T cells specific for HPV16 E7 epitopes in patients with squamous cell carcinoma of the oropharynx

Authors


Abstract

Squamous cell carcinomas of the oropharynx (SCCO) are often infected with oncogenic human papilloma virus (HPV) subtype 16. To determine the frequency of T cells specific for human leukocyte antigen (HLA)-A2.1 restricted HPV16 E7 protein-derived epitopes, tetramer analysis was performed using peripheral blood lymphocytes of 20 HLA-A2.1+ patients and 20 HLA-A2.1+ healthy individuals. Tetramers specific for 3 HPV16 peptides (E711–20, E782–90 and E786–93), an influenza matrix peptide (a model recall antigen) or an HIV reverse transcriptase peptide (a model novel antigen) were used in multicolor flow analysis. The HPV-specific T-cell frequencies were correlated with the HPV16 E7 and p16 status in tumor sections. In vitro stimulation (IVS) with autologous dendritic cells (DC) pulsed with HPV16 E7 epitopes was performed to demonstrate proliferation and antitumor activity of the HPV-responsive T cells. Frequencies of CD8+ T cells specific for HPV16 E7 peptides were not significantly different in patients with SCCO relative to normal donors. However, patients with tumors expressing HPV16 E7 (60%) and p16 (50%) had an increased frequency (p < 0.05) of T cells specific for the E711–20 epitope compared to those with tumors negative for both markers. HPV16 E711–20 and HPV16 E786–93 specific T cells were expandable upon IVS with cognate peptide-pulsed DC and were reactive against peptide-pulsed targets or, in case of the E711–20 epitope-specific T cells, against HPV16 E7 expressing CaSki cell line. Thus, in patients with HPV16+ SCCO, precursor T cells specific for E711–20 epitope are present (1/3,947) in the circulation, are responsive to stimulation with the cognate viral peptide and recognize in vitro HPV16 E7+ tumor cells. Further studies have to elucidate why those T cells are unable to eliminate the tumor in vivo and this might also allow for finding potential strategies that will increase the chances of developing a future HPV-based vaccine in patients with SCCO. © 2005 Wiley-Liss, Inc.

Squamous cell carcinomas (SCC) represent the most common type of head and neck cancer, accounting for 6% of all new cancer cases.1 Typical risk factors include alcohol and nicotine consumption. However, a significant proportion of the patients do not have these risk factors, and recent studies have suggested an association of SCC to viral pathogens such as high risk (oncogenic) human papilloma virus (HPV) types, particularly HPV16 or 18. These viruses are frequently found in squamous carcinoma of the head and neck.2, 3, 4, 5, 6, 7, 8, 9, 10, 11 HPV16 infections have been observed in approximately one half of squamous cell carcinomas of the oropharynx (SCCO).7, 10, 11, 12 The HPV16+ SCCO have a good prognosis and are not associated with conventional risk factors, suggesting they may represent a separate tumor entity.3, 7, 10, 11

The HPV-derived oncoproteins E6 and E7 are mainly responsible for both the onset and maintenance of malignant transformation through inactivation of the p53 and retinoblastoma (Rb) tumor suppressor genes, respectively.13, 14 Inhibition of Rb protein (pRb) by the HPV16 E7 oncogene product leads to upregulation of p16 via a loss of negative control of pRb expression. Thus, HPV16 E7+ SCCO coexpress p16 in nearly 90% of cases, and p16 has been suggested as a surrogate marker for the HPV status in SCCO.15

Since infections with HPV16 lead to expression of foreign proteins in the tumor, i.e., the HPV16 E7 oncoprotein, it could be expected that T-cell responses specific for these proteins develop and can be detected in peripheral circulation, similar to T-cell responses detected upon infection with the influenza virus. To investigate T-cell responses against HPV-derived peptides/epitopes, we used fluorochrome labeled tetrameric peptide–MHC class I complexes (tetramers) in multicolor flow cytometry. Tetramers allow for direct identification of antigen-specific T cells in the peripheral circulation.16, 17 To generate HPV16 E7-specific tetramers, immunogenic peptides had to be selected. To date, only a few human leukocyte antigen (HLA)-A2.1 restricted HPV16 E7-derived peptides, HPV16 E711–20, HPV16 E782–90 and HPV16 E786–93, have been shown to be immunogenic in cervical carcinomas.18, 19, 20, 21, 22, 23, 24, 25 We, therefore, selected these peptides for the production of tetramers to be used for the quantification of HPV16 E7-specific T cells in the peripheral blood of SCCO patients and normal controls.

In this manuscript, we report on the frequency of tetramer+ T cells in the peripheral circulation of SCCO patients with HPV+ and HPV tumors, and on the in vitro capability of these T cells to differentiate into antitumor effector T cells.

Abbreviations:

Ab, antibody; APC, antigen presenting cell; CTL, cytotoxic T lymphocyte; DC, dendritic cell; ELISPOT, enzyme linked immunospot; GM-CSF, granulocyte macrophage colony stimulating factor; HLA, human leukocyte antigen; HPV, human papilloma virus; IL, interleukin; IVS, in vitro sensitization; mAb, monoclonal antibody; MHC, major histocompatibility complex; PBMC, peripheral blood mononuclear cells; SCCO, squamous cell carcinoma of the oropharynx.

Material and methods

Peripheral blood mononuclear cells

Peripheral blood samples or leukapheresis products were obtained from 20 HLA-A2.1+ SCCHN patients and 20 HLA-A2.1+ healthy donors. Leukapheresis products were obtained from the Institute of Transfusion Medicine, Düsseldorf. The mean age of the patients (16 male and 4 female) was 49.4 years and that of normal controls (14 male and 6 female) was 50.2 years. None of the female individuals had a history of CIN. Peripheral blood mononuclear cells (PBMC) were isolated by centrifugation over Lympho Sep® lymphocyte separation medium (ICN, Aurora, Ohio) in Leucosep® Tubes (Greiner, Solingen, Germany). The study was approved by the local ethics committee at the University of Düsseldorf, and informed consent was obtained from each participating individual. PBMC were phenotyped for expression of HLA-A2 molecules by flow cytometry, using the anti-HLA-A2 monoclonal antibody (mAb), BB7.2, (ATCC, Manassas, VA) as well as an IgG isotype as a control and were additionally confirmed by the use of an HLA-A2.1-binding influenza-specific tetramer.

Tetrameric peptide–MHC class I complexes (tetramers)

Tetramers were purchased from Coulter (Krefeld, Germany) and included the following HLA-A2.1-binding peptides: GILGFVFTL, an influenza matrix immunodominant peptide (residues 58–66); ILKEPVHGV, the HIV-1 reverse transcriptase peptide (pol 476–484); YMLDLQPETT, corresponding to HPV16 E711–20; LLMGTLGIV, corresponding to HPV16 E782–90 or TLGIVCPI, corresponding to HPV16 E786–93.

To minimize background staining, each tetramer was titered and used at the lowest concentration that still gave a clearly discernible positive population in a donor vaccinated for influenza (for FLU58–66 tetramer) or in a HPV16 E711–20 specific cytotoxic T lymphocyte (CTL) line donated by Stephen Man, University of Wales College of Medicine, Cardiff, UK.26 Within a 2-fold range of tetramer concentrations bracketing the concentrations used here, the frequency of tetramer positive events and competition of CD3 binding27 were stable, and tetramer fluorescence intensity was within 80% of that obtained at saturating concentrations.

The default panel of antibodies used for these studies were CD3-FITC, CD14-PerCP and CD8-APC (BD PharMingen, Heidelberg, Germany). Isotype controls were used in all flow cytometry experiments.

Staining and flow cytometry analysis

Immediately prior to staining, cells were washed twice with medium, consisting of PBS, 0.1% (w/v) bovine serum albumin (BSA) and 0.1% (w/v) sodium-azide, and resuspended at a concentration of 5 × 106 ml−1 in a volume of 150 μl. Tetramer (5 μl of 1:10 dilution of the stock solution) was added and cells were incubated at room temperature for 30 min, followed by a 30 min incubation with antibodies (7.5 μl of each) at 4°C. After 2 additional washes, the cells were resuspended in ˜1 ml of 0.5% methanol-free formaldehyde in PBS. At least 1 × 106 events were collected using a four-color Beckton Dickinson, FACS Calibur Cytometer. Flow cytometry data were analyzed in real time using Beckman-Coulter System II software. In initial experiments, the region defining tetramer positive events was determined by evaluating PBMC stained with the antibody panel, but without tetramer. This region was held constant throughout the analysis. Data were saved for subsequent reanalysis in System II or WinMDI program (Version 2.8, Joseph Trotter, Scripps Inst., LaJolla, CA).

Immunohistochemistry for p16 and HPV16 E7

Tumor samples included in this study were available as paraffin blocks archived at the University of Düsseldorf, Institute of Pathology. The histology of each case was reviewed by a pathologist, and representative tissue sections containing areas of invasive SCCO were selected for immunohistochemistry. Normal appearing salivary gland tissue or skeletal muscle served as an internal nontumor control.

For immunohistochemistry, formalin fixed, paraffin-embedded tumor tissues were sectioned at 5 μm, air-dried overnight at 37°C, deparaffinized and dehydrated. After antigen retrieval and inactivation of endogenous peroxidase, the sections were stained with a mAb against HPV16 E7 (Clone 8C9, Zymed Lab. by WAK-Chemie, Bad Soden, Germany) and p16 (p16INK4a Research Kit, Dako, Glostrop, Denmark).

The avidin-biotin-peroxidase method was used to visualize the HPV16 E7 antigen (Vectastain-Elite-ABC Kit; Vector Laboratories, Burlingame, CA). Counterstaining was provided by Mayer's hemalaun. The immunostained slides were evaluated by light microscopy for HPV16 E7 expression. The tumor was considered HPV16 E7 positive when more than 25% of the tumor cells showed staining intensity of 2+ and higher on scale of 0–4+. IgG isotype mAb used at the same concentration as the primary mAb served as a negative control.

For p16 staining, the visualization reagent provided in the research kit was used and was completed by a counterstaining with Mayer's Hematoxilin. Nuclear as well as cytoplasmic staining was considered positive for p16 expression. Immunostaining was graded and scored on whole sections according to Klaes and colleagues,28 and tumors were scored p16 positive if >25% of the cells were stained diffusely positive.

Generation of anti-HPV16 specific T cells using peptide pulsed dendritic cells

Human dendritic cells (DC) were generated according to a modified method of Sallustro and Lanzavecchia.29 Briefly, PBMC were incubated for 2 hr at 37°C in AIM-V medium, and nonadherent cells were removed by gentle washing with warm medium. The remaining (plastic adherent) cells were incubated in AIM-V medium (Life Technologies), 1,000 U/ml granulocyte macrophage colony stimulating factor (GM-CSF; Immunex, Seattle, WA) and 1,000 U/ ml interleukin (IL)-4 (Schering Plough, Kennilworth, NJ). Immature DC were harvested on day 6, using cold Hanks' solution (Life Technologies) and used as antigen presenting cells (APC). DC were resuspended at the concentration of 2 × 106 cells/ml in PBS containing 10 μg/ml peptide and incubated at 37°C for 45 min. Subsequently, the peptide-pulsed DC were cocultured with PBMC in 24-well tissue culture plates (Costar, Corning, NY) in a final volume of 2 ml/well of X-Vivo medium (Cambrex, Apen, Germany) and 25 ng/ml of IL-7 (BD Biosciences) for the first 72 hr and, additionally, with 20 IU/ml of IL-2 (Chiron-Cetus, Munich, Germany) for the remaining time in culture. The lymphocytes were restimulated weekly with the peptide-pulsed autologous DC (10:1) and harvested after the third cycle. Reactivity of generated T cells was tested against various targets in 24 hr enzyme-linked immunospot (ELISPOT) assays. The specificity of generated T cells was determined in antibody blocking experiments and confirmed by tetramer staining.

Enzyme-linked immunospot assay for IFN-γ

The ELISPOT assay was performed in 96-well plates (MaxiSorp, Nunc, Roshilde, Denmark). The capture and detection Abs and AEC substrate reagent were purchased from BD Biosciences (Human IFN-gamma ELISPOT Pair, AEC Substrate Reagent Set). For Ab blocking experiments, target cells were preincubated with 10 μg/ml anti-HLA class I-specific monoclonal Ab (mAb), W6/32 (HB95, ATCC) or respective IgG isotype control (IgG2a Pharmingen, San Diego, CA) for 30 min. Target cells included HLA-A2.1+, HPV16 infected cervical carcinoma line CaSki and 10 μg/ml peptide pulsed HLA-A2.1-transfected K562, the latter were kindly provided by W. Herr, University of Mainz, Germany.30 The effector:target ratio was 1:1 using 10,000 cells/well for each group.

Culture of target cells

Adherent cell lines were grown in plastic culture flasks (Greiner, Solingen, Germany) under standard conditions (37°C, 5% CO2, fully humidified atmosphere), using modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum (both Gibco, Eggenstein, Germany), 2 mM L-glutamine, 50 μg/ml streptomycin and 50 IU/ml penicillin (all ICN, Meckenheim, Germany), as described previously. To transfer or passage the cell lines, almost confluent monolayers were detached with 0.05% trypsin/0.02% EDTA solution (Boehringer, Mannheim, Germany). Subsequently, cells were washed twice in medium and resuspended in 96-well flat-bottom microtiter plates (Becton Dickinson, Heidelberg, Germany) or culture flasks.

Statistical analysis

Tetramer-positive cells were quantified by flow cytometry and expressed as frequencies (e.g. 1/1,000) or reciprocal frequencies (e.g. 1,000) among of CD8+ T cells. We examined raw reciprocal frequency data and log-transformed these data, using normal probability plots.31 For all 3 tetramers, the log-transformed data had a normal distribution. Accordingly, descriptive statistics (means, standard deviations/SEM, confidence intervals) and statistical analyses (Student's t test, two-tailed), were performed on log-transformed reciprocal frequencies.

Results

Enumeration of HPV16 E7 specific CD8+ T cells in peripheral blood of SCCO patients and healthy controls

A 4-color flow cytometric assay that simultaneously measures tetramer, CD3, CD8 and CD14 binding was used to determine the frequency of HPV16 E7-specific T cells in the circulation.27 Binding of the HPV16 E7 tetramers was compared to that of the influenza virus HLA-A2.1-restricted matrix peptide GILGFVFTL (FLU), a model peptide for recall responses, and the HIV reverse transcriptase peptide ILKEPVHGV (HIV, a model peptide for a new antigen). As described previously, the lower limit of detection of the current assay was determined to be 1/12,821 and was applied to all the data obtained from testing of patients or healthy controls.27 The mean frequencies of FLU- and HIV- specific T cells were 1/2,515 and 1/11,992, respectively.

Mean frequencies of tetramer+CD8+ T cells obtained in SCCO patients (1/6,189) were not significantly different from those of normal controls (1/7,364, p = 0.54) for the HPV16 E711–20 epitope. Patients had lower frequencies of T cells specific for the HPV16 E782–90 epitope (1/12,961) compared to normal controls (1/10,008), but the difference was not significant (p = 0.39). For HPV16 E786–93 epitope, the frequencies obtained in SCCO patients were higher (1/9,497) compared to normal controls (1/13,225) without reaching significance (p = 0.47, Fig. 1 and Table I).

Figure 1.

Reciprocal frequencies of HPV16 E7 tetramer+ CD8+ CD3+ T cells in 20 HLA-A2.1+ healthy donors and 20 HLA-A2.1+ patients with SCCO. The symbols are geometric means, and error bars represent the standard error of the mean. The closed and open symbols denote patients and controls, respectively.

Table I. Summary of Clinical Characteristics (TNM) and Frequencies of HPV16 E7–Specific T Cells Among CD8+ T Cells in 20 Patients with SCCO
PatientTNM statusFrequencies of HPV16 E711–20–specific T cellsFrequencies of HPV16 E7 82–90–specific T cellsFrequencies of HPV16 E786–93–specific T cells
  1. NA, not available.

1T2N0M01/33601/229101/3543
2T1N0M01/136211/272531/83032
3T3N1M01/228051/199121/2533
4T1N1M01/255751/151051/15222
5T3N1M01/51431/108551/807
6T3N1M01/132621/73531/10275
7T2N1M01/16641/143041/22245
8T4aN3M01/122121/98231/25680
9T3N2cM01/324831/93991/14616
10T2N0M01/70721/311861/8807
11T4aN2bM01/7677NA1/1901
12T2N1M01/106241/373811/4872
13T3N2bM01/1458NA1/36695
14T3N3M01/11121/328501/33627
15T2N0M01/27731/58351/8194
16T2N0M01/23451/99341/10014
17T2N0M01/1441NA1/13641
18T4aN2bM01/281431/1019NA
19T1N2bM01/36911/92541/4459
20T2N2bM01/62171/19834NA

Immunohistochemistry of HPV16 E7 and p16 in patients' tumors

Immunohistochemistry of HPV16 E7 protein in patients' tumors was performed to evaluate the potential of these tumors to present the epitopes, and ultimately relate this information to the frequencies of tetramer+ CD8+ T cells detected in the peripheral circulation of these patients.

Using a HPV16 E7 specific antibody, immunohistochemistry was performed on 20 paraffin-embedded oropharyngeal tumors. We found nuclear and cytoplasmic staining in 60% of the SCCO, comparable to that in positive controls (CaSki (Fig. 2) and samples known HPV16 positive cervical cancers, not shown). In those positively stained tumors, almost all neoplastic cells showed immunoreactivity (Fig. 2, upper panel). Nondysplastic squamous epithelium served as internal negative control. There was no correlation between HPV16 E7 staining and tumor stage or grade (Table I).

Figure 2.

Immunohistochemistry for HPV16 E7 (upper panel) and p16 (lower panel) in representative tumors. HPV16 E7 and p16 staining was detected in the nucleus and the cytoplasm. For HPV16 E7 staining, CasKi cells served as a positive control (upper right corner in the upper panel).

Immunohistochemical staining for p16 revealed strong immunoreactivity in the nuclei and in the cytoplasm (Fig. 2, lower panel) of 50% of the SCCO. Surrounding mesenchymal cells showed no p16 staining, but some weak staining was found occasionally in lymphocytes and salivary glands. Out of 12 HPV16 E7+ SCCO, 10 were positive for p16 protein. There was no case of a HPV16 E7 negative tumor with p16 expression. (Table II).

Table II. Frequencies of HPV16 E711–20–Specific T Cells and The HPV16 E7 as well as The P16 Status in SCCO
PatientFrequencies of HPV16 E711–20–specific T cellsTumor HPV 16 E7 proteinTumor p16 protein
11/3360
21/13621
31/22805
41/25575
51/5143++
61/13262
71/1664++
81/12212++
91/32483+
101/7072++
111/7677
121/10624++
131/1458++
141/1112++
151/2773++
161/2345++
171/1441++
181/28143
191/3691+
201/6217

Association between HPV16 E7 expression in tumors and frequency of HPV16 E7 specific T cells

The patients were divided into 2 subsets relative to HPV16 E7 expression in their tumor: (i) patients whose tumors expressed HPV16 E7 and (ii) patients whose tumors were HPV16 E7 negative. A higher frequency of CD8+ T cells specific for the HPV16 E711–20 epitope in patients with HPV16 E7+ tumors was observed compared to those with HPV16 E7 negative tumors (1/3,947 vs. 1/12,152, p = 0.02, Fig. 3). Individual data for the frequency of T cells specific for the HPV16 E711–20 epitope in SCCO patients with or without HPV16 E7 expression in the tumor are shown in Figure 4. For the other HPV16 E7 epitopes no significant difference was found between the 2 patient subsets. Overall, the results suggest that HPV16 E7 expression in the tumor is associated with a significantly increased frequency of HPV16 E711–20 epitope-specific CD8+ T cells in the patients' circulation. Nevertheless, the frequency of these T cells (1/3,947) was considerably lower from that found for FLU58–66 epitope (1/2,515).

Figure 3.

Increased reciprocal frequency of HPV16 E711–20 epitope-specific CD8+ T cells in patients with SCCO expressing HPV16 E7. PBMC and tumors from 20 HLA-A2.1+ patients with primary SCCO were evaluated for tetramer frequencies and expression of HPV16 E7 by immunohistochemistry, respectively. The closed and open symbols denote positive and negative patient groups. The symbols are geometric means and error bars represent the standard error of the mean. The asterisk (*) indicates a significant difference (p < 0.05).

Figure 4.

Comparison of reciprocal frequencies of HPV16 E711–20 specific CD8+ T cells in patients with SCCO and HPV16 E7 expression in the tumor. PBMC and tumors from 20 HLA-A2.1+ patients with primary SCCO and 20 HLA-A2.1+ normal controls were evaluated. The asterisk (*) indicates a significant difference (p < 0.05).

In vitro stimulation of HPV16 E7 specific T-cells with autologous DC

To test the ability of the T-cell precursors specific for HPV16 E7 epitopes found in the peripheral circulation to respond to the cognate epitope, in vitro stimulation (IVS) was performed, using autologous DC as APC. IVS experiments were performed with PBMC of SCCO patients with a high precursor frequency for at least 1 HPV16 E7 epitope. Reactivity of the generated effector cells against HPV16 E7 peptides was tested by γ-IFN ELISPOT assay. The results of representative ELISPOT assays for IFN-γ production are shown in Figure 5. T cells generated by priming of PBMC with the HPV16 E711–20 peptide (panel A), but not with the HPV16 E782–90 peptide pulsed onto autologous DC were reactive against K562 cells pulsed with the respective peptide or against HPV16 E7 expressing tumor cell line CaSki. Both reactions were significantly inhibited by anti-HLA class I Ab w6/32. For peptide HPV16 E786–93 (panel B), the only response against K562 cells pulsed with the corresponding peptide was significantly blocked by anti-HLA class I Ab w6/32, which was not the case for HPV16 E7 expressing tumor cell line CaSki. IVS failed to induce a significant and specific response using PBMCs of 2 HLA-A2.1+ normal controls and 1 HLA-A2.1+ patient characterized by low frequencies of precursor T cells specific for all 3 HPV16 E7 epitopes.

Figure 5.

Results of a representative ELISPOT assay for IFN-γ production. T cells were generated in IVS cultures by priming of PBMC with the HPV16 E711–20 peptide (a) or HPV16 E786–93 (b) pulsed onto autologous DC. The ELISPOT assay was performed after 3 IVS cycles. The T cells were tested against HLA-A2.1+ K562 cells pulsed with the respective peptide or HPV16 E7 expressing tumor cell line CaSki were used as targets at a ratio of 1:1. Spots were counted by 2 independent investigators [TKH, KS]. Representative experiments performed with PBMC obtained from patient 15 are shown. Asterisks (*) indicate a significant difference (p < 0.05) between the number of spots.

These results suggest that precursor T cells specific for HPV16 E711–20 or HPV16 E786–93 peptides are present in the patients' circulation and are able to differentiate, at least in vitro, into HPV-specific effector cells.

Discussion

A predisposition of the oropharyngeal mucosa to malignant transformation by high risk HPV16 has been detected in a series of studies, including tumors of the tongue (base), tonsil and pharynx, but not in control tissues.2, 3, 4, 7, 10, 11 A causal relationship between this viral oncoprotein and up to 50% of oropharyngeal cancer has been well established over the recent years. However, efforts to characterize immune responses against the identified HLA-A2.1 restricted HPV16 E7 epitopes26 or to use them for immunization have been relatively few. So far, little is known about the interaction between T cells and the HPV16 E7 oncoprotein presented in oropharyngeal cancer. This is in contrast to extensive studies of HPV-specific responses in cervical cancer.32, 33

Here, we report that patients with HPV16-transformed SCCO have higher frequencies of T cells specific for the HPV16 E711–20 epitope detectable in the peripheral circulation compared to patients with tumors negative for this viral oncoprotein or normal controls. In our hands, this was the only immunogenic epitope, as the frequency of T cells specific for the other 2 epitopes, HPV16 E782–90 and E786–93, was not statistically different among the 3 groups. Interestingly, results obtained in HLA-A2.1 transgenic mice suggest that the latter 2 epitopes are not processed or are inefficiently presented by APC.34, 35 This could explain why we have failed to detect T cells specific for these epitopes in patients with HPV16 E7+ SCCO. Thus, only the HPV16 E711–20 epitope is apparently immunogenic in patients with SCCO leading to an increased frequency of epitope-specific T cells in their circulation. Supporting data also come from Youde and collegues, who described, a higher susceptibility of HPV16 E711–20-specific T cells to in vitro expansion in patients with cervical cancer compared to normal controls.26 Similarly, Ressing et al. showed that patients with HPV16+ lesions occasionally had memory CTLs against a HPV16 E711–20, providing evidence for natural CTL immunity against HPV16 in patients with cervical lesions.36

In previous studies, we have evaluated T-cell responses against the “self-antigen” p53 in patients with head and neck cancer and found increased frequencies in those patients whose tumors had normal p53 expression. In contrast, in patients with tumors overexpressing p53, low frequencies of p53 peptide-specific T cells were detectable, suggesting immunoselection of epitope loss-variants.27, 37 In this study, using a “non-self” viral epitope to detect T-cell responses, no evidence for immunoselection was observed. Instead, T cells specific for the HPV16 E711–20 epitope were found with increased frequencies in SCCO patients whose tumors were HPV16 E7+ and shown to be able to differentiate in vitro into antitumor effector cells. Therefore, the question remains why those T cells were unable to eliminate the tumor in vivo. It could be expected that HPV16 E711–20 specific T cells are as potent and effective as influenza-specific T cells. However, the response to influenza-associated epitopes seems to be stronger, which is also reflected by a higher frequency of epitope-specific T cells that we observed. The frequency of T cells specific for the influenza matrix immunodominant peptide (FLU58–66) was 1/2,515 of CD8+ T cells, whereas the frequency of T cells specific for the HPV16 E711–20 epitope in patients with HPV16 E7+ tumors was much lower (1/3,947). One might speculate that either HPV16 is not as immunogenic as the influenza virus or that the tumor is not recognized by the specific T cells because it fails to present the epitope in association with HLA-A2.1 on its surface. Alternatively, the tumor might actively suppress the antitumor immune response by elimination of tetramer positive effector cells by apoptosis. The latter possibility is currently tested in our laboratories.

In the current study, we were able to induce and expand T cells specific for 2 of 3 HPV16 E7 peptides, which were able to recognize peptide loaded targets (HPV16 E711–20 and HPV16 E786–93) as well as naturally HPV16 E7 expressing CaSki cells (only HPV16 E711–20). Our data indicate that precursor T cells able to respond to cognate peptides under optimized conditions in IVS cultures exist in the circulation of patients with SCCO. These peptide-specific responders could be further characterized not just by release of IFN-γ in ELISPOT assays but also for the ability to produce regulatory cytokines, e.g., IL-10. Unfortunately, we were limited in the available cell numbers and could not further characterize the responding T cells. According to our tetramer data, we had expected only the induction of T cells specific for HPV16 E711–20. We were surprised by expansion of T cells specific for the HPV16 E786–93 peptide in IVS cultures, since the frequency of T cells specific for this epitope was not elevated in the circulation of this particular donor or of other patients with SCCO. However, in contrast to the HPV16 E711–20 peptide, generated HPV16 E786–93 specific T cells failed to recognize naturally HPV16 E7 expressing cell line CaSki, providing support for the hypothesis that recognition of the epitope on the tumor surface is of critical importance for antitumor responses. Bauer et al.34 have made similar observations with the HPV16 E786–93 peptide in transgenic mice. They concluded that the HPV16 E786–93 peptide is able to induce CTL responses if loaded on the antigen presenting HLA class I molecules, but that the peptide appears not to be processed or presented by HPV16 infected cells.

The prevalence of HPV16 E7 infection in our patients was at 60%, which is comparable to that found in the literature.2, 3, 4, 5, 6, 7, 10, 11 The immunohistochemically detected presence of E7 in tumor tissue sections suggest that considerable quantities of this protein are present in the tumor. This opens a possibility of Western Blot analysis as a confirmatory method for HPV16E7. Instead, as a confirmation method, we have chosen the measurement of p16 expression, which has been described as a surrogate marker of HPV16 infection.15 We found p16 expression in most (10/12 cases) of the HPV16 E7-expressing SCCO. This is readily explainable, since the HPV E7 oncogene product inhibits the activity of the pRb protein, and p16 is up-regulated via a loss of the negative feedback control mediated by pRb expression.6 The other 2 cases (HPV16 E7 positive, p16 negative) might represent tumors with low HPV16 viral load as suggested by Klussmann and colleagues.15 In support of this hypothesis, when those 2 patients were studied for T-cell frequencies specific for HPV16 E711–20 epitope, it became apparent that 1 was similar (1/3,691) and the other significantly below (1/32,483) the geometric mean of specific CD8+ T cells in those patients with HPV16 E7-infected SCCO (1/3,947).

In summary, using tetramers we found that the frequency of HPV16 E7-specific T cells was comparable patients with SCCO and healthy controls. However, elevated frequencies of T cells specific for HPV16 E711–20 epitope were observed in patients with SCCO whose tumors expressed HPV16 E7 and p16 proteins. While the HPV16 E711–20 epitope is apparently immunogenic in patients with HPV16 E7+ SCCO, the epitope-specific T cells are unable to eliminate the tumor. Further studies are necessary to explain this resistance of tumor targets to cytotoxic T cells and to find potential strategies that will increase the chances of developing a future HPV-based vaccine.

Ancillary