**6. Malignant oral lesions and HPV**

In the oral cavity, 24 types of HPV (1, 2, 3, 4, 6, 7, 10, 11, 13, 16, 18, 30, 31, 32, 33, 35, 45, 52, 55, 57, 59, 69, 72 and 73) have been associated with benign lesions and 12 types (2, 3, 6, 11, 13, 16, 18, 31, 33, 35, 52 and 57) with malignant lesions [47-48]. Since the first report of the presence of HPV DNA in head and neck cancer, 65 high-risk types have been consistently detected at different sites; however, these types are specifically found in transcriptionally active tumor cells [74]. According to data from a review, 99% of HPV-infections in head and neck cancers are by high-risk types 16, 18, 31 and 33 [75]. Infection with HPV 33 accounts for up to 10% of positive head and neck cancers; however, the HPV 16 type is by far the most common subtype detected in head and neck cancer ([75-77], and also, oropharyngeal cancer (OPC) is more likely to have HPV 16 than other types at head and neck sites. Just to demonstrate the high levels of HPV-16 genotype in OPC, this genotype accounts for 78% to 100% of positive cases, while HPV-18 accounts for only 1% of cases [75]. An interesting prevalence profile of the HPV types has been observed in some investigations in the countryside of Sao Paulo state in Brazil, where a higher prevalence of HPV 18 than HPV 16 was found in oral and cervical carcinomas. Furthermore, the presence of HPV 18 was found to be associated with metastasis to the lymph nodes and shorter patient survival [78-80].

wild type p53, while classical HNSCC have usually a mutated form of the protein, in accord‐ ance with the expected better development of HPV-associated lesions. Indeed, among HPVpositive tumors, the worst outcome is related to smoking, showing evidence that tobaccoderived carcinogens could potentiate the transformation effect of HPV [89-90]. But p53 status in HPV-related tumors, especially those presenting HPV-16 infection led to a confusion involving HPV detection methods and even HPV-related carcinogenesis itself. Initially, it was expected that HPV-16 positive tumors to have a predominantly mutated p53 status, given that HPV-16 E6 inactivates p53, and therefore, mutations in TP53 would be, and indeed are rarely present in cervical carcinomas. But in HNSCC, TP53 is mutated in 60-80% of all cases, and it was expected that HPV-infected tumors would be among the 20-40% of wild-type TP53, although this is not what was found in HNSCC. These findings highlighted the importance of the detection method of HPV infections. For example, the HPV DNA PCR assay is too sensitive, since it detects only a few copies of viral DNA, and may detect more than oncogenic infections, but also productive infections, laboratory artefacts and virions [91]. The following additional techniques can also provide data regarding the presence of HPV: light and electron microsco‐ py, ELISA, gene expression by DNA microarray, Dot blot, Southern blot, hybrid capture and ligase chain reaction for probe amplification. Due to the existence of numerous options for HPV detection in HNSCC, a standardization of procedures for routine application has yet to be developed [77,85]. Among other important pathways in HPV-induced HNSCC are: (1) p53 and pRb pathways, involved in cell cycling; (2) EGFR pathway, which are an important therapeutical target in other cancers (as breast and lung cancers); (3) TGFβ pathway; (4) PI3K-

The role of Human Papillomavirus in Pre-Cancerous Lesions and Oral Cancers

http://dx.doi.org/10.5772/55943

253

PTEN-AKT pathway and (5) angiogenesis and hypoxia pathways [91].

Aimed to investigate the HPV frequency in Brazilian patients diagnosed with OSCC we performed a study to establish the HPV clinicopathological profile and its possible influence on prognosis of disease [84]. HPV expression in primary tumors (PTs), and their matched samples (MSs) of recidives, lymph nodal metastasis (LNM) or necropsies were correlated with survival of patients. Through polymerase chain reaction using one general and two typespecific HPV primers, 87 PTs and their corresponding MSs were tested. As first step, HPV-DNA detection was performed, using a GP5+/GP6+ primer (Bioneer Inc.) to amplify a 150-bp fragment from L1 gene of general HPV types (GP5+, 5'-TTTGTT ACTGTGGT AGA T ACT AC-3'; GP6+, 5'- GAAAAATAAACTGTAAATCATATTC-3'). At second step, PCR reaction was performed on HPV-positive DNA samples to determine if contained the genotypes -16 and -18, using specific primers targeting ~100 bp in the E7 ORF: HPV-16E7.667 (5'- GAT‐ GAAATAGATGGTCCAGC-3'), HPV-16E7.774 (5'-GCTTTGTACGCACAACCGAAGC- 3'), HPV-18E7.696 (5'-AAGAAAACGATGAAATAGATGGA-3') and HPV-18E7.799 (5'- GGCTTCACACTTACAACACA-3') (Bioneer Inc.). All of 87 OSCC patients analyzed, 17 (19.5%) presented tumors HPV-DNA positive. Analyses of all paraffin-embedded samples (87 primary tumors plus 87 matched samples) revealed the presence of HPV-DNA in 18 of 174 samples (10.4%), 10 samples (11.5%) from PTs, and 8 samples (9.2%) from MSs. In addition, no virus infection was detected in 7 (8.1%) MSs samples, and only one patient has demon‐ strated HPV-DNA positivity in both samples. HPV genotypes -16 and -18 were detected in 4 (22.2%) and 3 (16.7%) of the positive samples, respectively. Infection with both genotypes was found in 6 (33.3%) investigated samples, and HPV genotype was not identified in 5 (27.8%)

Several HNSCC have been analyzed for the presence of HPV, and HPV-DNA has been found in different proportions of tumors from different head and neck sites [75,81]. Some evidence has indicated that some subtypes of HPV are specifically linked to head and neck cancer, especially those arising from specific oropharyngeal subsites (e.g., tonsil and the base of the tongue) [82]. The HPV prevalence in HNSCC ranges from 3% to 40% and could vary more according to the specific site and HPV has been found in 4-80% of oral squamous cell carci‐ noma. Brazilian observations in the countryside of São Paulo state have found a low prevalence of HPV in tumors of the larynx [83] and an increase in the presence of HPV-DNA in oral cavity cancers during the past two decades [79-80,84]. The wide variation in HPV prevalence can be attributed to different detection techniques, small sample numbers, differences in the lesions and sampling techniques and epidemiological characteristics of the populations studied [85]. Among the many methods to detect HPV infections, both polymerase chain reaction (PCR) and in situ hybridization assays have been well validated, although not perfect.

In terms of incidence, it is now believed that HPV-infection could be responsible for approx‐ imately 20% of oral cancers and 60-80% of OPC. Recently, in 2011, International Agency of Research of Cancer (IARC) declared that there is sufficient evidence that HPV-16 is causally associated with oral cancer cases [86]. More important, these HPV-related oral cancers are now considered to be completely different entities, differing remarkably from HPV-negative tumors in their clinical response and overall survival [76,87].

Currently, the identification of distinct epidemiological profiles in HPV-positive and HPVnegative HNSCCs is possible. The main factors studied are heavy or no tobacco/marijuana exposure, heavy or mild alcohol consumption, poor or intact dentition, low or high oral sex exposure, age > 50 years or < 45 years, lower or higher socioeconomic status and deceasing or increasing incidence [82]. The epidemiological trend suggests that HPV-positive HNSCC occurs more often in younger patients (age < 50 years), which differs from the typical charac‐ teristics of head and neck cancer (which is more frequent in men above 40 years old). Tumors that show association with the presence of HPV usually appears strawberry-like and exophytic lesions on gross inspection and occur more frequently in the tonsil and the base of tongue with a basaloid aspect, poor differentiation and cystic changes within metastatic lymph nodes [82]. In addition, gene expression profiles are known to be different in HPV- positive OPCs compared with HPV-negative cases [88].

Molecular evidences have shown that HPV-associated oral tumors differ significantly from the classic "tobacco and alcohol"-associated oral tumors. First, HPV-positive HNSCCs harbor wild type p53, while classical HNSCC have usually a mutated form of the protein, in accord‐ ance with the expected better development of HPV-associated lesions. Indeed, among HPVpositive tumors, the worst outcome is related to smoking, showing evidence that tobaccoderived carcinogens could potentiate the transformation effect of HPV [89-90]. But p53 status in HPV-related tumors, especially those presenting HPV-16 infection led to a confusion involving HPV detection methods and even HPV-related carcinogenesis itself. Initially, it was expected that HPV-16 positive tumors to have a predominantly mutated p53 status, given that HPV-16 E6 inactivates p53, and therefore, mutations in TP53 would be, and indeed are rarely present in cervical carcinomas. But in HNSCC, TP53 is mutated in 60-80% of all cases, and it was expected that HPV-infected tumors would be among the 20-40% of wild-type TP53, although this is not what was found in HNSCC. These findings highlighted the importance of the detection method of HPV infections. For example, the HPV DNA PCR assay is too sensitive, since it detects only a few copies of viral DNA, and may detect more than oncogenic infections, but also productive infections, laboratory artefacts and virions [91]. The following additional techniques can also provide data regarding the presence of HPV: light and electron microsco‐ py, ELISA, gene expression by DNA microarray, Dot blot, Southern blot, hybrid capture and ligase chain reaction for probe amplification. Due to the existence of numerous options for HPV detection in HNSCC, a standardization of procedures for routine application has yet to be developed [77,85]. Among other important pathways in HPV-induced HNSCC are: (1) p53 and pRb pathways, involved in cell cycling; (2) EGFR pathway, which are an important therapeutical target in other cancers (as breast and lung cancers); (3) TGFβ pathway; (4) PI3K-PTEN-AKT pathway and (5) angiogenesis and hypoxia pathways [91].

positive head and neck cancers; however, the HPV 16 type is by far the most common subtype detected in head and neck cancer ([75-77], and also, oropharyngeal cancer (OPC) is more likely to have HPV 16 than other types at head and neck sites. Just to demonstrate the high levels of HPV-16 genotype in OPC, this genotype accounts for 78% to 100% of positive cases, while HPV-18 accounts for only 1% of cases [75]. An interesting prevalence profile of the HPV types has been observed in some investigations in the countryside of Sao Paulo state in Brazil, where a higher prevalence of HPV 18 than HPV 16 was found in oral and cervical carcinomas. Furthermore, the presence of HPV 18 was found to be associated with metastasis to the lymph

252 Human Papillomavirus and Related Diseases – From Bench to Bedside A Diagnostic and Preventive Perspective

Several HNSCC have been analyzed for the presence of HPV, and HPV-DNA has been found in different proportions of tumors from different head and neck sites [75,81]. Some evidence has indicated that some subtypes of HPV are specifically linked to head and neck cancer, especially those arising from specific oropharyngeal subsites (e.g., tonsil and the base of the tongue) [82]. The HPV prevalence in HNSCC ranges from 3% to 40% and could vary more according to the specific site and HPV has been found in 4-80% of oral squamous cell carci‐ noma. Brazilian observations in the countryside of São Paulo state have found a low prevalence of HPV in tumors of the larynx [83] and an increase in the presence of HPV-DNA in oral cavity cancers during the past two decades [79-80,84]. The wide variation in HPV prevalence can be attributed to different detection techniques, small sample numbers, differences in the lesions and sampling techniques and epidemiological characteristics of the populations studied [85]. Among the many methods to detect HPV infections, both polymerase chain reaction (PCR)

and in situ hybridization assays have been well validated, although not perfect.

tumors in their clinical response and overall survival [76,87].

compared with HPV-negative cases [88].

In terms of incidence, it is now believed that HPV-infection could be responsible for approx‐ imately 20% of oral cancers and 60-80% of OPC. Recently, in 2011, International Agency of Research of Cancer (IARC) declared that there is sufficient evidence that HPV-16 is causally associated with oral cancer cases [86]. More important, these HPV-related oral cancers are now considered to be completely different entities, differing remarkably from HPV-negative

Currently, the identification of distinct epidemiological profiles in HPV-positive and HPVnegative HNSCCs is possible. The main factors studied are heavy or no tobacco/marijuana exposure, heavy or mild alcohol consumption, poor or intact dentition, low or high oral sex exposure, age > 50 years or < 45 years, lower or higher socioeconomic status and deceasing or increasing incidence [82]. The epidemiological trend suggests that HPV-positive HNSCC occurs more often in younger patients (age < 50 years), which differs from the typical charac‐ teristics of head and neck cancer (which is more frequent in men above 40 years old). Tumors that show association with the presence of HPV usually appears strawberry-like and exophytic lesions on gross inspection and occur more frequently in the tonsil and the base of tongue with a basaloid aspect, poor differentiation and cystic changes within metastatic lymph nodes [82]. In addition, gene expression profiles are known to be different in HPV- positive OPCs

Molecular evidences have shown that HPV-associated oral tumors differ significantly from the classic "tobacco and alcohol"-associated oral tumors. First, HPV-positive HNSCCs harbor

nodes and shorter patient survival [78-80].

Aimed to investigate the HPV frequency in Brazilian patients diagnosed with OSCC we performed a study to establish the HPV clinicopathological profile and its possible influence on prognosis of disease [84]. HPV expression in primary tumors (PTs), and their matched samples (MSs) of recidives, lymph nodal metastasis (LNM) or necropsies were correlated with survival of patients. Through polymerase chain reaction using one general and two typespecific HPV primers, 87 PTs and their corresponding MSs were tested. As first step, HPV-DNA detection was performed, using a GP5+/GP6+ primer (Bioneer Inc.) to amplify a 150-bp fragment from L1 gene of general HPV types (GP5+, 5'-TTTGTT ACTGTGGT AGA T ACT AC-3'; GP6+, 5'- GAAAAATAAACTGTAAATCATATTC-3'). At second step, PCR reaction was performed on HPV-positive DNA samples to determine if contained the genotypes -16 and -18, using specific primers targeting ~100 bp in the E7 ORF: HPV-16E7.667 (5'- GAT‐ GAAATAGATGGTCCAGC-3'), HPV-16E7.774 (5'-GCTTTGTACGCACAACCGAAGC- 3'), HPV-18E7.696 (5'-AAGAAAACGATGAAATAGATGGA-3') and HPV-18E7.799 (5'- GGCTTCACACTTACAACACA-3') (Bioneer Inc.). All of 87 OSCC patients analyzed, 17 (19.5%) presented tumors HPV-DNA positive. Analyses of all paraffin-embedded samples (87 primary tumors plus 87 matched samples) revealed the presence of HPV-DNA in 18 of 174 samples (10.4%), 10 samples (11.5%) from PTs, and 8 samples (9.2%) from MSs. In addition, no virus infection was detected in 7 (8.1%) MSs samples, and only one patient has demon‐ strated HPV-DNA positivity in both samples. HPV genotypes -16 and -18 were detected in 4 (22.2%) and 3 (16.7%) of the positive samples, respectively. Infection with both genotypes was found in 6 (33.3%) investigated samples, and HPV genotype was not identified in 5 (27.8%) samples. The most prevalent infected anatomical site was the tongue. The main result of the present study was the significant number of positive HPV samples among non-smoking patients and although, a possible influence of HPV infection on carcinogenesis cannot be ruled out, the low frequency of HPV-positive OSCC cases found in our analysis leads us to suggest that this virus has not the same etiological influence on patients, as tobacco consumption does. Although we cannot to exclude a possible transient role for HPV in the OSCC induction, we believe that occasional detection of HPV-infection in OSCC resulting from the incidental colonization of tumoral lesions might reflect the true correlation of HPV in most analysis. [84].

tumors (combined HR: 0.85, 95% CI: 0.7–1.0). At the same study, HPV-positive patients had lower risk of disease-failure (recurrence of tumor) as compared to HPV-negative patients (meta HR, 0.62; 95%CI, 0.5–0.8). The evidence for association of OSCC with HPV-infection and its possible role as an oncogenic agent remains controversial. Schwartz et al. (2001) evaluating the HPV-16 influence on survival rate in OSCC demonstrated that patient's HPV-16 positive presented significantly reduced disease-specific mortality in OSCC (HR = 0.17, 95% CI = 0.04, 0.76) when compared with HPV-16 negative patients. This result suggests the HPV-16 infection could be associated with a favorable prognosis in OSCC. However, the mechanism responsible

The role of Human Papillomavirus in Pre-Cancerous Lesions and Oral Cancers

http://dx.doi.org/10.5772/55943

255

Several hypotheses have been proposed to explain the improved prognosis in tumors HPVpositive. The benefit on survival rate has been attributed to an enhanced radiosensitivity of tumors HPV-positive [95-96], and an improvement of apoptotic secondary response to the presence of unmutated p53 in HPV-associated tumors [95,97]. The improvement of diseasespecific survival rate could be associated with a reduction risk of second primary tumor, since these HPV-positive patients tend to have no prior history of tobacco and/or high alcohol consumption [95]. This finding reduces the field cancerization process (upper respiratory

Several epithelial lesions are originated by infection with human papillomaviruses (HPVs), mainly benign hyperplasia with low malignant potential like warts or papillomas. However, there is a subgroup of HPVs that are associated with precancerous lesions, which could become a cancer in a small fraction of people [99]. As example of those high-risk HPV subtypes, HPV 16 and 18 [100] are responsible for approximately 70% of cervical cancer cases and are present in more than 60% of HPV-infected penile cancer and HPV-16 is the genotype most frequently detected in head and neck carcinomas, found in up to 90% of HPV-positive cases [99]. Other high-risk HPV types account for virtually all of the remaining cases of cervical cancer, although in other primary sites they do not appear to have a similarly important role [101]. Therefore, cancer of the uterine cervix is most widely accepted malignancy as being associated with HPV infection. HPV high-risk subtypes are also associated with some others anogenital carcinomas, including penile, anal and vulvar cancers [102-103] and a subset of head-and-neck squamous

Taken together, these findings supports in several countries, vaccination against some HPV types on girls and young women with the goal of protecting them against HPV-induced cervical cancer [105-106]. Trials with vaccines against cervical cancer shown that crossprotection is possible, because this vaccines also have the potential to prevent other cancers that are caused by the same types of HPV, including some of head and neck cancers [107], and the most of anogenital cancers outside the cervix, including cancer of the vulva, vagina, penis, and anus [108-109]. In theory, these vaccines should target the same viruses at other anatomical sites, as head and neck. This approach could provide important information about the final

for this improved prognosis conferred by HPV is still unclear [94].

epithelium repeatedly exposed to carcinogens) [98].

cell carcinomas [104].

proof of HPV etiology in these tumors [110].

**8. HPV vaccines (Therapeutic and prophylactic)**
