Pathophysiology and carcinogenesis

Human papilloma virus is a small, non-enveloped, oncogenic DNA virus with more than 120 identified genotypes. ‘Low risk’ types (e.g. HPV 6 and 11) are associated with benign proliferative growths, such as common plantar and genital warts, and have been identified in respiratory papillomatosis. Human papilloma virus types 16 and 18, considered ‘high risk’ genotypes, have been found in the host DNA of patients with cervical cancer.Reference Hu and Goldie^3, Reference Campisi and Giovannelli⁸ These HPV types are also the most common genotypes identified in head and neck SCC, with HPV 16 identified in approximately 90 per cent of oropharyngeal SCC specimens. Human papilloma virus 16 has six global variants: European, North American, Asian American, Asian, African one and African two.Reference Yamada, Manos, Peto, Greer, Munoz and Bosch⁹ Human papilloma virus is an obligatory intracellular organism which infects epithelial tissue initially by mucosal damage, followed by transportation of the viral genome to the host's intracellular genome. It is believed that the HPV DNA fragments E6 and E7 bind to and inactivate the tumour suppressors p53 and pRb.Reference Slebos, Yi, Ely, Carter, Evjen and Zhang¹⁰ This results in defects in apoptosis, DNA repair mechanisms and cell cycle regulation.

In a systematic review of 5046 head and neck SCC cases from 26 different countries, it was noted that the prevalence of HPV was 25.9 per cent in all resection specimens. The prevalence of HPV was 35.6 per cent in oropharyngeal SCC, 23.5 per cent in oral SCC and 24 per cent in laryngeal SCC. The specific HPV 16 genotype was present in 86.7 per cent of HPV-positive oropharyngeal SCC, as compared with 68.2 per cent of oral SCC and 69.2 per cent of laryngeal SCC.Reference Kreimer, Clifford, Boyle and Franceschi²

The two possible pathways of HPV transmission to the head and neck mucosa comprise oral–genital contact and perinatal transmission.Reference Chaudhary, Singh, Sundaram and Mehrotra¹¹

There are a number of reasons cited for the predilection of HPV for both oral and cervical mucosa. The oral cavity and uterine cavity are easily accessible for infection. The squamous epithelium in these sites is derived embryologically from endoderm and is susceptible to metaplasia. Cytokines produced by nearby lymphoid tissue may influence HPV transcription and cellular transformation. Finally, the mucosa in these sites contain crypts lined with a reticulated epithelium, which may facilitate viral access to basal cells.Reference Hafkamp, Speel, Haesevoets, Bot, Dinjens, Ramaekers, Hopman and Manni¹²

Studies show HPV infection to be associated with an increased risk (up to three times) of head and neck SCC, independent of exposure to alcohol or tobacco. Patient demographics shown to be significantly associated with increased detection of HPV in head and neck SCC includes male gender, minimal tobacco and alcohol consumption, and a history of oro-genital sex.Reference Mork, Lie, Glattre, Hallmans, Jellum, Koskela, Møller, Pukkala, Schiller, Youngman, Lehtinen and Dillner^13, Reference Ringstrom, Peters, Hasegawa, Posner, Liu and Kelsey¹⁴

Gillison et al. conducted a case–control study to identify independent risk factors for head and neck SCC in a HPV-positive and a HPV-negative group.Reference Gillison, D'Souza, Westra, Sugar, Xiao, Begum and Viscidi¹⁵ They noted that an increased number of oral sexual partners and increased frequency of marijuana use were associated with HPV positivity. On the other hand, HPV-negative patients were noted to have a strong history of poor oral hygiene and alcohol and tobacco use. This suggests that patients who are exposed to HPV are probably exposed to many different potential risk factors, and thus make up a unique patient cohort.

Previous studies had suggested that HPV-positive cancer has a better response to treatment and also a better survival advantage. Fakhry et al. conducted a prospective study of 96 patients with stage III–IV head and neck SCC, to determine whether there was a difference between HPV-positive and HPV-negative cancers.Reference Fakhry, Westra, Li, Cmelak, Ridge and Pinto⁷ They noted that patients who were HPV-positive were more likely to be Caucasian, have a <20 pack year smoking history, be diagnosed with tonsillar carcinoma, have basaloid, poorly differentiated SCC, and have a higher Eastern Cooperative Oncology Group performance status. The response rate to induction chemotherapy was 82 per cent in HPV-positive tumours, versus 55 per cent in HPV-negative tumours. The response to chemo-radiotherapy was 84 per cent in HPV-positive tumours, versus 57 per cent in HPV-negative tumours. The survival rate for HPV-negative cancer patients was 90 per cent at one year and 62 per cent at two years; by comparison, the survival rate for HPV-positive cancer patients was 97 per cent at one year and 95 per cent at two years. After performing statistical analysis, these authors concluded HPV status was independently associated with mortality risk, and that patients with HPV-positive tumours had a 64 per cent lower risk of death and a 73 per cent lower risk of progression, compared with patients with HPV-negative tumours.

Virus identification

There is significant research being done in the area of HPV detection in oral specimens. Stratifying patients by HPV status may be necessary if the optimal treatment differs for these groups. Oral HPV does not produce as strong a result with polymerase chain reaction technology as do cervical specimens, and therefore a slightly different approach may be needed, compared with Papanicolaou smear collection.Reference Giovannelli, Campisi, Lama, Giambalvo, Osborn and Margiotta¹⁶ Research to date suggests that exfoliated superficial cells obtained by use of a mouth wash probably constitute the most suitable sample specimen. This method is non-invasive and yields large numbers of cells. Polymerase chain reaction technology currently produces the highest rates of HPV identification.Reference Chaudhary, Singh, Sundaram and Mehrotra^11, Reference Ha, Pai, Westra, Gillison, Tong and Sidransky¹⁷

Potential vaccination

The body mounts an immune response against HPV by initially recruiting T-lymphocytes (cluster of differentiation 4+ glycoprotein), which attract and activate B-lymphocytes, which ultimately produce antibodies against the HPV capsid proteins. However, these capsid proteins are not expressed in large numbers, so a vaccine which targeted non-structural proteins would be more efficacious. The E6 and E7 HPV DNA fragments are involved in the induction and maintenance of cellular transformation, and allow uncontrolled cell growth and possible malignant transformation. These proteins might be a more suitable target for an immunotherapeutic vaccine. Vaccines could be both therapeutic and prophylactic.Reference Devaraj, Gillison and Wu¹⁸

The vaccines currently licensed for use in pre-adolescent girls target HPV 16 and 18. Human papilloma virus 16 occurs in approximately 90 per cent of oropharyngeal cancers, whereas HPV 16 and 18 combined occur in approximately 75 per cent of cervical cancers. Oropharyngeal cancers are more prevalent in young males, who also may contribute significantly to the spread of genital HPV.

Human papilloma virus is the most common sexually transmitted infection worldwide, and is often asymptomatic. It is believed that the risk of HPV transmission is associated with the number of oral sex or open mouth kissing partners.Reference D'Souza, Agrawal, Halpern, Bodison and Gillison¹⁹ The prevalence of oral sex is thought to be increasing in all age groups, but most especially adolescents, who perceive it to be less risky. Therefore, it is hoped that these vaccinations might be offered to all pre-adolescent children, both male and female. This would presumably have a large impact on the future incidence of genital and oropharyngeal carcinomas, which generally affect young, otherwise healthy individuals.Reference Gillison²⁰

The ideal vaccine to prevent HPV infection would probably consist of virus-like proteins. These proteins are formed by a capsid protein, L1, which spontaneously forms empty capsids. A prophylactic vaccine needs to produce long-lasting antibodies that work at an immunoglobulin (Ig) A mucosal and IgG systemic level. This would not be sufficient to treat established infection or neoplasia, as it would not stimulate a strong enough immune response. A therapeutic vaccine would be more efficacious if aimed at the E6 and E7 DNA fragments, which are fundamental to neoplastic transformation.Reference Devaraj, Gillison and Wu^18, Reference Yan, Reichenbach, Corbitt, Hokey, Ramanathan and McKinney^21, Reference Lee, Anderson, Wu and Lee²² Ideally, of course, a combined multivalent vaccination with therapeutic and prophylactic elements would be most useful.Reference Devaraj, Gillison and Wu¹⁸

At present, there are two prophylactic HPV vaccines available: the bivalent Cervarix^® vaccine (Merck) and the quadrivalent Gardasil^® vaccine (GlaxoSmithKline). Gardasil has been approved by the US Food and Drug Administration. Gardasil protects against HPV 6, 11, 16 and 18. Merck, which initially produced the vaccine, conducted a randomised, double-blinded, placebo-controlled trial to determine Gardasil's efficacy. This had to be abandoned on ethical grounds, and the vaccine was subsequently offered to the placebo group. Both vaccines are composed of the L1 capsid protein and produce neutralising antibodies. Both vaccines need to be administered on three occasions over a six-month period. Early vaccination confers 90–100 per cent protection against HPV infection. Further studies are still needed to determine if long term protection is afforded. It is estimated that it would cost the Irish government €9.7 million to vaccinate all 12-year-old girls in Ireland; a national vaccination scheme is currently on hold due to financial difficulties in the Irish health service.

At Johns Hopkins University, the otolaryngologist and head and neck surgeon Sara Pai has been investigating the benefits of vaccinating patients with oropharyngeal cancers, including cancers of the tonsils, soft palate, posterior pharynx and tongue base. Dr Pai and colleagues have developed a vaccine for HPV-related head and neck cancers, with a clinical trial expected in the summer of 2009. Unlike the preventive HPV vaccine for cervical cancer, Pai's vaccine would be available to women and men with HPV who are already being treated for head and neck cancer.