens in most HPV-16–associated HNSCCs.2- 4,9 In the present study, our findings also strongly suggest that HPV-16 E6/E7–targeted vaccines can be a potential prophylactic tool against this disease.
One of the questions addressed by this study is whether functional HPV oncogenes can induce an immune response that results in antigen-specific tumor cell rejection. There is evidence that E6/E7 proteins are particularly immunogenic in human beings, with the production of both humoral and cell-mediated responses.41,42 We also show that it is possible to develop an immune response using an adenoviral strategy that expresses functional E6/E7. In the present study, mice with or without immunization against E6/E7 were challenged with E6/E7-expressing tumor cells, and their splenocytes were incubated with the identical tumor cells to assess spontaneous INF-γ production. Interferon-γ is used as a signature cytokine for Th1 responses and for effector function of CD8-positive T cells that are critical to tumor eradication.25 The in vivo clearance and in vitro INF-γ data together suggest that an immunization strategy against E6/E7 is an effective method for protecting a host from E6/E7-expressing HNSCC via generation of a potent immune response.
The route of vaccine administration was also examined. We used intratracheal vaccination as a route of immunization. This mimics the situation in human beings when the aerodigestive tract is confronted with HPV-16 E6/E7 antigens in cases of HPV-16–associated HNSCC. In these conditions, intraepithelial dendritic cells in the tracheobronchial epithelium may have a major role in E6/E7 antigen presentation by taking up antigens in the respiratory epithelium and carrying them to distant draining lymph nodes, where priming or the cytotoxic T-lymphocyte response occurs.43 It is known that the lungs, trachea, and tracheobronchial lymph nodes are the major immune inductive sites in intratracheal vaccination.44 This type of mucosal vaccination, rather than systemic, is easy to administer and has a low rate of secondary adverse effects. Furthermore, it can induce both mucosal and systemic humoral immunity and induce cellular immunity, including the cytotoxic T-lymphocyte response. In addition, there is reduced risk of injury with the use of syringes.45
Prophylactic vaccine for HPV-associated malignant tumors would have a substantial effect worldwide on morbidity and mortality, and it can be expected to decrease the incidence of all HPV-associated cancers, including HPV-16–associated HNSCCs. The current HPV vaccines are based on the induction of neutralizing antibodies against HPV capsid proteins and can be termed prophylactic for HPV infection. The development of viruslike particles that are indistinguishable from authentic virions (apart from lacking the viral genome) has accelerated the development of these vaccines, which show great promise as prophylactic vaccines in clinical trials.46 However, given that capsid proteins are not expressed at detectable levels in infected epithelial cells, vaccines targeting other nonstructural viral antigens such as E6/E7 proteins can have strong prophylactic and, more important, therapeutic potential to prevent HPV-associated cancers.
The vaccination method presented herein may expand the role of immunization to individuals with established HPV disease. Inasmuch as all persistent HPV lesions and HPV-associated cancers express E6/E7, such a strategy may be helpful as an adjuvant to treatment protocols. The target populations for therapeutic vaccination against HPV-associated malignant tumors would include patients with established HPV infection and HPV-associated cancers. Several studies have found that patients with HPV-associated HNSCC have improved survival compared with individuals with HPV-negative HNSCC. The reason for this is as yet unclear. It has been suggested that the difference in survival may be attributed to an immunologic response to the virus during therapy.47- 49 Vaccines as individual therapies for established antigenic cancer have had little effect. However, several successful investigational studies have used a vaccine as an adjuvant to established therapy.50,51 Therefore, generation of a cell-mediated immune response via therapeutic vaccination in HPV-16–associated HNSCCs may hold particular promise.
The determination that HPV-16 is an etiologic agent for HNSCCs, especially oropharyngeal squamous cell cancers, has elicited the development of preventive and therapeutic HPV-16 vaccines that may lead to the control of HPV-16–associated malignant tumors and their potentially lethal consequences. An understanding of the molecular carcinogenesis of HPV-16–associated cancers has led to the realization that HPV-16 E6/E7–associated concerns are important targets for the development of HPV-16 vaccines for the control of HPV-16–associated lesions. In accord with this, our data show that immunization with HPV-16 E6/E7 is an effective method for protecting a host from E6/E7-expressing HNSCCs via generation of a potent immune response. Safe and effective prophylactic HPV-16 vaccines such as intratracheally administered vaccine would show a protective effect and a reduction in the development of HPV-16–associated malignant tumors, including a subset of HNSCCs. Therapeutic vaccines may also be a practical option for patients with HPV-associated cancers such as those of the head and neck, with attention paid to efficacy, safety, and cost. Combining immunotherapy with traditional treatment such as surgery, chemotherapy, or radiotherapy may be another option for improving the prognosis and quality of life of individuals with HPV-16–associated HNSCCs.
Correspondence: John H. Lee, MD, Department of Otolaryngology–Head and Neck Surgery, Sanford Research/USD, Sanford School of Medicine, University of South Dakota, 1310 W 22nd St, Sioux Falls, SD 57105 (Leej@stanfordhealth.org).
Submitted for Publication: February 6, 2008; final revision received April 28, 2008; accepted May 5, 2008.
Author Contributions: Dr J. H. Lee had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: D. W. Lee and J. H. Lee. Acquisition of data: D. W. Lee, Anderson, and Wu. Analysis and interpretation of data: D. W. Lee and J. H. Lee. Drafting of the manuscript: D. W. Lee and J. H. Lee. Critical revision of the manuscript for important intellectual content: Anderson, Wu, and J. H. Lee. Statistical analysis: D. W. Lee. Obtained funding: D. W. Lee and J. H. Lee. Administrative, technical, and material support: Wu and J. H. Lee. Study supervision: J. H. Lee.
Financial Disclosure: None reported.
Funding/Support: This study was supported by grant KRF-2007-013-E00043 from the Korea Research Foundation funded by the Korean Government, Ministry of Education & Human Resources Development (Dr D. W. Lee). Dr J. H. Lee was supported by a Veterans Affairs Medical Center Merit Award.