International Reviews of Immunology, Early Online:1–19, 2014 C Informa Healthcare USA, Inc. Copyright ISSN: 0883-0185 print / 1563-5244 online DOI: 10.3109/08830185.2014.911857
REVIEW
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
The Role of Human Papilloma Virus (HPV) Infection in Non-Anogenital Cancer and the Promise of Immunotherapy: A Review Chris Cobos,1,∗ Jose´ A. Figueroa,1,2,3,∗ Leonardo Mirandola,1,2,∗ Michela Colombo,1 Gabby Summers,1 Alejandro Figueroa,1 Amardeep Aulakh,1 Venu Konala,1 Rashmi Verma,1 Jehanzeb Riaz,1 Raymond Wade,1 Charles Saadeh,1 Rakhshanda L. Rahman,4 Apurva Pandey,1 Saba Radhi,1 Diane D. Nguyen,1,3 Marjorie Jenkins,2,† Maurizio Chiriva-Internati,1,2,3,† and Everardo Cobos1,2,3,† 1
Department of Internal Medicine at the Division of Hematology & Oncology, Texas Tech University Health Sciences Center and Southwest Cancer Treatment and Research Center, Lubbock, TX, USA; 2 Laura W. Bush Institute for Women’s Health and Center for Women’s Health and Gender-Based Medicine, Amarillo, TX, USA; 3 Kiromic LLC, Lubbock, TX, USA; 4 Division of Surgical Oncology, Texas Tech University Health Sciences Center Amarillo, TX, USA
Over the past 30 years, human papilloma virus (HPV) has been shown to play a role in the development of various cancers. Most notably, HPV has been linked to malignant progression in neoplasms of the anogenital region. However, high-risk HPV has also been suggested to play a significant role in the development of cancers in other anatomic locations, such as the head and neck, lung, breast and bladder. In 2006, the first vaccine for HPV, Gardasil, was approved for the prevention of subtypes 6, 11, 16 and 18. A few years later, Cevarix was approved for the prevention of subtypes 16 and 18, the HPV subtypes most frequently implicated in malignant progression. Although increased awareness and vaccination could drastically decrease the incidence of HPV-positive cancers, these approaches do not benefit patients who have already contracted HPV and developed cancer as a result. For this reason, researchers need to continue developing treatment modalities, such as targeted immunotherapies, for HPV-positive lesions. Here, we review the potential evidence linking HPV infection with the development of non-anogenital cancers and the potential role of immunotherapy in the prevention and eradication of HPV infection and its oncogenic sequela. Keywords: anogenital cancer, cervical cancer, Gardasil, head and neck squamous cell carcinoma, vevarix
Accepted 1 April 2014. ∗ Equal first author contributors † Equal senior author contributors Address correspondence to Everardo Cobos, MD, Chief of the Division of Hematology & Oncology, Professor, Department of Internal Medicine, School of Medicine, Texas Tech University Health Sciences Center, 3601 4th St., Mail Stop 9410, Lubbock, USA. E-mail: [email protected]
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
INTRODUCTION According to data from the National Program of Cancer Registries (NPCR) and the Surveillance, Epidemiology, and End Results (SEER) program, an average of 33 369 human papilloma virus (HPV)-associated cancers are diagnosed annually, including 12 080 among males (8.1 per 100,000) and 21 290 among females (13.2 per 100,000) [1, 2]. HPV is thought to be responsible for 7–8% of all human malignancies and is associated with 96% of cervical cancers, 93% of anal cancers, 64% vaginal cancers, 51% of vulvar cancer, 36% of penile cancers and 63% of oropharyngeal carcinomas [1, 3, 4] (Table 1). HPV belongs to a family of non-enveloped DNA viruses that typically infect skin or mucosal epithelium, particularly in squamous epithelial cells. Commonly infected surfaces include the skin, vagina, transitional zone of the cervix, anus, vulva, glans penis, mouth, throat, trachea, bronchi and lungs [5]. HPV’s association with malignancy has been well established for more than 30 years. In 1972, Stefania Jablonska first demonstrated an association of HPV with a type of skin cancer called epidermodysplasia verruciformis [6]. Soon after this discovery, Hearald Zur Hausen published his work linking HPV with cervical cancer, and, with Lutz Gissmann in 1977, isolated the HPV-6 DNA from genital warts [6]. In 1983, Zur Hausen implicated HPV-16 and -18 as key etiologic factors for the development of cervical cancer [6]. The main mode of HPV transmission is direct skin contact. The infection may occur via skin-to-skin contact during sexual intercourse, oral–genital, or hand–genital contact. Transmission from an infected mother to a newborn’s trachea is possible during delivery [2]. In the majority of cases (approximately 90%), the human immune system opsonizes the virus and clears it from the system within 2 years [7, 8]. Thus, the vast majority of infected individuals remain asymptomatic and develop no clinical sequelae. In symptomatic patients, infection with low-risk HPV subtypes can lead to the formation of genital warts and/or recurrent respiratory papillomatosis (RRP) within weeks or months after contact with an HPV carrier [2, 9]. Visible genital warts can be treated with prescribed medication, and RRP can be treated either by medication or surgical removal [10, 11]. Non-genital forms can also manifest as warts on other anatomic sites such as hands and feet. For women, a regular screening papinicolaou (PAP) smear test can detect early epithelial changes due to HPV infection allowing for early intervention and prevention of malignant progression [5]. Recently, the U.S. Food and Drug Administration (FDA) has approved two vaccines for the prevention of HPV infection: Gardasil (Merck and Co.) and Cervarix (GlaxoSmithKline). Gardasil was approved for use in the prevention of genital cancers and precancerous lesions in women [12]. In recent years, the rate of administered HPV vaccinations has steadily risen, which is expected to result in the decline of HPV-related cervical and anal cancers and deaths associated with these diseases in high-risk populations. Unfortunately, these vaccines will not eliminate existing HPV infections, and TABLE 1.
Percentage of HPV-associated urogenital cancer.
Type of cancer
HPV association from 2004 to 2008 72
HPV-positive DNA cancers from 1999 to 200571
HPV 16/18% from 1999 to 200571
Anal Cervical Vulvar Vaginal Penile Oropharyngeal
93% 96% 51% 64% 36% 63%
91% 90% 69% 75% 63% 72%
79% 66% 55% 49% 48% 62% International Reviews of Immunology
HPV in Human Cancer and Immunotherapy
public awareness of the importance of this intervention is critical in order to prevent future HPV infections and related malignancies in women [13]. For men, there is no established screening test for HPV, and the diagnosis of infection relies on visual inspection and physical findings [14]. Further research is needed to provide appropriate screening and preventive options for males at risk for HPV infection.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
The Role of HPV in Cancer To date, more than 100 subtypes of HPV have been discovered. There are approximately 60 HPV strains that are known to be related to common body warts. Those strains infect the cutaneous epithelium leading to the formation of common skin warts. The other 40 HPV types are called mucosal or genital-type HPV strains as they infect the orogenital mucosal epithelium [7]. The mucosal HPV subtypes infect moist, squamous cell surfaces of the body generally in the anogenital area. The mucosal subtypes are further classified as low-risk and high-risk. Low-risk types, including HPV-6 and HPV-11, cause cauliflower-shaped warts on or around the anogenital region of men and women. These lesions are known as condyloma acuminata and rarely transform into cancer. The high-risk types of HPV includes 16, 18, 31, 33, 35, 39, 45, 51, 52 and 58 and can eventually lead to cancer in both men and women. Therefore, high-risk HPV types can be life-threatening, and pre-screening is vital to detect these changes during early stages [2,7]. Studies have shown that the role of HPV infections in epithelial cancers has been associated with the expression of viral oncogenes E6 and E7 (Figure 1). These proteins transform cells by interacting with growth-regulating host cell proteins and are important in the pathogenesis of malignant disease. Particularly in cervical cancer, E6 and E7 play an instrumental role at the molecular level through their interaction with P53 and retinoblastoma (Rb) genes, respectively [10,11]. P53 is involved in cell-cycle regulation, DNA repair process and apoptosis, and represents the most commonly mutated gene in human malignancy [10]. The HPV E6 protein is able to bind P53 and targets this tumor suppressor for proteasomal degradation. This degradation of p53 results in unchecked cellular cycle progression and accumulation of chromosomal mutations due to impaired DNA checkpoint regulation. Furthermore, E6 association with p53 also causes inhibition of the Src family of nonreceptor tyrosine kinases, important in cytoplasmic signal transduction and the cell’s mitotic activity [15]. The Retinoblastoma (Rb) protein is one of the first-recognized tumor suppressor gene products that acts as a growth inhibitor by binding and inactivating the E2F general transcription factor. E2F controls DNA synthesis and cyclin function and promotes the entrance of the cells in the S phase of cell cycling. The HPV E7 protein binds Rb, causing E2F to be released, become activated and promote cell cycling. This process leads to DNA damage and genomic instability at the molecular level, giving rise to abnormal cell proliferation [9,10]. The E7 protein also interacts with wild-type p53, and this interaction further contributes to the immortality and subsequent transformation of the cells, through an anti-apoptotic affect [9, 10]. HPV in Head and Neck Cancer The first evidence of HPV’s relation to squamous cell carcinoma of the head and neck (HNSCC) was provided by Syrj¨anen KJ et al in 1983 [16]. It was later discovered that 40% of the laryngeal and oral cancers had morphologic similarities with HPV-infected lesions, and 20% of samples examined demonstrated HPV structural proteins, detected by immunohistochemistry [17]. Further studies have shown that HPV infection is associated with 23% to 35% of all HNSCC biopsies worldwide [18, 19]. The prevalence of HPV in HNSCC appears to vary according to ethnic and geographic location, C Informa Healthcare USA, Inc. Copyright
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
C. Cobos et al.
FIGURE 1. Molecular mechanisms of HPV-induced cell cycle dysregulation in cancer cells. Entry into the S phase from the G1 phase of cell cycle requires an active E2F elongation factor to transcribe S-phase genes, such as DNA polymerase. The protein, Rb, binds E2F inactivating its activity. The complex of CDK4 and cyclin D (CyD) phosphorylates E2F-bound Rb, producing a pRb–E2F complex. This event releases E2F, allowing the transition into the S phase. Then, the association with cyclin E activates CDK2, which in turn phosphorylates the pRb-producing pppRb (tri-phosphorylated Rb), which is then degraded by the proteasome. The HPV protein, E6, blocks both pRb and p21, which in turn blocks the CDK2–CyE complex, ultimately resulting in increased formation of pppRb. However, E6 blocks p53, which activates p21, resulting in an additional loss of regulation in the pRb degradation process. Moreover, blockade of p53 reduces cell sensitivity to apoptotic stimuli and the ability to repair damaged DNA.
tumor sub-site and HPV-detection methods [20]. The majority of HPV-positive HNSCC (up to 90%) originate in the oropharynx, and tend to arise mostly from the lingual and palatine tonsils. [20, 21]. The implicated HPV subtypes are similar to those in cervical cancer, with HPV-16 found in 68% to 87% of HPV-positive head and neck cancers worldwide. Less common HPV types also found in HNSCC include HPV-18, -31, -33, -35, -45, -51, -52, -56, -58, -59 and -68 [19]. In a case-control study of 100 patients with newly diagnosed oropharyngeal cancer, a high number of vaginal-sex partners (26 or more) during life was associated with oropharyngeal cancer (odds ratio [OR], 3.1; 95% confidence interval [CI], 1.5–6.5), as was a high lifetime number of oral-sex partners (6 or more; OR, 3.4; 95% CI, 1.3–8.8) [18]. The odds of acquiring an oral HPV infection were significantly increased in HIVpositive patients, compared with those who were HIV negative, after controlling for sexual behavior [22]. In this study, the use of marijuana was also found to be associated with HPV-16-positive HNSCC. This association was stronger with greater numbers of oral sex partners and increasing intensity and duration of marijuana use. However, tobacco, alcohol and poor oral hygiene were associated primarily with HPV-negative
International Reviews of Immunology
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
HPV in Human Cancer and Immunotherapy
HNSCC [23]. A defective immune system also appears to increase the risk of HPVpositive HNSCC and other HPV-associated malignancies [20]. Several differences exist between HPV-positive and -negative HNSCC. For example, patients with HPV-positive oropharyngeal tumors tend to be younger (40–60 years) as compared with HPV-negative patients (60 years or more), it is more common in men as compared with women (3:1 ratio) and is more prevalent in Caucasian patients with higher education and economic status [20, 23, 24]. HPV-positive HNSCC commonly present with poorly differentiated non-keratinizing, basaloid features compared with HPV-negative HNSCC, which presents with a more differentiated and keratinized morphology. HPV-positive HNSCC also commonly presents with more advanced TNM stage (III and IV), reflecting the presence of more advanced lymph node metastases [19, 20, 25]. Interestingly, several studies have shown that locally advanced HPV-positive HNSCC treated with radiation and chemotherapy has a better outcome in terms of survival, response to treatment, and rates of local recurrence, when compared with HPVnegative cancers. In the Radiation Therapy Oncology Group (RTOG) 0129 trial, 743 patients with stages III and IV HNSCC were retrospectively analyzed and 63.8%, of patients with oropharyngeal cancer were found to have HPV-positive tumors. The results of this study showed a better 3-year overall survival (82.4%, vs. 57.1% P < 0.001) for HPV-positive compared with HPV-negative cancers [26]. Another retrospective analysis done on an HPV-positive (detected by p16 overexpression) cohort of patients with oropharyngeal cancer of RTOG study 02.02, also demonstrated an improved 2year OS and failure-free survival [27]. Patients with HPV-positive oropharyngeal carcinoma in the TAX 324 trial also had better 5-year OS (82% vs 35%), and PFS (78% vs 28%) [28] than their HPV-negative counterparts. In another retrospective study, survival of HPV-positive and HPV-negative patients was compared in 271 oropharyngeal cancers (1984–2004) collected by the SEER program. The median survival was significantly longer for HPV-positive than for HPV-negative patients, and it persisted beyond 15 years after cancer diagnosis (131 vs 20 months; log-rank p < 0.001). In addition, a 69% reduction in the risk of death was observed among HPV-positive cancers. Survival was better among HPV-positive patients who were treated by radiation compared with those who were not [29]. A study by Fakhry and coworkers prospectively evaluated the association of tumor HPV status with therapeutic response and survival among 96 patients with stage III or IV HNSCC of the oropharynx or larynx [24]. In this Eastern Cooperative Oncology Group (ECOG) phase II study, patients received two cycles of induction followed by concomitant chemoradiation therapy. Patients with HPV-positive tumors had higher response rates after induction chemotherapy (82% vs 55%, p = 0.01) and chemoradiation treatment (84% vs 57, p = .007). After a median follow-up of 39.1 months, patients with HPV-positive tumors had improved overall survival (2-year overall survival = 95% vs 62%, p = 0.005) [24]. The rate of progression and locoregional disease, but not distant metastasis, was also significantly lower for patients with HPV-positive than HPV-negative tumors [24]. Another important difference between HPV-positive and HPV-negative HNSCC is the observation that the cumulative incidence of second primary tumors appears to be significantly lower among patients with HPV-positive tumors [26, 27]. Interestingly, a recent meta-analysis has found no difference in survival between HPV-positive and HPV-negative HNSCC of non-oropharyngeal origin [29]. Smoking has also been associated with increased mortality and cancer relapse in both HPV-positive and HPVnegative cancers, and this increment is estimated to be approximately 1% for each additional pack-year of tobacco smoking [26]. C Informa Healthcare USA, Inc. Copyright
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
HPV in Less-Studied Malignancies HPV has been well studied in urogenital as well as head and neck malignancies, but it has also been implicated in many other types of cancers. Here, we discuss the available evidence linking other types of malignancies to HPV infection. These data are summarized in Table 2. HPV in Lung Cancer Lung cancer is the leading cause of cancer-related mortality worldwide. HPV was first implicated in the etiology of lung cancer in 1979 and, over the past 30 years, many studies have tried to confirm this relationship [30–32]. There is increasing evidence that high-risk HPV might be associated with tumorigenesis and progression of nonsmall cell lung cancer (NSCLC) [33–39]. It has been proposed the route of transmission could be either through mucosal barrier–blood–lung tissue infection or through oral cavity–larynx–lung tissue during sexual contact [40, 41]. A recent study by Munoz et al has also proposed a possible relationship between tobacco carcinogens and HPV16 E6/E7 oncoproteins, their interaction leading to malignant transformation and tumorigenesis of lung epithelial cells [41]. However, in a case-control study Koshiol et al selected 450 Italian patients with lung cancer and discovered that none of the specimens were positive for HPV subtypes. This study concluded that there was no evidence to support a relationship between HPV and lung cancer in this particular representative sample of a Western population [42]. A recent meta-analysis by Syrjanen suggested these discrepancies in HPV prevalence in patients with lung cancer were likely due to differences in geographic location and histologic subtype, rather than the technique used for HPV isolation [43]. Therefore, the definitive role of HPV infection in lung cancer remains unclear, and prospective cohort studies are needed to confirm or disprove this association. HPV in Esophageal Cancer It was estimated that 17 460 subjects (13 950 men and 3510 women) would be diagnosed with esophageal cancer in 2012, and 15 070 men and women would die of cancer [44]. It is well recognized that the incidence of esophageal carcinoma is increasing due to a higher prevalence of chronic gastroesophageal reflux disease (GERD) and Barrett’s esophagus. Nonetheless, it is conceivable that HPV could be implicated in the etiology of esophageal squamous cell cancer (ESCC), because oral squamous epithelium and the epithelium of the upper esophagus are histologically similar. Two large studies were done in China in 2000 and 2006 to explore this hypothesis. In the first study, 700 esophageal carcinomas were analyzed for HPV DNA sequences, which was found to be present in 17% of the samples examined. High-risk HPV-16 and HPV-18 strains were found to be present in 27% of the cases studied [45]. In the second study, 702 patients underwent endoscopy and cytologic evaluation of biopsy specimens and demonstrated HPV positivity in 8%, 7% and 16% of subjects with mild, moderate and severe dysplasia, respectively [46]. Recently, a study by Yahyapour et al used archived tissue blocks from 177 patients with esophageal squamous cell carcinoma (ESCC) and evaluated them for the presence of HPV DNA by PCR. HPV was detected in 49 patients (28%), of which 25 were positive for at least one high and one low-risk HPV genotypes [47]. Another study by Gupta et al identified HPV DNA in 17 of 37 (46%) cases of nonkeratinizing SCC of the esophagus [48]. A review by Liyanage et al identified 139 studies with results indicating higher levels of HPV detection in areas with a high-risk of ESCC, than in areas with a lower risk of contracting the disease [49]. These investigators concluded these inconsistencies were due to variations in technique, study design and sample types [49]. Another study using 1500 serum samples from patients with ESCC found a significant association International Reviews of Immunology
Not definitively established
Not definitively established
No association
Not definitively established
Not definitively established
Not definitively established
Lung
Esophageal
Prostate
Bladder
Breast
Skin
Role of HPV
90% in immunocompromised and 50% in immunocompetent patient [111].
0–61% [77,78, 80–82, 110]
Meta- analysis 16%–19.4% (range from 3%–35%) [68,69, 72–75]
—
7–46% [45,46, 48]
22.4% [73]
Incidence of HPV in studies
Summary of HPV role in other forms of cancer.
Type of cancer
TABLE 2.
Disruption of DNA repair activity by UV radiation along with ubiquitously present cutaneous HPV types increases the risk of NMSC
Only one study demonstrated Id-1 is an important target for E6/E7 oncoproteins of HPV type 16, 18, 31, 33 and 35 in breast cancer cells.2
Epithelial tropism of HPV and the anatomic proximity of the urogenital zone1
—
Mucosal barrier–blood–lung tissue infection or through oral cavity-larynx-lung tissue during sexual contact. Oral squamous epithelium and upper esophagus similar histologically.
Possible mechanism
Summary High-risk HPV might be associated with the tumorigenesis of NSCLC mainly in Asian nonsmokers. No association established in Western countries. High-risk HPV has 27% incidence in Esophageal cancer but pathogenesis remains inconclusive. Pathogenesis questionable since it is adenocarcinoma doesn’t occur by direct contact with the exposure Meta-analysis points out to an association with the incidence of HPV infection from bladder infection, no clear evidence and also needs to subcategorize the data based on subtype of bladder cancer No association was found in multiple studies and further all definite associations with HPV are squamous cell cancers and breast cancer is exclusively adenocarcinoma. Beta genus HPV might be co-carcinogen with UV radiation, but a causal relationship needs to be established.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
with the presence of HPV-16 E6, HPV-33 L1, HPV-6 E6, HPV-6 L1 and HPV-11 L1 [50]. In this study, no definite conclusion could be reached regarding HPV association in the oncogenesis of ESCC, although an association was suggested. Based on the available information, the relationship between HPV and the etiology of ESCC remains to be determined.
HPV in Prostate Cancer The role of HPV in the pathogenesis of prostate cancer is a topic of debate largely fueled by the notion that increased and/or risky sexual behavior and more frequent exposure to sexually transmitted diseases (STDs) may play a role in the etiology of this malignancy [51, 52]. Although a clear association between prostate cancer and HPV has not been supported [53], the presence of oncogenic viruses, including high-risk HPV types, have been documented in normal and neoplastic prostate tissues. In a recent study, Whittaker and collaborators detected both HPV-18 and Epstein–Barr Virus (EBV) DNA sequences in normal, benign and cancerous prostate specimens [54]. In this study, 24% of prostate cancer specimens demonstrated the presence of koilocytes, an indication of HPV-induced cellular infection. These findings suggest HPV infection may be a ubiquitous phenomenon in prostate tissues with possible oncogenic potential. However, as HPV carcinogenesis depends on the expression of viral oncogenes which inhibit Rb and p53 function, the observation that these tumor suppressor gene products are inactivated by mutations and/or deletions in prostate cancer suggest a mechanism of oncogenic transformation independent of HPV effects [53]. Based on the available evidence, the etiologic relation of HPV and prostate cancer remains uncertain.
HPV in Bladder and Urothelial Cancer Several studies have demonstrated the presence of HPV DNA in carcinomas of the bladder, suggesting HPV might be associated with the development of this malignancy [55–66]. However, a definitive association between HPV and bladder cancer still remains controversial [67–69]. For example, Noel and co-investigators analyzed 75 cases of transitional cell carcinoma of the bladder for the expression of DNA for HPV subtypes 6, 11, 16, 18 and 33 and identified the virus in only two cases (2.7%) [70]. Ben Selma et al analyzed 125 archival specimens of bladder carcinoma (119 urothelial carcinomas, five squamous carcinomas, and one adenocarcinoma) by PCR and did not detect anogenital-type HPV DNA in any of the specimens [71]. However, Barghi et al used PCR to identify HPV DNA in bladder tissue specimens from 59 patients with bladder cancer and 20 controls, and found HPV DNA was present in 21 (35.6%) of the bladder cancer cases and 1 (5%) of the control group (p = 0.008) [72]. In this study, the HPV subtype 18 was isolated from HPV-positive bladder cancer specimens in 81% of the subjects. A meta-analysis of 39 studies conducted by Gutierrez et al in 2006 found an association of HPV infection and bladder cancer in 16% of the studies analyzed [73]. In a different meta-analysis by Pacheco et al including 38 publications, HPV DNA was detected in 19.4% of bladder cancer cases [74]. The pooled OR in this study was 3.2 (95% CI 1.19–8.60, p = 0.02) indicating a significant relationship between HPV and bladder cancer [74]. More recent meta-analyses have supported an association between HPV infection and bladder cancer. Li et al used data from 52 studies to calculate a prevalence of HPV infection in bladder cancer of 16.88 [75] and, in 2012, a meta-analysis of 21 observational, case–controlled studies showed a pooled OR of 2.13 for the presence of HPV DNA in bladder cancer specimens (95% confidence interval [CI] 1.54–2.95) [76]. Despite the lack of prospective data, these International Reviews of Immunology
HPV in Human Cancer and Immunotherapy
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
large pooled studies suggest the possibility that HPV infection may be linked to the development of bladder cancer in a percentage of patients. HPV in Breast Cancer The possible association of HPV with breast cancer has been studied extensively over the past years. In a study by Gopalkrishna et al, fine-needle aspirates from 26 patients with breast cancer and four breast tumor biopsies were analyzed for the presence of HPV-16 and -18 DNA sequences, by both PCR and Southern blot hybridization. None of the samples studied demonstrated the presence of the virus [77]. More recently, Hedau et al prospectively collected a total of 228 biopsies and 142 blood samples from 252 patients with breast cancer and tested them for HPV DNA sequences using conventional PCR. They were unable to detect HPV DNA in any of the biopsy or blood samples [78]. Another study by Silva et al used PCR to analyze 79 samples of invasive ductal cancer of the breast for the presence of HPV DNA subtypes 6, 11, 16 and 18, and none were found to be positive for viral DNA [79]. A recent study reported HPV expression in only four of the 70 specimens (5.7%) analyzed using in-situ hybridization (ISH), and in two of 70 specimens (2.9%) using in-situ PCR (IS-PCR) [80]. In another study, HPV DNA was found in 4.4% of 67 breast cancer specimens analyzed [81]. These earlier studies suggest that high-risk HPV types are either not associated with breast cancer, associated only with a very small proportion of breast cancer cases (
REVIEW
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
The Role of Human Papilloma Virus (HPV) Infection in Non-Anogenital Cancer and the Promise of Immunotherapy: A Review Chris Cobos,1,∗ Jose´ A. Figueroa,1,2,3,∗ Leonardo Mirandola,1,2,∗ Michela Colombo,1 Gabby Summers,1 Alejandro Figueroa,1 Amardeep Aulakh,1 Venu Konala,1 Rashmi Verma,1 Jehanzeb Riaz,1 Raymond Wade,1 Charles Saadeh,1 Rakhshanda L. Rahman,4 Apurva Pandey,1 Saba Radhi,1 Diane D. Nguyen,1,3 Marjorie Jenkins,2,† Maurizio Chiriva-Internati,1,2,3,† and Everardo Cobos1,2,3,† 1
Department of Internal Medicine at the Division of Hematology & Oncology, Texas Tech University Health Sciences Center and Southwest Cancer Treatment and Research Center, Lubbock, TX, USA; 2 Laura W. Bush Institute for Women’s Health and Center for Women’s Health and Gender-Based Medicine, Amarillo, TX, USA; 3 Kiromic LLC, Lubbock, TX, USA; 4 Division of Surgical Oncology, Texas Tech University Health Sciences Center Amarillo, TX, USA
Over the past 30 years, human papilloma virus (HPV) has been shown to play a role in the development of various cancers. Most notably, HPV has been linked to malignant progression in neoplasms of the anogenital region. However, high-risk HPV has also been suggested to play a significant role in the development of cancers in other anatomic locations, such as the head and neck, lung, breast and bladder. In 2006, the first vaccine for HPV, Gardasil, was approved for the prevention of subtypes 6, 11, 16 and 18. A few years later, Cevarix was approved for the prevention of subtypes 16 and 18, the HPV subtypes most frequently implicated in malignant progression. Although increased awareness and vaccination could drastically decrease the incidence of HPV-positive cancers, these approaches do not benefit patients who have already contracted HPV and developed cancer as a result. For this reason, researchers need to continue developing treatment modalities, such as targeted immunotherapies, for HPV-positive lesions. Here, we review the potential evidence linking HPV infection with the development of non-anogenital cancers and the potential role of immunotherapy in the prevention and eradication of HPV infection and its oncogenic sequela. Keywords: anogenital cancer, cervical cancer, Gardasil, head and neck squamous cell carcinoma, vevarix
Accepted 1 April 2014. ∗ Equal first author contributors † Equal senior author contributors Address correspondence to Everardo Cobos, MD, Chief of the Division of Hematology & Oncology, Professor, Department of Internal Medicine, School of Medicine, Texas Tech University Health Sciences Center, 3601 4th St., Mail Stop 9410, Lubbock, USA. E-mail: [email protected]
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
INTRODUCTION According to data from the National Program of Cancer Registries (NPCR) and the Surveillance, Epidemiology, and End Results (SEER) program, an average of 33 369 human papilloma virus (HPV)-associated cancers are diagnosed annually, including 12 080 among males (8.1 per 100,000) and 21 290 among females (13.2 per 100,000) [1, 2]. HPV is thought to be responsible for 7–8% of all human malignancies and is associated with 96% of cervical cancers, 93% of anal cancers, 64% vaginal cancers, 51% of vulvar cancer, 36% of penile cancers and 63% of oropharyngeal carcinomas [1, 3, 4] (Table 1). HPV belongs to a family of non-enveloped DNA viruses that typically infect skin or mucosal epithelium, particularly in squamous epithelial cells. Commonly infected surfaces include the skin, vagina, transitional zone of the cervix, anus, vulva, glans penis, mouth, throat, trachea, bronchi and lungs [5]. HPV’s association with malignancy has been well established for more than 30 years. In 1972, Stefania Jablonska first demonstrated an association of HPV with a type of skin cancer called epidermodysplasia verruciformis [6]. Soon after this discovery, Hearald Zur Hausen published his work linking HPV with cervical cancer, and, with Lutz Gissmann in 1977, isolated the HPV-6 DNA from genital warts [6]. In 1983, Zur Hausen implicated HPV-16 and -18 as key etiologic factors for the development of cervical cancer [6]. The main mode of HPV transmission is direct skin contact. The infection may occur via skin-to-skin contact during sexual intercourse, oral–genital, or hand–genital contact. Transmission from an infected mother to a newborn’s trachea is possible during delivery [2]. In the majority of cases (approximately 90%), the human immune system opsonizes the virus and clears it from the system within 2 years [7, 8]. Thus, the vast majority of infected individuals remain asymptomatic and develop no clinical sequelae. In symptomatic patients, infection with low-risk HPV subtypes can lead to the formation of genital warts and/or recurrent respiratory papillomatosis (RRP) within weeks or months after contact with an HPV carrier [2, 9]. Visible genital warts can be treated with prescribed medication, and RRP can be treated either by medication or surgical removal [10, 11]. Non-genital forms can also manifest as warts on other anatomic sites such as hands and feet. For women, a regular screening papinicolaou (PAP) smear test can detect early epithelial changes due to HPV infection allowing for early intervention and prevention of malignant progression [5]. Recently, the U.S. Food and Drug Administration (FDA) has approved two vaccines for the prevention of HPV infection: Gardasil (Merck and Co.) and Cervarix (GlaxoSmithKline). Gardasil was approved for use in the prevention of genital cancers and precancerous lesions in women [12]. In recent years, the rate of administered HPV vaccinations has steadily risen, which is expected to result in the decline of HPV-related cervical and anal cancers and deaths associated with these diseases in high-risk populations. Unfortunately, these vaccines will not eliminate existing HPV infections, and TABLE 1.
Percentage of HPV-associated urogenital cancer.
Type of cancer
HPV association from 2004 to 2008 72
HPV-positive DNA cancers from 1999 to 200571
HPV 16/18% from 1999 to 200571
Anal Cervical Vulvar Vaginal Penile Oropharyngeal
93% 96% 51% 64% 36% 63%
91% 90% 69% 75% 63% 72%
79% 66% 55% 49% 48% 62% International Reviews of Immunology
HPV in Human Cancer and Immunotherapy
public awareness of the importance of this intervention is critical in order to prevent future HPV infections and related malignancies in women [13]. For men, there is no established screening test for HPV, and the diagnosis of infection relies on visual inspection and physical findings [14]. Further research is needed to provide appropriate screening and preventive options for males at risk for HPV infection.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
The Role of HPV in Cancer To date, more than 100 subtypes of HPV have been discovered. There are approximately 60 HPV strains that are known to be related to common body warts. Those strains infect the cutaneous epithelium leading to the formation of common skin warts. The other 40 HPV types are called mucosal or genital-type HPV strains as they infect the orogenital mucosal epithelium [7]. The mucosal HPV subtypes infect moist, squamous cell surfaces of the body generally in the anogenital area. The mucosal subtypes are further classified as low-risk and high-risk. Low-risk types, including HPV-6 and HPV-11, cause cauliflower-shaped warts on or around the anogenital region of men and women. These lesions are known as condyloma acuminata and rarely transform into cancer. The high-risk types of HPV includes 16, 18, 31, 33, 35, 39, 45, 51, 52 and 58 and can eventually lead to cancer in both men and women. Therefore, high-risk HPV types can be life-threatening, and pre-screening is vital to detect these changes during early stages [2,7]. Studies have shown that the role of HPV infections in epithelial cancers has been associated with the expression of viral oncogenes E6 and E7 (Figure 1). These proteins transform cells by interacting with growth-regulating host cell proteins and are important in the pathogenesis of malignant disease. Particularly in cervical cancer, E6 and E7 play an instrumental role at the molecular level through their interaction with P53 and retinoblastoma (Rb) genes, respectively [10,11]. P53 is involved in cell-cycle regulation, DNA repair process and apoptosis, and represents the most commonly mutated gene in human malignancy [10]. The HPV E6 protein is able to bind P53 and targets this tumor suppressor for proteasomal degradation. This degradation of p53 results in unchecked cellular cycle progression and accumulation of chromosomal mutations due to impaired DNA checkpoint regulation. Furthermore, E6 association with p53 also causes inhibition of the Src family of nonreceptor tyrosine kinases, important in cytoplasmic signal transduction and the cell’s mitotic activity [15]. The Retinoblastoma (Rb) protein is one of the first-recognized tumor suppressor gene products that acts as a growth inhibitor by binding and inactivating the E2F general transcription factor. E2F controls DNA synthesis and cyclin function and promotes the entrance of the cells in the S phase of cell cycling. The HPV E7 protein binds Rb, causing E2F to be released, become activated and promote cell cycling. This process leads to DNA damage and genomic instability at the molecular level, giving rise to abnormal cell proliferation [9,10]. The E7 protein also interacts with wild-type p53, and this interaction further contributes to the immortality and subsequent transformation of the cells, through an anti-apoptotic affect [9, 10]. HPV in Head and Neck Cancer The first evidence of HPV’s relation to squamous cell carcinoma of the head and neck (HNSCC) was provided by Syrj¨anen KJ et al in 1983 [16]. It was later discovered that 40% of the laryngeal and oral cancers had morphologic similarities with HPV-infected lesions, and 20% of samples examined demonstrated HPV structural proteins, detected by immunohistochemistry [17]. Further studies have shown that HPV infection is associated with 23% to 35% of all HNSCC biopsies worldwide [18, 19]. The prevalence of HPV in HNSCC appears to vary according to ethnic and geographic location, C Informa Healthcare USA, Inc. Copyright
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
C. Cobos et al.
FIGURE 1. Molecular mechanisms of HPV-induced cell cycle dysregulation in cancer cells. Entry into the S phase from the G1 phase of cell cycle requires an active E2F elongation factor to transcribe S-phase genes, such as DNA polymerase. The protein, Rb, binds E2F inactivating its activity. The complex of CDK4 and cyclin D (CyD) phosphorylates E2F-bound Rb, producing a pRb–E2F complex. This event releases E2F, allowing the transition into the S phase. Then, the association with cyclin E activates CDK2, which in turn phosphorylates the pRb-producing pppRb (tri-phosphorylated Rb), which is then degraded by the proteasome. The HPV protein, E6, blocks both pRb and p21, which in turn blocks the CDK2–CyE complex, ultimately resulting in increased formation of pppRb. However, E6 blocks p53, which activates p21, resulting in an additional loss of regulation in the pRb degradation process. Moreover, blockade of p53 reduces cell sensitivity to apoptotic stimuli and the ability to repair damaged DNA.
tumor sub-site and HPV-detection methods [20]. The majority of HPV-positive HNSCC (up to 90%) originate in the oropharynx, and tend to arise mostly from the lingual and palatine tonsils. [20, 21]. The implicated HPV subtypes are similar to those in cervical cancer, with HPV-16 found in 68% to 87% of HPV-positive head and neck cancers worldwide. Less common HPV types also found in HNSCC include HPV-18, -31, -33, -35, -45, -51, -52, -56, -58, -59 and -68 [19]. In a case-control study of 100 patients with newly diagnosed oropharyngeal cancer, a high number of vaginal-sex partners (26 or more) during life was associated with oropharyngeal cancer (odds ratio [OR], 3.1; 95% confidence interval [CI], 1.5–6.5), as was a high lifetime number of oral-sex partners (6 or more; OR, 3.4; 95% CI, 1.3–8.8) [18]. The odds of acquiring an oral HPV infection were significantly increased in HIVpositive patients, compared with those who were HIV negative, after controlling for sexual behavior [22]. In this study, the use of marijuana was also found to be associated with HPV-16-positive HNSCC. This association was stronger with greater numbers of oral sex partners and increasing intensity and duration of marijuana use. However, tobacco, alcohol and poor oral hygiene were associated primarily with HPV-negative
International Reviews of Immunology
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
HPV in Human Cancer and Immunotherapy
HNSCC [23]. A defective immune system also appears to increase the risk of HPVpositive HNSCC and other HPV-associated malignancies [20]. Several differences exist between HPV-positive and -negative HNSCC. For example, patients with HPV-positive oropharyngeal tumors tend to be younger (40–60 years) as compared with HPV-negative patients (60 years or more), it is more common in men as compared with women (3:1 ratio) and is more prevalent in Caucasian patients with higher education and economic status [20, 23, 24]. HPV-positive HNSCC commonly present with poorly differentiated non-keratinizing, basaloid features compared with HPV-negative HNSCC, which presents with a more differentiated and keratinized morphology. HPV-positive HNSCC also commonly presents with more advanced TNM stage (III and IV), reflecting the presence of more advanced lymph node metastases [19, 20, 25]. Interestingly, several studies have shown that locally advanced HPV-positive HNSCC treated with radiation and chemotherapy has a better outcome in terms of survival, response to treatment, and rates of local recurrence, when compared with HPVnegative cancers. In the Radiation Therapy Oncology Group (RTOG) 0129 trial, 743 patients with stages III and IV HNSCC were retrospectively analyzed and 63.8%, of patients with oropharyngeal cancer were found to have HPV-positive tumors. The results of this study showed a better 3-year overall survival (82.4%, vs. 57.1% P < 0.001) for HPV-positive compared with HPV-negative cancers [26]. Another retrospective analysis done on an HPV-positive (detected by p16 overexpression) cohort of patients with oropharyngeal cancer of RTOG study 02.02, also demonstrated an improved 2year OS and failure-free survival [27]. Patients with HPV-positive oropharyngeal carcinoma in the TAX 324 trial also had better 5-year OS (82% vs 35%), and PFS (78% vs 28%) [28] than their HPV-negative counterparts. In another retrospective study, survival of HPV-positive and HPV-negative patients was compared in 271 oropharyngeal cancers (1984–2004) collected by the SEER program. The median survival was significantly longer for HPV-positive than for HPV-negative patients, and it persisted beyond 15 years after cancer diagnosis (131 vs 20 months; log-rank p < 0.001). In addition, a 69% reduction in the risk of death was observed among HPV-positive cancers. Survival was better among HPV-positive patients who were treated by radiation compared with those who were not [29]. A study by Fakhry and coworkers prospectively evaluated the association of tumor HPV status with therapeutic response and survival among 96 patients with stage III or IV HNSCC of the oropharynx or larynx [24]. In this Eastern Cooperative Oncology Group (ECOG) phase II study, patients received two cycles of induction followed by concomitant chemoradiation therapy. Patients with HPV-positive tumors had higher response rates after induction chemotherapy (82% vs 55%, p = 0.01) and chemoradiation treatment (84% vs 57, p = .007). After a median follow-up of 39.1 months, patients with HPV-positive tumors had improved overall survival (2-year overall survival = 95% vs 62%, p = 0.005) [24]. The rate of progression and locoregional disease, but not distant metastasis, was also significantly lower for patients with HPV-positive than HPV-negative tumors [24]. Another important difference between HPV-positive and HPV-negative HNSCC is the observation that the cumulative incidence of second primary tumors appears to be significantly lower among patients with HPV-positive tumors [26, 27]. Interestingly, a recent meta-analysis has found no difference in survival between HPV-positive and HPV-negative HNSCC of non-oropharyngeal origin [29]. Smoking has also been associated with increased mortality and cancer relapse in both HPV-positive and HPVnegative cancers, and this increment is estimated to be approximately 1% for each additional pack-year of tobacco smoking [26]. C Informa Healthcare USA, Inc. Copyright
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
HPV in Less-Studied Malignancies HPV has been well studied in urogenital as well as head and neck malignancies, but it has also been implicated in many other types of cancers. Here, we discuss the available evidence linking other types of malignancies to HPV infection. These data are summarized in Table 2. HPV in Lung Cancer Lung cancer is the leading cause of cancer-related mortality worldwide. HPV was first implicated in the etiology of lung cancer in 1979 and, over the past 30 years, many studies have tried to confirm this relationship [30–32]. There is increasing evidence that high-risk HPV might be associated with tumorigenesis and progression of nonsmall cell lung cancer (NSCLC) [33–39]. It has been proposed the route of transmission could be either through mucosal barrier–blood–lung tissue infection or through oral cavity–larynx–lung tissue during sexual contact [40, 41]. A recent study by Munoz et al has also proposed a possible relationship between tobacco carcinogens and HPV16 E6/E7 oncoproteins, their interaction leading to malignant transformation and tumorigenesis of lung epithelial cells [41]. However, in a case-control study Koshiol et al selected 450 Italian patients with lung cancer and discovered that none of the specimens were positive for HPV subtypes. This study concluded that there was no evidence to support a relationship between HPV and lung cancer in this particular representative sample of a Western population [42]. A recent meta-analysis by Syrjanen suggested these discrepancies in HPV prevalence in patients with lung cancer were likely due to differences in geographic location and histologic subtype, rather than the technique used for HPV isolation [43]. Therefore, the definitive role of HPV infection in lung cancer remains unclear, and prospective cohort studies are needed to confirm or disprove this association. HPV in Esophageal Cancer It was estimated that 17 460 subjects (13 950 men and 3510 women) would be diagnosed with esophageal cancer in 2012, and 15 070 men and women would die of cancer [44]. It is well recognized that the incidence of esophageal carcinoma is increasing due to a higher prevalence of chronic gastroesophageal reflux disease (GERD) and Barrett’s esophagus. Nonetheless, it is conceivable that HPV could be implicated in the etiology of esophageal squamous cell cancer (ESCC), because oral squamous epithelium and the epithelium of the upper esophagus are histologically similar. Two large studies were done in China in 2000 and 2006 to explore this hypothesis. In the first study, 700 esophageal carcinomas were analyzed for HPV DNA sequences, which was found to be present in 17% of the samples examined. High-risk HPV-16 and HPV-18 strains were found to be present in 27% of the cases studied [45]. In the second study, 702 patients underwent endoscopy and cytologic evaluation of biopsy specimens and demonstrated HPV positivity in 8%, 7% and 16% of subjects with mild, moderate and severe dysplasia, respectively [46]. Recently, a study by Yahyapour et al used archived tissue blocks from 177 patients with esophageal squamous cell carcinoma (ESCC) and evaluated them for the presence of HPV DNA by PCR. HPV was detected in 49 patients (28%), of which 25 were positive for at least one high and one low-risk HPV genotypes [47]. Another study by Gupta et al identified HPV DNA in 17 of 37 (46%) cases of nonkeratinizing SCC of the esophagus [48]. A review by Liyanage et al identified 139 studies with results indicating higher levels of HPV detection in areas with a high-risk of ESCC, than in areas with a lower risk of contracting the disease [49]. These investigators concluded these inconsistencies were due to variations in technique, study design and sample types [49]. Another study using 1500 serum samples from patients with ESCC found a significant association International Reviews of Immunology
Not definitively established
Not definitively established
No association
Not definitively established
Not definitively established
Not definitively established
Lung
Esophageal
Prostate
Bladder
Breast
Skin
Role of HPV
90% in immunocompromised and 50% in immunocompetent patient [111].
0–61% [77,78, 80–82, 110]
Meta- analysis 16%–19.4% (range from 3%–35%) [68,69, 72–75]
—
7–46% [45,46, 48]
22.4% [73]
Incidence of HPV in studies
Summary of HPV role in other forms of cancer.
Type of cancer
TABLE 2.
Disruption of DNA repair activity by UV radiation along with ubiquitously present cutaneous HPV types increases the risk of NMSC
Only one study demonstrated Id-1 is an important target for E6/E7 oncoproteins of HPV type 16, 18, 31, 33 and 35 in breast cancer cells.2
Epithelial tropism of HPV and the anatomic proximity of the urogenital zone1
—
Mucosal barrier–blood–lung tissue infection or through oral cavity-larynx-lung tissue during sexual contact. Oral squamous epithelium and upper esophagus similar histologically.
Possible mechanism
Summary High-risk HPV might be associated with the tumorigenesis of NSCLC mainly in Asian nonsmokers. No association established in Western countries. High-risk HPV has 27% incidence in Esophageal cancer but pathogenesis remains inconclusive. Pathogenesis questionable since it is adenocarcinoma doesn’t occur by direct contact with the exposure Meta-analysis points out to an association with the incidence of HPV infection from bladder infection, no clear evidence and also needs to subcategorize the data based on subtype of bladder cancer No association was found in multiple studies and further all definite associations with HPV are squamous cell cancers and breast cancer is exclusively adenocarcinoma. Beta genus HPV might be co-carcinogen with UV radiation, but a causal relationship needs to be established.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
C. Cobos et al.
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
with the presence of HPV-16 E6, HPV-33 L1, HPV-6 E6, HPV-6 L1 and HPV-11 L1 [50]. In this study, no definite conclusion could be reached regarding HPV association in the oncogenesis of ESCC, although an association was suggested. Based on the available information, the relationship between HPV and the etiology of ESCC remains to be determined.
HPV in Prostate Cancer The role of HPV in the pathogenesis of prostate cancer is a topic of debate largely fueled by the notion that increased and/or risky sexual behavior and more frequent exposure to sexually transmitted diseases (STDs) may play a role in the etiology of this malignancy [51, 52]. Although a clear association between prostate cancer and HPV has not been supported [53], the presence of oncogenic viruses, including high-risk HPV types, have been documented in normal and neoplastic prostate tissues. In a recent study, Whittaker and collaborators detected both HPV-18 and Epstein–Barr Virus (EBV) DNA sequences in normal, benign and cancerous prostate specimens [54]. In this study, 24% of prostate cancer specimens demonstrated the presence of koilocytes, an indication of HPV-induced cellular infection. These findings suggest HPV infection may be a ubiquitous phenomenon in prostate tissues with possible oncogenic potential. However, as HPV carcinogenesis depends on the expression of viral oncogenes which inhibit Rb and p53 function, the observation that these tumor suppressor gene products are inactivated by mutations and/or deletions in prostate cancer suggest a mechanism of oncogenic transformation independent of HPV effects [53]. Based on the available evidence, the etiologic relation of HPV and prostate cancer remains uncertain.
HPV in Bladder and Urothelial Cancer Several studies have demonstrated the presence of HPV DNA in carcinomas of the bladder, suggesting HPV might be associated with the development of this malignancy [55–66]. However, a definitive association between HPV and bladder cancer still remains controversial [67–69]. For example, Noel and co-investigators analyzed 75 cases of transitional cell carcinoma of the bladder for the expression of DNA for HPV subtypes 6, 11, 16, 18 and 33 and identified the virus in only two cases (2.7%) [70]. Ben Selma et al analyzed 125 archival specimens of bladder carcinoma (119 urothelial carcinomas, five squamous carcinomas, and one adenocarcinoma) by PCR and did not detect anogenital-type HPV DNA in any of the specimens [71]. However, Barghi et al used PCR to identify HPV DNA in bladder tissue specimens from 59 patients with bladder cancer and 20 controls, and found HPV DNA was present in 21 (35.6%) of the bladder cancer cases and 1 (5%) of the control group (p = 0.008) [72]. In this study, the HPV subtype 18 was isolated from HPV-positive bladder cancer specimens in 81% of the subjects. A meta-analysis of 39 studies conducted by Gutierrez et al in 2006 found an association of HPV infection and bladder cancer in 16% of the studies analyzed [73]. In a different meta-analysis by Pacheco et al including 38 publications, HPV DNA was detected in 19.4% of bladder cancer cases [74]. The pooled OR in this study was 3.2 (95% CI 1.19–8.60, p = 0.02) indicating a significant relationship between HPV and bladder cancer [74]. More recent meta-analyses have supported an association between HPV infection and bladder cancer. Li et al used data from 52 studies to calculate a prevalence of HPV infection in bladder cancer of 16.88 [75] and, in 2012, a meta-analysis of 21 observational, case–controlled studies showed a pooled OR of 2.13 for the presence of HPV DNA in bladder cancer specimens (95% confidence interval [CI] 1.54–2.95) [76]. Despite the lack of prospective data, these International Reviews of Immunology
HPV in Human Cancer and Immunotherapy
Int Rev Immunol Downloaded from informahealthcare.com by University of Laval on 06/09/14 For personal use only.
large pooled studies suggest the possibility that HPV infection may be linked to the development of bladder cancer in a percentage of patients. HPV in Breast Cancer The possible association of HPV with breast cancer has been studied extensively over the past years. In a study by Gopalkrishna et al, fine-needle aspirates from 26 patients with breast cancer and four breast tumor biopsies were analyzed for the presence of HPV-16 and -18 DNA sequences, by both PCR and Southern blot hybridization. None of the samples studied demonstrated the presence of the virus [77]. More recently, Hedau et al prospectively collected a total of 228 biopsies and 142 blood samples from 252 patients with breast cancer and tested them for HPV DNA sequences using conventional PCR. They were unable to detect HPV DNA in any of the biopsy or blood samples [78]. Another study by Silva et al used PCR to analyze 79 samples of invasive ductal cancer of the breast for the presence of HPV DNA subtypes 6, 11, 16 and 18, and none were found to be positive for viral DNA [79]. A recent study reported HPV expression in only four of the 70 specimens (5.7%) analyzed using in-situ hybridization (ISH), and in two of 70 specimens (2.9%) using in-situ PCR (IS-PCR) [80]. In another study, HPV DNA was found in 4.4% of 67 breast cancer specimens analyzed [81]. These earlier studies suggest that high-risk HPV types are either not associated with breast cancer, associated only with a very small proportion of breast cancer cases (
