Cytokines and Markers of Immune Response to HPV Infection

Cervical cancer is the third most commonly diagnosed cancer in women worldwide (Ferlay, Shin et al. 2010) and is a result of infection with cancer-causing types of human papillomavirus (HPV) (Bouvard, Baan et al. 2009). HPV is a very common infection, although in most circumstances, infection does not usually result in cervical disease (Trottier and Franco 2006). In fact, the natural history of HPV infection suggests that additional factors are required to drive progression from infection to the development of cancer. Most women are thought to clear their HPV infections within two years, but in approximately 10% of women, infection persists (Schiffman, Castle et al. 2007). Persistent HPV infection is, in effect, the strongest risk factor for progression to cervical precancer and cancer (Koshiol, Lindsay et al. 2008), and a dysfunctional immune response is likely to underlie the amplified risk that leads to HPV persistence and cervical cancer. Although efficacious prophylactic vaccines against the two types of HPV (16 and 18) that cause about 70% of cervical cancers (Munoz, Castellsague et al. 2006) are available, these vaccines are expensive, difficult to administer in poorer countries and will not protect women who have already been exposed to the virus (FUTURE II Study Group 2007; Hildesheim, Herrero et al. 2007) (Su, Wu et al. 2010). Thus, it is important to understand factors that predispose some women infected with a carcinogenic HPV infection to persist and progress.


Introduction
Cervical cancer is the third most commonly diagnosed cancer in women worldwide (Ferlay, Shin et al. 2010) and is a result of infection with cancer-causing types of human papillomavirus (HPV) (Bouvard, Baan et al. 2009).HPV is a very common infection, although in most circumstances, infection does not usually result in cervical disease (Trottier and Franco 2006).In fact, the natural history of HPV infection suggests that additional factors are required to drive progression from infection to the development of cancer.Most women are thought to clear their HPV infections within two years, but in approximately 10% of women, infection persists (Schiffman, Castle et al. 2007).Persistent HPV infection is, in effect, the strongest risk factor for progression to cervical precancer and cancer (Koshiol, Lindsay et al. 2008), and a dysfunctional immune response is likely to underlie the amplified risk that leads to HPV persistence and cervical cancer.Although efficacious prophylactic vaccines against the two types of HPV (16 and 18) that cause about 70% of cervical cancers (Munoz, Castellsague et al. 2006) are available, these vaccines are expensive, difficult to administer in poorer countries and will not protect women who have already been exposed to the virus (FUTURE II Study Group 2007;Hildesheim, Herrero et al. 2007) (Su, Wu et al. 2010).Thus, it is important to understand factors that predispose some women infected with a carcinogenic HPV infection to persist and progress.
HPV uses a variety of methods to avoid immune detection, such as maintaining an unobtrusive infectious cycle (e.g., non-viremic and non-cytolytic since replication occurs in cells already destined for natural cell death), suppressing interferon response, and downregulating toll-like receptor (TLR)-9 (Stanley 2010).By employing such immune evasion tactics, HPV infection itself does not lead to a direct or obvious inflammatory response.Rather, inflammation due to other co-factors such as smoking, parity, oral contraceptive use, co-infection with other sexually transmitted diseases, multiple sexual partners etc. have long been hypothesized to lead to HPV incidence, persistence, and progression to cervical precancer and cancer (Castle and Giuliano 2003).Studies that directly evaluate women's immune response to HPV infection may provide better insights into the role of inflammation and immunity in HPV persistence and cervical carcinogenesis.
Although humoral response to HPV infection has been well-characterized (Bhat, Mattarollo et al. 2011), cell-mediated response has not been well established.Numerous approaches have of numerous diseases.Further, the availability of inexpensive enzyme-linked immunosorbent assays (ELISAs), radioimmunoassays (RIAs), and other bioassays to reliably measure cytokines in these samples make them enticing targets for discovery.Currently, seven studies that met our inclusion criteria have directly examined HPV-infection-related immune responses in either serum or plasma (Table 1).All of these studies have focused on associations with carcinogenic infection using a Hybrid Capture assay.Hildesheim et al. (Hildesheim, Schiffman et al. 1997) was among the first to use plasma to evaluate markers of immunity.However, their comparison of carcinogenic HPV positive women with low-grade lesions to carcinogenic negative women with low-grade lesions failed to find a statistical difference in the soluble IL-2 receptor (sIL-2R; p=0.63).Adam et al. (Adam, Horowitz et al. 1999) similarly compared 10 women with high risk HPV infection to 10 HPV negative women and reported that high risk HPV infection was indeed associated with higher mean serum CSF-1 levels.Abike et al. (Abike, Engin et al. 2011) measured neopterin, often considered a marker of immune activation, and found lower concentrations in HPV-positive versus HPV-negative women with normal through high-grade histology.Unlike the earlier studies, Bais et al. (Bais 2005) measured numerous cytokines simultaneously (IL-2, IL-4, IL-10, IL-12, IFN-, TNF-), as well as soluble markers (sTNFRI and sTNFRII) in plasma.They discovered that higher mean IL-2 levels alone were associated with carcinogenic HPV positivity.Baker et al. (Baker, Dauner et al. 2011) evaluated eleven circulating markers (adiponectin, resistin, tPAI-1, HGF, TNF-, leptin, IL-8, sVCAM-1, sICAM-1, sFas, MIF) and found elevated levels of resistin [odds ratio(OR) for 3 rd versus 1 st tertile, 103.3; 95 confidence interval (CI), 19.3-552.8;P < 0.0001], sFas (OR, 4.2; 95% CI, 1.5-11.7;P = 0.003), 59.8;95% CI,; P < 0.0001), and TNA-(OR, 38.6; 95% CI, 9.1-164.3,P < 0.0001) were in women with persistent HPV infection compared to HPV-negative women.Kemp et al. (Kemp, Hildesheim et al. 2010) evaluated an even broader spectrum of cytokines in their comparison of 50 HPV-positive women older than 45 years and 50 HPV-negative similarly aged women from their population-based cohort study in Guanacaste, Costa Rica.Plasma levels of IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-13, IL-17, IL-1 , IFN-, GM-CSF, TNF-, MCP-1, MIP-1 , IP-10, RANTES, eotaxin, G-CSF, IL-12, IL-15, IL-7, and IL-1 were measured by Lincoplex assay, IFN-was measured by bead array, and TGF-1 was measured by ELISA.Their analysis revealed statistically significant differences between cases and controls in levels of IL-6, IL-8, TNF-, and MIP-1 , GM-CSF, IL-1 (all P < 0.0001) and IL-1 (P = 0.02).However, it should be noted that this study was intentionally designed to explore differences between the extremes of the immunological spectrum.Thus, differences between these groups are likely to be biased away from the null (upward) in comparison to the general population.All six of these studies failed to concurrently evaluate potential confounders, and with the possible exception of TNF-, none of their findings have been confirmed by other studies.
Unlike the other studies, Hong et al. (Hong, Kim et al. 2010) evaluated several potential confounders (parity, menopausal status, smoking, oral contraceptive use, histological findings of colposcopic-directed biopsy) in their recently published report of HPV persistence and clearance among 160 carcinogenic HPV positive Korean women (normal women or women with histologically confirmed mild dysplasia).While their univariate analysis revealed that the number of women who were serum negative for TNF-was significantly higher in the carcinogenic HPV clearance group (N=107) than their persistence group (N=53, P = 0.0363), their multivariate logistic regression analysis indicated that none of the four cytokines measured (IFN-, TNF-, IL-6, and IL-10) had a significant association with clearance of the carcinogenic HPV infection, pointing to the importance of these factors in future study design.In fact, they found that only age was significantly associated with clearance of carcinogenic HPV infections (OR, 0.95; 95% CI, 0.92-0.98;P = 0.001).2).Unlike the studies of circulating cytokines above, most of these studies have tested for a broad range of HPV types, although one (Guha and Chatterjee, 2009) only tested for carcinogenic HPV types using the Hybrid Capture 2 assay, and another only analyzed results for women with carcinogenic HPV infection compared to women without carcinogenic HPV infection (Marks, Viscidi et al. 2011).Scott et al. (Scott, Stites et al. 1999) (Gravitt, Hildesheim et al. 2003) found no statistical differences in IL-10 and IL-12 concentrations by HPV-positivity versus HPV-negativity in women selected from a population-based cohort study in Guanacaste, Costa Rica, after adjusting for stage of menstrual cycle, recent oral contraceptive use secretion volume, and pH.Lieberman et al. (Lieberman, Moscicki et al. 2008) used a multiplex immunoassay kit to measure IL-1 , IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12 (p40/p70), IL-13, IFN-in young women attending a family-planning clinic or university health center, or their friends.Although no significant differences were observed for women with incident or persistent HPV infections compared to women without HPV, there was some suggestion that IL-1 and IL-13 levels were reduced in women with incident or persistent HPV infections and that IL-6 and IL-2 levels were reduced in women with incident infections.Guha et al. (Guha and Chatterjee 2009) measured IL-1 , IL-6, IL-10, and IL-12 cytokine levels in commercial sex workers or spouses of HIV-positive men coming in for an HIV test.After taking HIV status into account, IL-1 , IL-10, and IL-12 seemed to be elevated in HPV-positive women compared to HPV-negative women.IL-6 was also higher in HPV-positive women compared to HPV-negative women (P ≤ 0.0004).After stratifying by HIV status, however, IL-6 was only notably elevated in in women positive for both HPV and HIV, making the association with HPV less clear.This study also evaluated cytokine levels by abnormal versus normal cervical cytology and found that only IL-6 was related to abnormal cytology (P = 0.03).Finally, a recent study by Marks et al. (Marks, Viscidi et al. 2011) evaluated 27 different cytokines in a multiplex assay in cervical secretions from 35-60-year-old women attending outpatient obstetrics and gynecology clinics for routine examination.Similar to Gravitt et al. (Gravitt, Hildesheim et al. 2003) and Lieberman et al. (Lieberman, Moscicki et al. 2008), this study found no association between IL-12p70 and HPV status.However, IL-5 (p = 0.03), IL-9 (p = 0.04), IL-13 (p = 0.01), IL-17 (p = 0.003), EOTAXIN (p = 0.04), GM-CSF (p = 0.01), and MIP-1 (p = 0.005) levels were elevated in women with carcinogenic HPV infection compared to those without carcinogenic HPV.In addition, T-cell and pro-inflammatory cytokines tended to be correlated with EOTAXIN in women with carcinogenic HPV, while they were correlated with IL-2 in women without carcinogenic HPV.The authors conclude that this shift from IL-2 to EOTAXIN may reflect a shift away from antigen-specific adaptive responses toward innate responses.Although there were no significant differences between groups, IL-1 and IL-13 seemed to be depressed in women with incident or persistent HPV infections.IL-6 and IL-2 also seemed to be depressed in women with incident infections.

www.intechopen.com
There is little consistency in the cytokines evaluated in these seven studies, but where there is overlap, the results tend to be contradictory.For example, one study found evidence that IL-6 levels were reduced in women with incident HPV infections (Lieberman, Moscicki et al. 2008), while another found that IL-6 levels tended to be elevated in HPV-positive women (Guha and Chatterjee 2009).Similarly, one study found no evidence that IL-12 levels varied by HPV status (Gravitt, Hildesheim et al. 2003), while two others (Crowley-Nowick, Ellenberg et al. 2000;Tjiong, van der Vange et al. 2001) observed higher levels of IL-12 in HPV-positive versus HPV-negative women.In addition, results from the study by Guha et al. (Guha and Chatterjee 2009) suggested a tendency toward increased levels of IL-1 in HPV-positive women versus HPV-negative women, while the results from Lieberman et al. (Lieberman, Moscicki et al. 2008) showed a trend toward decreased levels of IL-1 in women with incident or persistent HPV infection compared to HPV-negative women.These inconsistencies are not yet resolved.

Cytokine Responses in Patient-derived PBMCs.
There is evidence that cell-mediated immune responses play an important role in the control of HPV infections.Cell-mediated immune responses are regulated by T lymphocytes [T-helper (Th) lymphocytes and cytotoxic lymphocytes (CTLs)] in cooperation with antigen-presenting cells such as monocytes and dendritic cells.These cells all are modulated by and release cytokines that can influence one another's synthesis.Characterization (including quality and quantity) of lymphocytes directed against HPV epitopes has been examined with the goal of providing insights into the clinical outcomes of HPV-positive patients.To this end, analyses of cytokines and concurrent lymphoproliferative and CTL responses in patient-derived peripheral blood mononuclear cells (PBMCs), T-cell fractions isolated from PBMCs or whole blood cultures after stimulation with several antigens and/or HPV peptides has been evaluated in 10 publications (Table 3).
Tsukui et al. (Tsukui, Hildesheim et al. 1996) was one of the first to measure IL-2 levels in culture supernatants of PBMCs stimulated with predominantly 15mer overlapping peptides from HPV-16 E6 and E7 oncoproteins.The HPV early proteins E2, E6 and E7 are among the first of proteins that are expressed in HPV-infected epithelia.Stimulation with influenza served as a specificity control, and stimulation with phytohemagglutinin (PHA) served as a positive control since it is known to activate lymphocytes and induce rapid cell proliferation as well as lead to the release of inflammatory and immune cytokines.While the report itself focused on associations with IL-2 and disease progression, the study included both HPV typing data and IL-2 response data for each subject included in the study.Interestingly, by using the data presented in the paper for statistical calculation, we found that IL-2 levels were significantly increased in a group of 32 HPV positive healthy women and women with LSIL compared to a group of 51 HPV negative healthy women and women with LSIL (P=0.006).Among 18 women with HSIL with HPV typing and adequate IL-2 data, only 2 women had positive IL-2 levels (1 HPV positive, 1 HPV negative).
Several other studies also attempted to evaluate IL-2 levels in a similar manner.These findings are consistent with Tsukui et al. (Tsukui, Hildesheim et al. 1996) and suggest that IL-2 responsiveness may differ by cytological and/or disease stage.In 1999, deGruijl et al. (de Gruijl, Bontkes et al. 1999) again evaluated IL-2 levels, as well as IgG responses, in this same population.This time, they used HPV16 L1-VLP or synthetic L1-derived 15-mer peptides P1 (amino acids 311-325) and P2 (amino acids 321-335) to stimulate the PBMCs and sorted them as before.Importantly, they found IgG responsiveness was significantly associated with HPV16 persistence alone, but that CD4 T helper IL-2 responsiveness was significantly associated with both HPV clearance and persistence.Further, they reported that neither cell-mediated nor humoral immune responses against HPV16 L1 seemed adequate for viral control.In another publication, this group took their study one step further and measured IL-2 levels in response to HPV E2 N-terminal and C-terminal protein fragments (Bontkes, de Gruijl et al. 1999).They reported that HPV16 infection was not associated with IL-2 responsiveness against the N-terminal domain of E2, but HPV clearance was associated with IL-2 responsiveness against the C-terminal E2 domain.The following year, Bontkes et al. (Bontkes, de Gruijl et al. 2000) evaluated HPV 16 E6-and E7-specific memory cytotoxic T-cell precursor (mCTLp) activity in the same cohort of patients with cervical dysplasia.They found that activity was significantly associated with persistent HPV16 infection but not observed in HPV negative women or women with viral clearance.Kadish et al. (Kadish, Ho et al. 1997) had previously observed a similar phenomenon.Subjects with positive lymphoproliferative responses to E6 and/or E7 peptides were more likely to be HPV negative at the same clinic visit than were nonresponders (P = 0.039).Subjects who were negative for HPV and those with a low viral load were also more likely to respond than were those with a high viral load (P for trend = 0.037).These data suggest that lymphoproliferative responses to specific HPV 16 E6 and E7 peptides appear to be associated with the clearance of HPV infection.
In 2007, three additional reports evaluating patient-derived PBMCs were published.Sharma et al. (Sharma, Rajappa et al. 2007) focused on IL-2, IFN-g, IL-4, and IL-10 levels in PBMCs stimulated with PHA.They observed that increasing levels of IL-4 and IL-10 levels were significantly associated with HPV infection and that decreasing levels of IL-2 and IFNwere associated with HPV status.Seresini et al. (Seresini, Origoni et al. 2007) measured lymphoproliferative responses and IL-2, IFN-g, IL-4, and IL-10 levels in PBMCs stimulated not with HPV16 peptides, but rather with HPV18-specific E6 peptides.Their analyses revealed that one or more HPV18 E6 peptides were able to induce a response in 40-50% of the women evaluated.Response percentages increased to 80-100% when HPV18-positive women alone were considered.Levels of IFN-released were also shown to predict HPV persistence and/or disease relapse after surgery.In addition, they showed that a higher number of infiltrating CD4(+) and T-bet(+) T cells in lesions correlated with favorable clinical outcomes.Finally, Molling et.al. (Molling, de Gruijl et al. 2007) evaluated cultured PBMCs again stimulated with 14 different 20mer HPV16 E7 peptides or with PHA and measured both IL-2 levels and CTL activity.Importantly, they also measured invariant natural killer T-cells (iNKT) and FoxP3+ regulatory T cells (Tregs) levels by flow cytometry (FACSCalibur).While iNKT cells did not appear to be associated with HPV persistence, Treg frequencies were significantly increased in women with persistent HPV16 infection; and the Tregs were significantly more common in women who had detectable HPV16 E7 specific IL-2 producing T-helper cells.These data suggest that HPV infection may affect Treg development -a finding that opens the door for a whole new avenue of research related to HPV-related immune research.
Markers in Cervical Tissues PBMC responses and circulating or secreted cytokines can be useful indicators of immune response, but the best indications may come www.intechopen.comfrom the actual site of interaction between HPV infection and the immune system: tissue.A number of studies have attempted to measure immune markers in HPV-positive compared to HPV-negative women in different ways.Among studies included in this review, these markers fall into three major categories: immune presentation molecules, cytokines or cytokine receptors, and immune cells.
Several studies used immunohistochemistry (IHC) to stain for major histocompatibility complex (MHC) proteins in cervical tissue (Table 4).MHC class I molecules present endogenous antigens (cytoplasmic proteins) to cytotoxic (CD8+) T cells and are typically present on all nucleated cells (Murphy, Travers et al. 2011).In contrast, MHC class II molecules present exogenous antigens from outside the cell to helper (CD4+) T cells and are typically present only on antigen presenting cells, such as dendritic cells macrophages.Thus, normal cervical epithelial cells should be MHC class I positive and MHC class II negative.In humans, MHC class I consists of major human leukocyte antigens (HLA) A, B, and C and minor antigens E, F, and G, while MHC class II consists of HLA-DM, -DO, -DP, -DQ, and -DR.
Using a polyclonal stain specific for HLA-A, -B and -C heavy chains in formalin-fixed, paraffin-embedded (FFPE) tissue from biopsies and resection specimens from women with CIN1-3 or cancer, Cromme et al. (Cromme, Meijer et al. 1993) found that normal MHC class I expression, defined positive staining in ≥75% of cells, was reduced in women with HPV16, 18, or 31 infection versus HPV-negative women (p=0.04).MHC class II expression, as measured through a polyclonal HLA-DR antigen stain, was also altered, with normal staining (<25% positively stained cells) in 42% of women with HPV16, 18, or 31 infection v e r s u s 6 4 % o f H P V -n e g a t i v e w o m e n .T h i s a l teration was not statistically significant, however (p=0.14).Gonclaves el al (Goncalves, Le Discorde et al. 2008) also examined MHC class I expression in FFPE biopsy blocks, but in women with normal through cancerous histology.They found that HLA-A/B/C expression was not significantly elevated in HPVpositive compared to HPV-negative women (OR, 2.29; 95% CI, 0.77-11.00;P = 0.14).Strangely, HPV16/18 infection was inversely associated with HLA-A/B/C expression (OR, 0.12; 95% CI, 0.02-0.79;P = 0.04), but as reported, it was unclear whether this association was based on comparison to HPV-negative women, or a combination of both HPV-negative women and women with HPV infections other than HPV16 and 18. HLA-E expression tended to be increased in HPV-positive versus HPV-negative women (OR, 3.83; 95% CI, 0.49-30.10;P = 0.22), especially for HPV16/18 infections (OR,11.25;;P = 0.003).Similarly, Dong et al. (Dong, Yang et al. 2010) stained for HLA-G in FFPE blocks from CIN1-3 patients and found higher HLA-G expression in HPV16/18-positive patients than HPV16/18-negative patients (P = 0.02).
In addition to interaction with an antigen MHC complex, T-cells require costimulation with an antigen nonspecific molecule to be full activated.T cells that encounter antigen MHC complex without costimulation may be come anergic and thus tolerant to the presence of HPV.To investigate this possibility, Ortiz-Sanchez et al. (Ortiz-Sanchez, Chavez-Olmos et al. 2007) evaluated expression of the CD80 and CD86 MHC class II costimulatory molecules through immunohistochemistry (IHC), quantitative reverse transcriptase PCR (qRT-PCR), and RNA in situ hybridization (ISH) in FFPE biopsies from histologically normal HPVnegative women and HPV16-positive women with LSIL.They found that CD86, but not CD80, was expressed in all HPV-negative normal cervical epithelial samples, while CD86     expression was lower (73% by IHC) in HPV16-positive LSIL samples.This decrease in CD86 expression in HPV-positive women could represent and immune evasion mechanisms through the down-regulation of costimulatory molecules.
The next major category of immune markers measured in cervical tissue includes cytokines and their receptors.In addition to testing for MHC costimulatory molecules, Ortiz-Sanchez et al. (Ortiz-Sanchez, Chavez-Olmos et al. 2007) used IHC to stain for IL-10, which inhibits CD86 expression.IL-10 detection was likewise poor in both HPV-negative normal tissue and HPV16-positive LSIL tissue, but detection was higher in a high-grade SIL (HSIL) control sample.Fernandez et al. 2005 tested for IFN-, TNF-, and IL-10 protein from snap frozen cervical biopsies from HIV-positive or HIV-negative LSIL and HSIL patients infected with HPV using a double-sandwich ELISA approach.They reported that HPV16 was associated with higher IL-10 (P = 0.03) and IFN-(P = 0.04) intra-lesional levels than other HPV types, but HPV18 was associated with reduced TNF-(P = 0.009) and INF-levels (P = 0.01) suggesting that immune responses may vary by the infecting HPV type.
Another major category of immune markers is the immune cells themselves.Two studies in this review evaluated infiltrating immune cells in cervical tissue by visually counting the cells.Butsch Kovacic et al. (Butsch Kovacic, Katki et al. 2008) counted lymphocytes, neutrophils, macrophages, plasma cells, and eosinophils among women with typical squamous cells of undetermined significance or LSIL and found that cervical inflammation varies with type of HPV infection, as well as risk of persistence and progression.Women with carcinogenic HPV infections also had more severe epithelial inflammation and less severe stromal inflammation than HPV-negative women.These associations were limited to carcinogenic and not the non-carcinogenic HPV types.Dong et al. (Dong, Yang et al. 2010) determined that among HPV16/18-positive CIN lesions, moderate to strong HLA-G expression was associated with weak immune response, as measured by few tumor infiltrating lymphocytes (TIL), whereas weak HLA-G expression was associated with strong immune response (high numbers of TIL).HLA-G expression was not associated with TIL in HPV-negative women, suggesting that the increased HLA-G expression in HPV-positive lesion may reflect an inhibition of immune response against HPV.Brismar Wendel et al. (Brismar Wendel, Kaldensjo et al. 2010) used qRT-PCR to measure CD3, CD4, CD8, CD19, and CD27 expression but found no difference by HPV status.Finally, Øvestad et al. (Ovestad, Vennestrom et al. 2011) used IHC to stain for cell surface marker in biopsies from CIN2-3 patients and found that HPV16 and related types were correlated with lower CD8positive cell counts in the stroma compared to other HPV types (P = 0.02).

Conclusions and future perspectives
Taken together, these studies support the role of cell-mediated immune response in HPVrelated carcinogenesis although their findings, particularly for those measuring cytokines, are largely inconsistent.There are many potential explanations.There has been a real lack of consistency in sample collection methods, cytokine measurement methods and even the outcome definitions used for analyses.
For example, some studies assessed HPV positivity, regardless of timing and/or disease state, while others evaluated incident HPV infection or HPV persistence or clearance.Further, these studies more than often focused on HPV 16, on carcinogenic HPV types, or any HPV type infection together.However, those few studies that did evaluate immune markers by individual HPV type found evidence that immune responses vary by HPV type (Fernandes, Gonçalves et al. 2005;Butsch Kovacic, Katki et al. 2008;Ovestad, Vennestrom et al. 2011).Thus, HPV type is an important consideration.Moreover, while we chose to focus on immune markers' associations with HPV infection, most of the studies reviewed in this chapter predominantly assessed associations between immune markers and disease state ( L S I L , H S I L , c a n c e r o r C I N 1 -3 and cancer).Ideally, future studies would evaluate differences by individual HPV type and better consider the timing of disease.
There are also other notable differences in the study populations considered by these studies (e.g., sample size, young versus old women, inclusion of HIV-positive women).Many studies used convenience samples of women.In fact, there is a general lack of consideration for factors that could confound or modify both cytokine production and the infectious outcomes.Only eight of the 35 studies made any attempt to account for co-factors that may influence cytokine level.The importance of adjusting for such potential confounders was recently highlighted at an international workshop that addressed best practices for sampling techniques and assessment of mucosal immune responses.The workshop identified a number of characteristics that should be considered when studying female genital tract immunity, including age, race, body mass index, sexually transmitted infections, other genital tract infections, vaginal flora, alcohol or substance use, recent immunization, pregnancy, phase of menstrual cycle, genital inflammation, recent douching, gynecologic procedures, recent intercourse/semen, and contraception (Anderson and Cu-Uvin 2011).
The number of women included in each study is another important consideration in the evaluation of these studies.Twenty-two of the 32 (69%) studies meeting our criteria included less than 30 women in one or more groups (e.g., the HPV-positive or HPV-negative group).Fifteen (47%) included less than 20 in one or more groups.Small numbers of women in the comparison groups can lead to unstable results and may help explain why results for individual immune markers are so inconsistent.
Most studies have measured only a few cytokines, and few have evaluated infiltrating immune cells concurrently with cytokines, making it challenging to explore the activation pathways of cells involved in the immune response against HPV.One research group has made extensive use of their study population to characterize several aspects of immune response as measured in PBMCs (de Gruijl, Bontkes et al. 1998;Bontkes, de Gruijl et al. 1999;de Gruijl, Bontkes et al. 1999;Bontkes, de Gruijl et al. 2000;Molling, de Gruijl et al. 2007).However, few studies have been so thorough.In fact, more than half of the studies of PBMCs (five of nine studies), have come from this same research group with the same study population.Additional studies characterizing many aspects of immune response in different study populations would help clarify whether the results are broadly applicable.
Many studies focused on T-helper type 1 (TH1) versus T-helper type 2 (TH2) polarization, using a single cytokine (or small group of cytokines) to characterize the T-helper phenotype.However, advances in immunology have led to the shift of the TH1/TH2 paradigm to the TH1/TH2/TH17/T-reg hypothesis, a multi-lineage commitment from the same T-helper precursor cells.TH17 cells, in fact, have been shown to inhibit both TH1 and TH2 cells, and therefore are likely to play a critical role in HPV-related immune responses as well.For studies of cervical secretions, the collection method can have a large impact on the results.Of the seven cervical secretion studies included in this review, two collected cervical secretions through cervicovaginal lavage (Tjiong, van der Vange et al. 2001;Guha and Chatterjee 2009), four used Weck-cel® (Crowley-Nowick, Ellenberg et al. 2000;Gravitt, Hildesheim et al. 2003) or Merocel® (Lieberman, Moscicki et al. 2008;Marks, Viscidi et al. 2011) ophthalmic sponges, and one used cytobrush sample suspensions (Scott, Stites et al. 1999).Cervicovaginal lavages may not be specific enough to the cervix and may overly dilute the specimen.Even studies that used ophthalmic sponges tended to use Weck-cel® sponges (Gravitt, Hildesheim et al. 2003;Moscicki, Ellenberg et al. 2004), which may not provide adequate cytokine recovery, especially compared to Merocel® sponges (Castle et al.(Castle, Rodriguez et al. 2004).
Studies evaluating tissue have seldom considered both stroma and epithelium.In this chapter, only two studies examined inflammation in both stroma and epithelium (Butsch Kovacic, Katki et al. 2008;Ovestad, Vennestrom et al. 2011).One study found opposite inflammatory patterns by HPV status ((Butsch Kovacic, Katki et al. 2008)).This study also found that neutrophils tended to be found only in the superficial epithelial layers, whereas mononuclear cells were found mainly near the basement membrane, suggesting that inflammatory patterns in the stroma and the epithelium may depend on the specific cell type.The second study only reported differences by CD8 in the stroma (Øvestad 2011).
Tissue-based studies of cytokines are also heterogeneous.Only two studies evaluated cytokine proteins in tissue (Fernandes, Gonçalves et al. 2005;Ortiz-Sanchez, Chavez-Olmos et al. 2007).It is not surprising that few studies have evaluated cytokine proteins in tissue since it can be challenging to find an appropriate antibody and optimize the assay.For example, antibodies that perform well in western blots may not work for staining since staining requires fixation, which can change the confirmation of the cytokine protein, thereby preventing antibody binding (Sachdeva and Asthana 2007).Most tissue-based studies of cytokines in this review measured RNA expression, but accurate measurement of RNA expression requires high quality tissue.If the tissue was not snap frozen immediately after surgery and well maintained, endogenous RNases may have degraded the RNA.RNA quality is rarely addressed.Although the presence of cytokine transcripts in tissue may be meaningful, the absence is not given the short-lived nature of RNA, even for high quality tissues (Sachdeva and Asthana 2007).
To clarify the role of immune response in cervical carcinogenesis, future studies should be conducted in well-characterized epidemiologic studies that can address most or all of the characteristics and considerations described above.Studies should include large numbers of women, evaluate a broader spectrum of cytokines/immune markers and measure and adjust for potential confounders concurrently.Possible usefulness of tissue microarrays and multiplex arrays with well-defined phenotypes should be considered as they are likely to make these studies more feasible.Emerging results must be repeated in different study populations and specimen types, but are encouraging.Accumulating evidence indicates that there is a cell-mediated immune response to HPV.As technologies improve, it should become possible to better characterize these responses to distinguish between women at risk of developing cervical cancer and women who can effectively resolve their HPV infections.

Table 1 .
Studies of circulating immune markers in plasma and serum.

Table 2
. Studies of immune markers in cervical secretions.
(Ortiz- Sanchez, Chavez-Olmos et al. 2007 2007The recent study byMolling et al. (Molling, de Gruijl et al. 2007) is among the few that have measured these cells.Using flow cytometry in HPV16 E7 stimulated PBMCs, they determined that Treg frequencies were significantly www.intechopen.comgreater in women with persistent HPV16 infection and in women with detectable HPV16 E7 specific IL-2 producing T-helper cells, suggesting that HPV infection may affect Treg development.These findings may also be supported by tissue-based studies of MHC class II expression.Although data are limited, one study found evidence of increased MHC class II expression in HPV-positive versus negative patients(Cromme, Meijer et al. 1993), while another found reduced expression of the CD86 MHC class II costimulatory molecule(Ortiz- Sanchez, Chavez-Olmos et al. 2007).It could be hypothesized that HPV upregulates MHC class II expression and down-regulates MHC class II costimulatory molecules in order to increase T-cell anergy through incomplete signaling.Additional studies are needed to better understand these relationships.