Relationship Between Human Papillomavirus and Penile Cancer

Implications for Prevention and Treatment

Laura C. Kidd; Sharon Chaing; Juan Chipollini; Anna R. Giuliano; Philippe E. Spiess; Pranav Sharma


Transl Androl Urol. 2017;6(5):791-802. 

In This Article

HPV Pathogenesis

Population Infection Rates

HPV infection has been implicated in many cancers, including cervical,[7] penile[1] and oropharyngeal malignancies.[8] However, most HPV infections do not develop into pathogenic external lesions, and the majority are immunologically cleared by 12 months.[9] Yet, HPV DNA is detected in more than 90% of cervical tumor cells,[10] 68% of tonsillar tumor cells[11] and an estimated more than 20% of penile tumor cells (although this value varies depending on the literature).[12] The variations in prevalence of HPV-positive penile cancer may be due to differences in sampling methods, genital sites sampled (e.g., glans, shaft, or scrotum), molecular testing, and populations studied. For instance, a higher estimate of HPV prevalence can be found when samples are collected from a larger number of anatomic sites.[13,14] Higher estimates can also result from the use of more sensitive sampling techniques, such as a pre-wetted Dacron swab versus a cytobrush or urine sample.[14]

It appears that there may be gender disparities in HPV infections. The prevalence of HPV infection in women has been estimated to be around 11.7% with specifically higher prevalence seen in sub-Saharan Africa, Eastern Europe, and Latin America.[15] The prevalence of HPV infection in men is more variable, ranging from 1.3% to 72.9%. However, most studies report estimates greater than 20% in men, with higher prevalence in uncircumcised versus circumcised men.[12,16]

Both genital[17–19] and oral[20] HPV prevalence have been shown to be higher in men than in women. Additionally, there does not seem to be an association between age and HPV prevalence in men. In contrast, HPV prevalence in women is highest among 14 to 24 years old, which then decreases until middle age.[18] These differences indicate that HPV infection and clearance may differ by gender. In fact, it has been shown that men take significantly longer to clear an oral HPV infection than women (5.3 vs. 3.0 months; P<.001).[21] Additionally, in serologic studies women have been shown to have a higher prevalence of HPV antibodies in comparison to men, regardless of age.[22,23] In another study, women were shown to have a higher prevalence of antibodies for one or more of HPV types 6, 11, 16 and 18 than men (32.5% vs. 12.2%).[22] Thus, evidence suggests that men do not exhibit as robust an immune response to HPV infection as women do. In the general population, the penile epithelium of asymptomatic men has consistently higher prevalence of HPV infection than the cervix of women with normal cytological testing.[24] It is possible that these gender differences occur because HPV infection of keratinized epithelium does not induce an immune response as effectively as that of mucosal epithelium, the tissue type most commonly infected in women.

Penile cancer has been considered a relatively rare malignancy in the western world. In fact, between 1973 and 2002, the overall incidence of primary malignant penile cancer had decreased in the United States.[3] Additionally, between 1943 and 1990 there was a statistically significant decline in the overall rate of penile cancer in Denmark.[25] However, recent reports indicate an increase in incidence rates in various countries. For example, the Netherlands had an increase from 1.4 to 1.6 per 10,000 person-years from 1989 to 2006.[26] In Denmark, despite older studies showing a decline in penile cancer incidence as noted above, there was actually an increase from 1.0 to 1.3 per 100,000 men-years more recently from 1978 to 2008.[27] More recently, England had an increase from 1.10 to 1.33 per 100,000 men from 1979 to 2009.[28]

This increase in incidence of SCC penile cancer observed in England, Denmark and the Netherlands mirror increases in incidence of HPV-positive oropharynx cancer rates seen in the U.S., Sweden and Finland. HPV infection is a risk factor for both oropharyngeal cancer and penile cancer. From 1988 to 2004 in the U.S., the prevalence of HPV-positive SCC oropharynx cancer (SCCOP) has increased 225%, while HPV-negative SCCOP fell by 50% from 1988 to 2004.[8] The drop in HPV-negative SCCOP incidence correlates with decreased smoking, while the rise in HPV-positive SCCOP suggests increased oral HPV exposure.[29] From 1970 to 2002 in Sweden, there has been a 2.8 fold increase in the incidence of tonsillar cancer and a significant increase (2.9-fold; P<0.001) in the proportion of HPV-positive tonsillar cancer.[11] From 1956 to 2000 in Finland, the age standardized incidence rate of tonsillar cancer doubled.[30]

Interestingly, treatment with radiation of HPV-positive SCCOP had more pronounced improvements in 2-year survival rates than HPV-negative SCCOP.[31] In another study investigating the viral load of HPV-16 in tonsillar cancer, six patients with tumors with greater than or equal to the median value of 190 HPV-16 copies per B-actin had significantly better survival than those with lesser than or equal to 60 HPV-16 copies/B-actin (P=0.039, log rank test) with the higher viral load group tumor free 3 years after diagnosis versus 2/5 tumor free in the lower viral load group (P=0.026, ×2 test).[32] These results seem to indicate that HPV-presence may influence disease outcome however, more research is needed in this line of inquiry.

HPV Serotypes

HPV is a DNA virus with more than 100 different known genotypes which can affect both cutaneous and mucosal sites. There are 20 types which are known to infect the genital tract.[33] These types are generally classified as "low-risk" or "high-risk" depending on their association with cervical malignancy. HPV types 16, 18, 31, 33, 45, 56, and 65 are considered high-risk and are associated with penile malignancies, whereas types 6 and 11 are found in benign lesions (e.g., condyloma acuminata, or genital warts) and are considered low risk for malignant transformation. In patients with HPV, coinfection with more than one HPV type is common.[34]

High-risk HPV-16 and HPV-18 are particularly implicated in genital malignancies. In cervical cancer, HPV-16 was present in more than 50% of cases and HPV-18 in another 11%.[10] In penile cancer, HPV-16 or HPV-18 were implicated in approximately 31% of penile cancers, with HPV-16 being the predominant subtype.[35] HPV-16 DNA is also a risk factor for developing tonsillar cancer and causes the majority of oropharyngeal SCC in the United States.[8,36,37]

HPV type influences the rate of progression from genital HPV infection to disease.[38] In one study, 16% of genital HPV-6 infections in men developed into HPV-6-positive condyloma and 22% of genital HPV-11 infections in men developed into HPV-11-positive condyloma. Both types of infections had rapid rates of progression to disease after initial genital infection with a median of 7.7 months. However, only 2% of genital HPV-16 developed into penile intraepithelial neoplasia (PeIN) within a 24-month period. HPV-16 infection had a relatively slow rate of progression to disease, with 50% taking more than 19 months for PeIN to be detected.[39] PeIN is considered to be a precursor lesion of invasive SCC and is defined as a change of the penile squamous epithelium indicated by dysplastic changes with an intact basement membrane.[34]

The role of HPV as a prognostic factor in penile cancer remains unclear. It is uncertain whether cancers involving HPV infection have better survival profiles than cancers without HPV infection. In a study with 82 penile cancer patients, 30.5% of tumors had HPV DNA, with HPV-16 being the most prevalent. This study demonstrated no association between HPV negative and positive patients when considering lymph node metastasis (P=0.386) and 10-year survival rate (68.4% vs. 69.1%; P=0.83).[40] In another study with 29 patients with invasive penile SCC, 31% of tumors had either HPV-16 or HPV-18 DNA. This study found no difference between HPV negative and positive patients in terms of nodal metastasis or survival even after adjustment controlling for tumor stage.[35] However, these results differ from another study which examined HPV status as a prognostic indicator in 171 penile cancer patients. In this study, high-risk HPV DNA was found in 29% of tumors, with 76% containing HPV-16; high-risk HPV was associated with improved 5-year disease-specific survival (78% vs. 93%; P=0.03). Additionally, high-risk HPV was an independent predictor of disease specific mortality in multivariate analysis [hazard ratio (HR), 0.14; 95% CI, 0.03–0.63; P=0.01].[41] It is possible that these conflicting results may be due to differing study designs, sample sizes and sampling methods for HPV DNA, especially since there has yet to be a universal protocol for HPV DNA testing of penile SCC tumors. The prognostic role of HPV in the literature remains unclear. A summary of the available studies is shown in Table 1.

HPV Pathogenesis/Carcinogenesis

HPV-mediated pathogenesis of human epithelial cells is a multistep process. Although the HPV-mediated pathogenesis of cervical squamous cell cancer is well understood, it is not as well understood in penile cancer. Penile cancers are predominantly squamous cell in origin. However, there are several subtypes, including basaloid, warty, keratinizing, and verrucous. HPV infection has been associated with basaloid and warty subtypes and less so with other subtypes. A study found HPV DNA in 42% of penile carcinomas, 90% of dysplasias and 100% of condylomas. Additionally, basaloid and warty SCC subtypes had 80% and 100% of HPV DNA, respectively, whereas keratinizing and verrucous subtypes had the lowest percentage of HPV DNA, at 34.9% and 33.3%, respectively.[48]

HPV affects the squamous epithelium in two ways, either as a viral infection or as a viral-associated precancerous lesion. A HPV viral infection is largely transient and occurs when the squamous epithelium supports virion production and develops into a morphologic low-grade lesion (e.g., condyloma and mild dysplasia). In contrast, an HPV viral-associated precancerous lesion (e.g., HPV-associated PeIN) occurs when the viral genome integrates into the host genome, leading to virally-induced overexpression of oncogenes that drive cell proliferation, which can develop into malignant transformation.[34] In HPV-16 mediated penile cancer, integration occurs at the chromosomal 8q21.3 locus (FAM92A1 gene) and at the 16p13.3 locus (TRAP1 gene). This HPV integration marks an end point in the clonal selection of cells, resulting in functional genes that are altered. This allows for uncontrolled growth and dedifferentiation which can progress into invasive cancer.[24]

HPV is a double-stranded DNA virus that has a circular genome which encodes eight genes, including E5, E6 and E7 oncogenes. However, only the E6 and E7 oncogenes are necessary for malignant transformation and maintenance of malignant phenotype of the host cells.[34] The activation of viral E5 oncogene is not necessary for malignant transformation. However, it may contribute to carcinogenesis through manipulating viral uptake of host target cells, leading to progression of dysplasia related to premalignant lesions. The E5 gene product is a transmembrane protein that regulates activation of epidermal growth factor receptor (EGFR); EGFR upregulation leads to a decrease in E-cadherin expression. This decrease, combined with an associated increase in MMP-9, results in a reduction of cell-to-cell adhesion.[49]

Viral oncogenes E6 and E7 are actively transcribed in HPV-infected cells. E6 and E7 contribute carcinogenesis by disrupting centrosome synthesis, which is central to mitosis. Thus, a hallmark feature of both HPV-mediated premalignant lesions and malignant tumors is the development of multipolar mitosis. Additionally, viral E6 oncoprotein targets p53 tumor suppressor protein whereas viral E7 oncoprotein primarily targets retinoblastoma-1 (RB1) tumor suppressor protein.[50] These oncoproteins inactivate their respective target tumor suppressor proteins through binding. Since the p53 and RB1 are negative regulators of cellular proliferation, the inactivation leads to uncontrolled cellular growth. In HPV-negative penile, head and neck squamous cell carcinomas (HNSCC), p53 is mutated and may be associated with lymph node metastases.[51,52] However, wild-type p53 function is crucial to the development of cervical tumors, but p53 mutations frequently occur later in the metastatic progression of HPV-positive primary cervical cancers.[53] In HPV-positive tonsillar cancer, HPV-16 oncogenes E6 and E7 oncogenes are generally expressed.[36,37] In the matter of HPV-mediated cervical cancer pathogenesis, it is known that HPV and chromosomal recombination is frequent and required for progression to cancer; DNA hypermethylation of the L1 gene is established as a biomarker for cancer progression. Some studies have shown that there are high rates of HPV-16 methylation in penile cancer similar to those seen in cervical cancers.[54]

In high-risk HPVs, viral E7 oncoprotein binds with Rb tumor-suppressor protein with much higher affinity than E7 produced by low-risk HPVs, such as HPV-6 and HPV-11. One of the major functions of Rb is to bind and inhibit transcription factors of the E2F-family, which leads to the repression of transcription of genes with products involved in DNA and chromosomal replication.[55] Viral E7 oncoprotein interferes with the Rb and E2F interaction, leading to the release of E2F factors in their transcriptionally active forms, resulting in autonomous cell proliferation without G1 cell-cycle stops.[56] This allows for cyclin-dependent kinase inhibitor p16ink4a to accumulating in the nucleus, which inhibits G1 cyclin-dependent kinase 4 (CDKN4), CDKN6, and cyclin-D dependent kinases, thus initiating phosphorylation (inactivation) of the Rb tumor-suppressor protein. Hence, in high-risk HPV transcriptionally-active infection, p16(INK4a) overexpression can serve as a surrogate immunohistochemical marker.[57,58] However, this inactivation of RB1 and enhancement of p16 expression can occur both in HPV-associated tumors and in HPV-negative tumors that have mutations leading to RB1 loss.[58]