Control of Epstein-Barr Virus (EBV) in the Oral Cavity Is Associated With Persistence of Oral Human Papillomavirus (HPV)16/18 Among Men From the HPV Infection in Men Study

Abstract

Background

Human papillomavirus (HPV)-related oropharyngeal cancer (OPC) incidence is increasing among men. Biomarkers that can identify oral HPV16/18 infections likely to persist, the obligate precursor for HPV-OPC, are needed.

Methods

We assessed the association between oral Epstein-Barr virus (EBV) and oral HPV16/18 persistence among 63 men in the HPV Infection in Men Study who tested positive for HPV16/18 at the baseline visit. Control of oral coinfections, including EBV, could serve as a biomarker of immune competence and the ability to control oral HPV.

Results

Detection of oral EBV was significantly associated with oral HPV16/18 ≥12-month persistence.

Conclusions

Detection of oral EBV deserves evaluation as a biomarker for oral HPV persistence and HPV-related OPC.

Detection of oral EBV shedding was significantly associated with the persistence of an oral HPV16/18 infection to at least 12 months. EBV coinfection therefore deserves further evaluation in its association with oral HPV persistence.

Globally, rates of oropharyngeal cancer (OPC) associated with human papillomavirus (HPV) infection are increasing in men [1]. More than 90% of HPV-associated OPC is attributed to HPV16 infection [2], an oncogenic HPV type that is also responsible for most cervical cancers in women [3] and anal cancers globally [4].

Most HPV infections resolve within 1 year through natural immunity [5], but a proportion persist longer [6]. Viral persistence is considered an obligate precursor to HPV-associated precancer and cancer at multiple anatomic sites. Although there is no precancer lesion that can aide in early detection of HPV-associated OPC, a persistent infection with an oncogenic type of HPV (eg, HPV16 or HPV18) is accepted as a reliable intermediate biomarker [6]. Factors that influence the persistence and clearance of oral high-risk HPV infections remain unknown.

It is likely that the ability to clear an oral HPV infection is correlated with the ability to control other infections at the same anatomic site. One infection that occurs in the oral cavity that has not been explored as a biomarker for persistent oral HPV infection is Epstein-Barr virus (EBV). Epstein-Barr virus is a saliva-transmitted gamma herpes virus that infects approximately 90% of individuals by adulthood and replicates in epithelial cells of the oral cavity [7, 8]. A previous study demonstrated that detectable seminal shedding of EBV was significantly associated with the detection and persistence of anal high-risk HPV in men with human immunodeficiency virus (HIV), provocative evidence that EBV control may identify which HPV infections are most likely to persist [9]. We therefore assessed the association between oral EBV shedding, which indicates active viral replication, and oral HPV16/18 persistence among 63 men with prevalently detected oral HPV16 and HPV18 infections.

METHODS

Study Population

The HPV Infection in Men (HIM) Study is an international HPV natural history study of >4000 men from Tampa, Florida, USA, Cuernavaca, Mexico, and São Paulo, Brazil, the details of which have been published previously [10]. Men were enrolled beginning in June of 2005 and followed biannually for a median of 4.1 years, with an average interval of 6.9 months between visits. After providing informed consent, demographic information and behavioral data were obtained using computer-assisted self-interviewing. Approval was obtained from the Human Subjects committees of the University of South Florida (United States), Ludwig Institute for Cancer Research (Brazil), Centro de Referencia e Treinamento em Doencas Sexualmente Transmissíveis e AIDS (Brazil), and Instituto Nacional de Salud Publica de Mexico (Mexico). In 2007, oral sampling was added to the HIM study biannual visits. A total of 3591 men have at least 2 archived oral gargle specimens. For this study, men who were found to be positive for HPV16 or HPV18 at the baseline oral specimen visit (n = 63) were included.

Specimen Processing

Human Papillomavirus Genotyping

At each 6-month oral visit, oral gargle specimens were collected by a 30-second rinse/gargle with 15 mL mouthwash. Within 24 hours of collection, the gargle specimen was centrifuged, the supernatant was decanted, and the resulting cell pellet was resuspended in 20 mL phosphate-buffered saline (PBS) (4°C). Centrifugation and pellet washing were repeated twice, with the final cell pellet resuspended in 1.2 mL PBS and maintained at −80°C until deoxyribonucleic acid (DNA) extraction for HPV genotyping. The DNA was extracted using the automated BioRobot MDx kit (QIAGEN, Inc.). The HPV genotyping was performed using the SPF10 PCR-DEIA-LiPA25 system (DDL Diagnostic Laboratory, Rijswik, the Netherlands) [11].

Epstein-Barr Virus Deoxyribonucleic Acid Assessment

Extracted DNA from the oral baseline and 6-month visit specimens was assayed using PCR reagents from QIAGEN and the Rotor-Gene Q MDx instrument for the amplification of a 97-base pair region of the EBV EBNA-1 gene. Analyses of PCR output were completed using Rotor-Gene Q software. Cellular beta-globin DNA was quantified for each specimen as a positive control (ie, to ensure adequate DNA was extracted to allow for EBV detection). In addition, all samples were tested in duplicate to assess reproducibility. Output for these duplicate samples was highly correlated (Spearman coefficient = 0.87; 95% confidence interval [CI], 0.82–0.91).

Statistical Analyses

Human Papillomavirus Persistence

For this study, we considered each participant’s baseline, 6-month, and 12-month HPV16/18 status to categorize oral HPV16/18 persistence. Transient oral HPV16/18 infections included men who were only oral HPV positive at the baseline visit but at no subsequent visits (n = 44). Participants were considered to have a 6-month persistent HPV infection if the same oral HPV type present at baseline was detected at their 6-month visit (n = 8), whereas participants with oral HPV16/18 infection that was present at baseline, the 6-month visit, and the 12-month visit were considered to have ≥12-month persistence (n = 11).

Epstein-Barr Virus Detection

Oral EBV detection was categorized as detectable (PCR reading >zero) versus undetectable. Using the baseline and 6-month oral visit specimens, participants were categorized into the following 3 groups: (1) “No EBV Detection” if no EBV DNA was detected at either visit; (2) “Transient EBV Detection” if EBV was detected at only 1 of the 2 study visits; and (3) “Persistent EBV Detection” if EBV was detected at both visits.

The mean logEBV viral load in the oral gargle collected at the baseline visit was compared using an unpaired t test between the following: (1) men with 6-month HPV16/18 persistence versus men with no persistence; (2) men with 12-month HPV16/18 persistence versus men with no persistence; and (3) men with 6-month versus 12-month HPV16/18 persistence.

An alpha of 5% was considered statistically significant. Multinomial logistic regression was used to determine the association between oral EBV shedding and oral HPV16/18 persistence, adjusted for age, country of residence, and smoking status.

RESULTS

Of the 63 men who were oral HPV16/18 positive at the baseline visit, 44 had transient HPV16/18 infection, 8 persisted 6 months only, and 11 persisted ≥12 months. Characteristics of those with transient, 6-month persistent, and ≥12-month persistent oral HPV16/18 infections were not statistically significantly different between groups, although men with oral HPV16/18 infections that persisted at least 12 months were more often current smokers (54.6% vs 27.3% of transient and 25.0% of 6-month persistent HPV infections) and men who have sex with men (18.2% vs 4.6% of transient and 0% of 6-month persistent HPV infections). Prevalence of oral EBV detection at the baseline visit was the only variable that was statistically significantly different between groups (P = .036), with the highest detection (81.8%) observed among men with ≥12-month persistent oral HPV16/18 infection compared with men with transient (47.7%) or 6-month persistent (25.0%) HPV infection.

Of the 11 men who had persistent oral HPV16/18 at the 12-month visit, 5 did not persist past 12 months (median baseline EBV = 2.85 copies/mL), 3 persisted up to the 18-month visit (median EBV at baseline = 269 copies/mL), and 3 persisted between 36 and 66 months (median baseline EBV = 214 copies/mL) (data not shown). In Figure 1, mean logEBV viral load was compared according to oral HPV16/18 persistence. Oral EBV shedding was highest among men with a baseline oral HPV16/18 infection that persisted ≥12-months (mean = 5.2 log copies/mL) and was significantly different than the mean logEBV viral load among men with a transient oral HPV16/18 infection (2.5 log copies/mL; P = .047) or oral HPV16/18 infection that only persisted 6 months (0.82 log copies/mL; P = .020). Mean logEBV viral load was not significantly different between men with a transient oral HPV16/18 infection and men with an HPV16/18 infection that persisted only 6 months (P = .23).

Figure 1.

Oral Epstein-Barr virus (EBV) detected at baseline was log transformed, and the distribution (mean and 95% CI represented by horizontal bars) was compared across men with a transient oral human papillomavirus (HPV)16/18 infection (n = 44), those with oral HPV16/18 infection that persisted to 6 months only (n = 8), and those with oral HPV16/18 infection that persisted to at least 12 months (n = 11). *, P < .05; (NS [not significant]), P > .05.

Baseline oral EBV detection was not statistically significantly associated with oral HPV16/18 infections persisting only 6 months (adjusted odds ratio [aOR], 0.27; 95% CI, 0.043–1.74) but was associated with ≥12-month oral HPV 16/18 persistence in crude (OR, 4.93; 95% CI, 0.95–25.47; P = .06) and adjusted models (aOR, 8.60; 95% CI, 1.18–62.51; P = .03) compared with men with no EBV detected at baseline (Table 1).

Table 1.

Association of Baseline Oral EBV DNA Shedding in Saliva With a Persistent Oral HPV16/18 Infection to 6 or 12 Months

EBV at Baseline	Persist to 6 Monthsa (n = 8)			≥12 Month HPV Persistenceb (n = 11)
	N (%)	OR (95% CI)	aORc (95% CI)	N (%)	OR (95% CI)	aORc (95% CI)
No EBV Detected (n = 31)	6 (75.0)	REF	REF	2 (18.2)	REF	REF
EBV Detected (n = 32)	2 (25.0)	0.37 (0.07–2.01)	0.27 (0.043–1.74)	9 (81.8)	4.93 (0.95–25.47)	8.60 (1.18–62.51)
P Value		.25	.17		.06	.03

Abbreviations: aOR, adjusted odds ratio; CI, 95% confidence interval; DNA, deoxyribonucleic acid; EBV, Epstein-Barr virus; HPV, human papillomavirus; REF, referent group.

Assessed whether men with HPV16/18 infection persisted to only 6 months or had transient infection.

Assessed whether men with HPV16/18 infection persisted ≥12 months or had transient infection.

Adjusted for age category, smoking, and country.

When examining sustained oral EBV DNA detection (ie, oral EBV shedding at baseline and 6-months), small sample sizes precluded stable estimation, but we did observe preliminary evidence of an association between increased frequency of EBV detection and the odds of ≥12-month oral HPV16/18 persistence (details in Supplementary Table 1), indicating sustained oral EBV shedding was associated with ≥12-month oral HPV16/18 persistence.

DISCUSSION

We leveraged the largest existing HPV natural history study (HIM Study) to understand whether oral EBV coinfection is associated with persistence of prevalently detected oral HPV16/18 infections. Results from this study indicate that EBV coinfection is significantly associated with prevalent oral HPV16/18 infections persisting to at least 12 months.

In prior studies, sexual behavior, advanced age, oral health, and living with HIV have been associated with 6-month and 12-month oral HPV persistence [12, 13]. This is the first study to examine associations between oral EBV coinfection and oral HPV16/18 persistence. We observed that both detection of oral EBV DNA shedding in saliva and higher EBV viral loads were associated with prolonged oral HPV16/18 persistence in men. In a study of 131 men with HIV, presence of seminal EBV shedding at baseline was associated with increased anal HPV persistence over 48 weeks of follow-up compared with the EBV nonshedders (P < .01). Considered alongside our findings, these data provide evidence to support a possible correlation between EBV coinfection and persistence of oncogenic HPV infections.

Epstein-Barr virus acquired through saliva can infect epithelial cells lining the oral cavity, subsequently establish a latent infection in B cells, and then periodically reactivate to shed in the oral cavity, which allows forward transmission via saliva. Although EBV itself may not directly influence biological mechanisms of oral HPV persistence, active EBV shedding in the oral cavity may be indicative of an impaired immune response and therefore serve as a biomarker of oral HPV immune control. Evidence that immunity may play a role in oral HPV persistence is supported by the fact that people with HIV (PWH), who experience consistent immunosuppression, are at increased risk of HPV-related OPC. Specifically, PWH have a 60% higher risk of developing HPV-related OPC compared with the HIV-uninfected population [14]. Furthermore, regardless of HIV status, poor oral health (ie, periodontitis linked to ongoing inflammation) that can promote multiple oral pathogens is associated with HPV-related OPC risk [15]. In sum, the results of this study suggest that poor immunity that can adversely impact viral control in the oral cavity, as measured by presence of oral EBV DNA shedding in our study, may be associated with effective control over oncogenic oral HPV infections.

This study had a limited sample size. Our post-hoc power analysis revealed that we had 80% power to detect a log mean difference of (1) 1.0 between men with transient oral HPV and 11 men with oral HPV16/18 infections persisting ≥12 months and (2) 1.2 between transient oral HPV and 8 oral HPV16/18 infections persisting to 6 months. Notably, shedding of EBV in the oral cavity was associated with a higher likelihood of oral HPV16/18 persistence after adjustment for age, country of residence, and smoking status. In the current study, we were able to examine persistence of prevalently detected oral HPV infections. Future studies are needed to evaluate the impact of EBV coinfection on both prevalent and incident oral HPV16/18 infections. Expansion to include other high-risk oral HPV genotypes would also add valuable insight into the degree to which EBV influences oral HPV persistence. Finally, although EBV is a common infection shed in saliva, other viruses with oral epithelial tropism exist, including other herpesviruses and oral bacteria associated with periodontitis. Such pathogens may also be good candidates for future investigation as biomarkers of high-risk oral HPV persistence and HPV-related OPC.

CONCLUSIONS

Both EBV and HPV are ubiquitous viruses that infect most adults but are carcinogenic in only a small proportion of affected individuals. Evaluations of coinfection as a tool to identify those at highest risk of disease onset are lacking. This study provides preliminary evidence that oral EBV detection is associated with 12-month or greater persistence of oral HPV16 or 18 infections in men. Epstein-Barr virus DNA detected in the oral cavity may be a useful biomarker that reflects a lack of immunity to respond to oral infections.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Financial support. This work was funded by the National Institutes of Health (Grant Number R01 CA214588), the Center for Immunization and Infection Research in Cancer at Moffitt Cancer Center, and the State of Florida. B. L. D.’s work was supported by the T32 Postdoctoral Fellowship in Molecular Epidemiology (T32CA147832; Principal Investigator P. Kanetsky).