Abstract
Background:
Effective screening for Oropharyngeal cancer (OPC) is lacking. Four oncogenic HPV clearance definitions were explored to understand long-term natural history for persistent oncogenic oral HPV (oncHPV), the precursor of OPC.
Methods:
Prospective multicenter cohort of participants living with/at-risk for HIV, with oral rinse and gargle samples collected every 6–12 months for up to 10 years and tested for oncHPV. HPV clearance definitions included 1 (clear1), 2 (clear2), 3 (clear3) consecutive negatives, or being negative at last 2 visits (clearlast).
Results:
Median time to clearance of oncHPV exceeded 2 years for conservative definitions (clear3: 2.38,clearlast: 2.43), but not lenient (clear1:0.68, clear2:1.15). By clear3, most incident infections cleared at 2,5,8 years (55.1%,75.6%,79.1%), contrary to prevalent infections (37.1%,52.5%,59.5%, respectively). In adjusted analysis, prevalent oncHPV, older age, male sex and living with HIV were associated with reduced clearance. Of 1833 subjects screened, 13.8% had prevalent oncHPV and 47.5% of those infections persisted ≥5 years, representing 6.5% of persons screened. Two men with prevalent oHPV16 developed incident OPC (IR=1.62 per 100 person-years,95%CI=0.41,6.4). Many with oHPV16 persisted ≥5 years (and/or developed HPV-OPC) among those with 2 (72.2%), ≥2 of first 3 (65.7%), or 3 (80.0%) consecutive positive oHPV16 tests, but not after 1 (39.4%).
Conclusions:
In our 10 year study, most incident infections cleared quickly. However, half of prevalent oncHPV persisted ≥5 years, suggesting increased risk with persistent oncHPV at > 2 visits.
Impact:
We identified groups with persistent oncHPV at increased risk of OPC and contextualized risk levels for those with oral HPV16 infection.
Keywords: oral, HPV, natural history, clearance, persistence, oropharyngeal cancer, risk factors
INTRODUCTION
Oncogenic oral HPV (oncHPV) infection precedes diagnoses of HPV-associated oropharyngeal cancer (HPV-OPC). A large nested case-control study detected oncHPV a median of 42 months prior to HPV-OPC diagnosis.(1) Additionally, oncHPV has been detected among healthy adults who have steadily increasing HPV viral loads prior to HPV-OPC development.(2) Natural history studies suggest that most infections clear spontaneously within two years and few persist beyond 4 years.(3) Long-term oncHPV natural history and progression to cancer remain unexplored. While there is no recognized HPV-OPC pre-cancer,(4) persistent oncHPV is a potential surrogate. Therefore, screening for oncHPV may be a viable strategy for earlier detection of HPV-OPC. Indeed, type-specific oncHPV persistence is a surrogate endpoint for OPC risk in preventive vaccine trials,(5) and in recurrent HPV-OPC, persistent oncHPV is considered equivalent to subclinical microscopic (i.e. low-volume) disease.(6,7)
To understand risk factors for progression to HPV-OPC, defining HPV clearance is key. Previously, six-month sampling compared to two-week intervals estimated short-term persistence well, although cumulative prevalence of oral HPV DNA is higher with more frequent interval sampling.(8) Because one negative test can over-estimate clearance, the anogenital HPV literature commonly defines HPV persistence as two or more positive tests at six-month intervals.(9) Using 10 years of oncHPV natural history data, we examine the impact of differing definitions on inferences of oncHPV clearance.
MATERIALS AND METHODS:
Study Design and Population
This longitudinal observational cohort study collected oral rinse and gargle samples tested for oncHPV DNA for up to 10 years. Persistent Oral Papillomavirus Study (POPS)(10) consisted of 1833 men and women living with or without HIV from the multi-institutional MACS WIHS Combined Cohort Study (MWCCS). Participants underwent semiannual visits for four years. Inclusion criteria allowed all interested participation at participating MWCCS study sites (MACS: Baltimore, Chicago, Pittsburgh, Los Angeles; WIHS: Brooklyn, Bronx, San Francisco, Chicago, Chapel Hill, Atlanta) to enroll if interested.
131 participants with persistent HPV DNA at the end of POPS were enrolled in the Men and Women Understanding Throat HPV (MOUTH)(11) study, after a median of 2.5 years (IQR=1.9–3.0) between studies, and had 3 additional years of annual visits Supplemental Figure 1). The study population included participants with ≥1 oncogenic oral HPV type detected and ≥1 follow-up visit(s) thereafter (N=692 infections among 435 of 1833 participants in POPS]). This study was approved by the institutional review boards of all sites, and informed written consent was obtained from all participants. The study was conducted in accordance with recognized ethical guidelines of the Declaration of Helsinki.
HPV Testing
HPV detection was performed using PCR amplification in POPS (PGMY09/11 primer-system)(10) and MOUTH (SPF10, Labo Bio-medical Products, Netherlands) and genotyped using the Roche linear array (POPS) and SPF10 DEIA/LIPA system (MOUTH) for oncogenic HPV types: 16,18,31,33,35,39,45,51,52,56,58,59,66, as previously described.(3,12) HPV16 copy number (CN) was measured by qPCR and standardized to CN/cell using RNase P for MOUTH and using ERV3 for POPS.
Blood samples were obtained from a subset of participants at POPS baseline (N=133, 30.5 % of 435 subjects in this analysis), and additionally from MOUTH baseline among MOUTH re-enrolled persons (N=120, 92% of the 131 in MOUTH extension), and tested for HPV antibodies. HPV antibody testing was performed centrally by the German Cancer Research Center (DKFZ) using multiplex serology method.(13) Serum was tested for E1, E2 antibodies of HPV types 16 and 18, and E6 and E7 oncoprotein antibodies for oncogenic HPV types 16,18,31,33,35,45,52,58, as previously described.(12)
Statistical Analysis
The primary analysis evaluated time to clearance of oncHPV DNA. All analyses were type-specific HPV infection, clustered by person to account for multiple infections within the same person. Type-specific oncHPV was classified as prevalent if detected at first study visit and incident if it was absent at first visit but detected later. oncHPV clearance was explored using four different definitions: “clear1”, “clear2”, and “clear3”, meaning the occurrence of 1, 2 and 3 consecutive negative results after a positive result. The fourth definition, “clearlast” only considered any infection cleared if it was negative at the last two visits (i.e., not cleared if detected at either of last two visits despite intervening negatives). For all four definitions, time of clearance was considered at the first negative of consecutive negatives meeting that clearance definition.
Data were organized based on the assumption of multiple unordered events of similar types per participant. Hence, a participant could have multiple oncHPV types and could clear them in any order. Upon clearing a type-specific HPV, participants remain in the risk set for clearance of other types, if any. Differences in time to clearance were explored using Kaplan-Meier curves and log-rank tests. Clear3 was used as the primary clearance definition for exploring risk factors for oncHPV clearance, using Cox proportional hazard models with robust standard errors, clustered by participant. Risk factors explored included HIV characteristics (HIV serostatus, current CD4 count, HIV RNA viral load), demographics (age, sex), HPV characteristics (incident/prevalent, HPV16/non-16 HPV type], HPV16 E6 seropositivity) and behaviors (current smoking, alcohol and marijuana use). Among HPV16 infections, HPV16 copynumber (HPV16CN) was explored using qPCR results. Given that HPV16CN was always 0 at visit of clearance, its effect was considered at first detection carried forward to explore utility as a predictor of clearance in an HPV-OPC screening context.
A sub-analysis of subjects who developed an incident OPC during follow-up was performed to examine timing of cancer diagnosis related to oncHPV. OPC incidence rates with 95% CI were estimated. For groups with 0 events, Hanley’s rule of 3 was used to estimate the upper confidence interval limit.(14). Among participants persistent by clear3, we examined oncHPV detection patterns to visualize differences in those considered persistent by clear3 but not by other definitions. All analyses were conducted using STATA16.1 and R4.2.2.
Data Availability
The data generated in this study are available upon request from the corresponding author.
RESULTS
This analysis included 435 people with 692 type-specific oncHPVs. At the visit of first detection, participants had a median age of 49.6 years (IQR=43.3–55.5), and the majority were male (58.4%), living with HIV (73.8%), current/ever smokers (45.6%/ 81.3%) and current drinkers (69.5%), Table 1. Among those living with HIV, median current CD4 cell count was 528 cells/μl (IQR=326–723), 61.6% had undetectable current HIV viral load (<50 copies per mL), and 81.6% were on ART. 58.2% of subjects with oncHPV had ≥1 prevalent HPV type, representing 13.8% of subjects tested (Table 1). About half of the infections detected were incident (52.5%). At time of this analysis, two participants had been diagnosed with an HPV-OPC (Figure 1).
Table 1 –
Characteristics of 692 oncogenic oral HPV at visit of first detection (among 435 participants included in this analysis).
| Number of people | Number of Type-specific Infections | % among 1833 screened | ||||
|---|---|---|---|---|---|---|
| Characteristic | N | % | N | % (of 692 observed) | % (of 5655 tested) | |
| Sex | ||||||
| Female | 181 | 41.6 | 289 | 41.8 | 5.1 | -- |
| Male | 254 | 58.4 | 403 | 58.2 | 7.1 | -- |
| Age in years | ||||||
| 22–45 | 138 | 31.7 | 215 | 31.1 | 3.8 | -- |
| 46–55 | 179 | 41.2 | 281 | 40.6 | 5.0 | -- |
| >55 | 118 | 27.1 | 196 | 28.3 | 3.5 | -- |
| HIV | ||||||
| HIV negative | 114 | 26.2 | 163 | 23.5 | 2.8 | -- |
| Person Living with HIV (PLWH) | 321 | 73.8 | 529 | 76.5 | 9.4 | -- |
| HIV and current CD4 cell count | ||||||
| HIV negative | 114 | 26.3 | 163 | 23.6 | 2.8 | -- |
| PLWH, CD4>500 | 171 | 39.4 | 266 | 38.5 | 4.7 | -- |
| PLWH, CD4 250–500 | 90 | 20.7 | 144 | 20.8 | 2.5 | -- |
| PLWH, CD4<250 | 59 | 13.6 | 118 | 17.1 | 2.1 | -- |
| Current Antiretroviral Therapy (among PLWH) | ||||||
| No | 59 | 18.4 | 92 | 17.4 | 1.6 | |
| Yes | 261 | 81.6 | 436 | 82.6 | 7.7 | |
| HIV Viral load (among PLWH) | ||||||
| Undetectable | 197 | 61.6 | 305 | 57.9 | 5.4 | -- |
| 50–9999 copies/mL | 70 | 21.8 | 122 | 23.1 | 2.2 | -- |
| 10000+ copies/mL | 53 | 16.6 | 100 | 19.0 | 1.8 | -- |
| Current smoker | ||||||
| No | 236 | 54.4 | 356 | 51.5 | 6.3 | -- |
| Yes | 198 | 45.6 | 335 | 48.5 | 5.9 | -- |
| Current pack/day (among ever smokers) | ||||||
| None | 170 | 48.3 | 254 | 45.1 | 4.5 | -- |
| <0.5 packs/day | 98 | 27.8 | 180 | 32.0 | 3.2 | -- |
| 0.5-<1 pack/day | 54 | 15.3 | 80 | 14.2 | 1.4 | -- |
| 1+ packs/day | 30 | 8.5 | 49 | 8.7 | 0.9 | -- |
| Current drinker | ||||||
| No | 131 | 30.5 | 214 | 31.2 | 3.8 | -- |
| Yes | 298 | 69.5 | 471 | 68.8 | 8.3 | -- |
| Oncogenic oral HPV | ||||||
| Incident | 182 | 41.8 | 363 | 52.5 | 6.4 | 9.9 |
| Prevalent | 253 | 58.2 | 329 | 47.5 | 5.8 | 13.8 |
| Oncogenic oral HPV type | ||||||
| Ever HPV16 | 107 | 24.6 | 107 | 15.5 | 1.9 | 5.8 |
| Ever non HPV16 oncogenic types | 328 | 75.4 | 585 | 84.5 | 10.3 | 17.9 |
| HPV vaccination status (self-reported) | ||||||
| Never Vaccinated | 360 | 85.7 | --- | --- | -- | |
| Ever Vaccinated | 17 | 4.1 | --- | --- | -- | |
| Unsure | 43 | 10.2 | --- | --- | -- | |
| Diagnosed with oropharyngeal cancer anytime after initial oncogenic oral HPV infection detection | 2 | 0.4 | --- | --- | 0.1 | |
Figure 1:
Natural history of oral HPV16 (shaded circle indicates positive and unshaded circle indicates negative) in semi-annual testing and HPV16 E6 serology results (red triangle indicates positive and unshaded triangle indicates negative) among participants later diagnosed with HPV related oropharyngeal cancer (blue diamond).
Median follow-up for participants was 4.4 years (IQR: 2.7–7.1, range: 0.4–11.5) from study entry to censor, and 8.8 years (IQR: 7.7–9.7, range: 3.9–11.5) among the subset re-enrolled in MOUTH. Reasons for censoring among the 435 enrolled included: study completion (N=362, 83%), loss to follow-up (N=68, 16%; COVID-related missed visits and closures of study sites. Characteristics of those lost to follow-up were similar to those completing the study. There were 5 deaths, all were HIV-related.
Median time to clearance varied dramatically by clearance definition. When using more stringent definitions requiring three repeated negative oral rinses (clear3) or absence of oncHPV at last samplings (clearlast), median clearance was over two years: (clear3: 2.38 years, IQR=0.52, undetermined; clearlast: 2.43 years, IQR=0.52, 7.78). In contrast, requiring two consecutive negatives (clear2), had median clearance of 1.15 years (IQR: 0.50, 4.04). Requiring only one negative oral rinse (clear1) reduced median time to clearance to less than one year (0.68 years, IQR: 0.49, 1.98). Median times to clearance were similar by clear3 and clearlast definitions, but both were significantly longer than those by clear2 or clear1 (Figure 2). Among all visits from first detection of type-specific oncHPV to clearance (by clear3) or censoring, 38.4% of tests were intermittently negative for that type, explaining the higher clearance by clear1 and clear2 than clear3, (Figure 3). Among persistent infections (i.e., never observed to clear during study), 28.6% of tests from first visit detected to censoring were HPV DNA negative.
Figure 2.
Clearance of prevalent (panel A) and incident (panel B) oncogenic oral HPV infections over 10 years.
Figure 3:
Examples of the pattern of oncogenic oral HPV DNA among subjects who did not clear their infections (by clear3). Included here are participants with oral HPV16 detected, who had at least 5 years of follow-up, stratified by whether clear1 and clear2 also considered these infections not to have cleared, or whether these alternative definitions considered the infections to have cleared. The top panel includes infections considered persistent (i.e. not to have cleared) by all three definitions (clear1, clear2 and clear3); the middle panel includes infections considered persistent by clear2 and clear3 but cleared by clear 1; the bottom panel includes infections considered persistent by clear3 but cleared by clear2 or clear 1. Legend: 1= HPV16 positive (Dark grey), 0=HPV16 negative (Light grey), * = missed visit, no test result (white).
Intermittent HPV16 detection (“flickering”) was significantly more common among those with low oral HPV16 copynumber (55% of those with ≤0.0001 copies/cell, 26% of those with 0.0001- <0.001, 16% of those with 0.001-<0.1) than those with high oral HPV16 copynumber (3.2% of those with ≥0.1 copies/cell). Higher HPV16 copynumber was associated with significantly lower odds of “flickering” (ie, p-trend <0.0001). Odds of flickering was significantly reduced in those with with ≥0.1 copies/cell (OR=0.05 95%CI=0.01,0.27) and 0.0001-<0.1 copies/cell (OR=0.25 95%CI=0.11,0.57) compared to those with <0.0001 copies/cell.
Most oncHPVs cleared during follow-up, even when using clear3 as the definition. Clearance of incident oncHPV was: 55.1%, 75.6%, and 79.1% at 2, 5 and 8 years, respectively. Clearance of prevalent oncHPV was lower with 37.1%, 52.5%, and 59.5% respectively at 2, 5 and 8 years (Figure 2). Thus, by 8 years, most (79.1%) incident infections had cleared, while only 59.5% of prevalent infections had cleared (p<0.001). The group without clearance of oncHPV had variable patterns of HPV DNA detected, some consistently detected and others fluctuating (intermittently detected; Figure 3).
Among the 1833 participants, 13.8% had prevalent oncHPV (Supplemental Table 1) and 47.5% of those did not clear by clear3 at 5 years, representing 6.5% of those screened. Oral HPV16 (oHPV16) prevalence was lower with 56 prevalent oHPV16 detected (3.1%), 53.6% of which persisted 5 years (Supplemental Figure 2), representing 1.6% of subjects screened. Among 35 male subjects with prevalent oHPV16 (123.4 years of follow-up), two developed HPV-OPC. Risk of oHPV16 persisting ≥5 years and/or developed HPV-OPC (5-year-risk) was high among those with 2 (72.2%), ≥2 of first 3 (65.7%), 3 (80.0%), or ≥3 of first 4 (75.8%) consecutive positive oHPV16 tests, compared to only 39.4% after 1 oHPV16 positive test. Among subjects with oHPV16 persisting 1-year, 5-year-risk was >50% by clear2 (57.2%) and clear3 (68.2%), but not clear1 (25.4%). Among subjects with oHPV16 persisting 2-years, 5-year-risk was >70% (clear2=72.1%, clear3=75.4%). Allowing intervening negatives expanded the proportion defined as oHPV16 persisting ≥5 years from 11.1% (clear1) to 31.3% (clear2) and 39.4% (clear3). Among men with prevalent oHPV16, the incidence rate (IR) of HPV-OPC was 1.62 per 100 person-years (95%CI=0.41,6.47) overall, and was higher in 26 men living with HIV than 9 men without HIV (2.28 per 100 person-years 95%CI=0.57–9.11 vs. 0). IR was also 0 among 21 women with prevalent oHPV16 and among 19 men and 32 women with incident oHPV16.
Both subjects who developed HPV-OPC had oHPV16 detected at first visit and persisted at follow-up visits (Figure 1). At study entry, both were men living with HIV, former smokers, and had a history of performing oral sex on ≥20 lifetime partners. Subject A was HPV16 E6- and E7-seronegative (L1 and E4 positive). At first HPV16 detection, oHPV16CN was detectable by RT-PCR at a low level (0.00038 copies per cell). Six years after first oHPV16 detection, while in his 60s, on ART, with CD4 count >500 cells/μl, he was diagnosed with HPV-OPC. At first visit, Subject B was HPV16 E1, E2, E6, E7 and L1 seropositive, oHPV16 was detected with low HPV16CN/cell (<RT-PCR limit). Six years later when diagnosed with HPV-OPC, he was in his 40s, on ART, with a CD4 count of 189 cells/μl.
Older age, male sex, prevalent oncHPV, and higher oHPV16CN were each associated with reduced oncHPV clearance in unadjusted analysis (Table 2, Supplemental Figures 3–7). Risk factors were similar when explored by all four clearance definitions, except that HPV16 was less likely to clear than other oncogenic types by clear1 and clear2, but not clear3 (Supplemental Table 2). In adjusted analysis, prevalent oncHPV (aHR=0.58, 95%CI=0.46,0.73), older age (aHR=0.82 , 95%CI=0.74,0.91), living with HIV (aHR=0.77, 95%CI= 0.60,0.99), male sex (aHR=0.83, 95%CI=0.66,1.03) and increasing oHPV16CN (per category of CN; aHR =0.63, 95%CI= 0.46,0.86, p-trend=0.004) remained significant predictors of reduced oncHPV clearance. Smoking, drinking, current and nadir CD4 cell count were not associated with oncHPV clearance (Table 2, Supplemental Figures 6–7).
Table 2.
Unadjusted and adjusted predictors of oncogenic oral HPV clearance (as defined by clear3)
| Characteristic | # visits | HR (95% CI) | P value | aHR (95% CI) | P value |
|---|---|---|---|---|---|
| Gender | |||||
| Female | 898 | 1 | 1 | ||
| Male | 1441 | 0.77 (0.62,0.96) | 0.02 | 0.83 (0.66,1.03) | 0.097 |
| Age (per 10 year increase) | 2339 | 0.81 (0.73, 0.90) | <0.0001 | 0.82 (0.74,0.91) | <0.0001 |
| Age categories, in years | |||||
| 22–39 | 213 | 1 | |||
| 40–49 | 527 | 1.13 (0.81,1.57) | 0.46 | ||
| 50–59 | 911 | 0.79 (0.57,1.09) | 0.15 | ||
| ≥60 | 688 | 0.53 (0.36,0.79) | 0.002 | ||
| HIV | |||||
| HIV negative | 516 | 1 | 1 | ||
| PLWH | 1823 | 0.83 (0.64,1.07) | 0.15 | 0.77 (0.60,0.99) | 0.046 |
| HIV and current CD4 cell count (cells/μL) | |||||
| HIV negative | 516 | 1 | |||
| PLWH, CD4>500 | 995 | 0.84 (0.63,1.10) | 0.21 | ||
| PLWH, CD4 250–500 | 482 | 0.85 (0.60,1.18) | 0.34 | ||
| PLWH, CD4<250 | 315 | 0.83 (0.59,1.16) | 0.28 | ||
| HIV and Nadir CD4 cell count (cells/μL) | |||||
| HIV negative | 516 | 1 | |||
| PLWH, nadir CD4>500 | 221 | 0.74 (0.46,1.17) | 0.21 | ||
| PLWH, nadir CD4250–500 | 829 | 0.79 (0.59,1.05) | 0.11 | ||
| PLWH, nadir CD4<250 | 743 | 0.92 (0.69,1.23) | 0.59 | ||
| HIV Viral load (copies/mL) | |||||
| Undetectable | 1216 | 1 | |||
| 50–10000 copies | 360 | 0.90 (0.66,1.22) | 0.50 | ||
| >10000 copies | 213 | 1.01 (0.71,1.43) | 0.96 | ||
| Current smoker | |||||
| No | 1226 | 1 | |||
| Yes | 1072 | 1.11 (0.89,1.36) | 0.34 | ||
| Current drinker | |||||
| No | 822 | 1 | |||
| Yes | 1453 | 1.16 (0.93,1.46) | 0.17 | ||
| HPV16 E6 Serum Antibodies * | |||||
| Seronegative | 1389 | 1 | |||
| Seropositive | 49 | 1.66 (0.83,3.32) | 0.15 | ||
| Oncogenic oral HPV | |||||
| Incident | 819 | 1 | 1 | ||
| Prevalent | 1520 | 0.55 (0.44,0.69) | <0.0001 | 0.58 (0.46,0.73) | <0.0001 |
| Oncogenic oral HPV type | |||||
| Non-16 oncogenic HPV type | 1913 | 1 | 1 | ||
| Type 16 | 426 | 1.01 (0.76,1.33) | 0.94 | 1.00 (0.75,1.34) | 0.97 |
| Among those with oral HPV 16 DNA detected only * | |||||
| HPV16 copy number/cell (qPCR) at first detection (carry forward) | |||||
| <0.0001 | 96 | 1 | |||
| 0.0001-<0.001 | 113 | 0.57 (0.31,1.04) | 0.07 | ||
| 0.001-<0.1 | 137 | 0.25 (0.12,0.52) | <0.0001 | ||
| >=0.1 copy/cell | 62 | 0.17 (0.06,0.48) | 0.001 | ||
| Change per qPCR category increase: B and 95% CI | 408 | 0.53 (0.40,0.69) | <0.0001 | 0.63 (0.46,0.86) | 0.004 |
Characteristic in this were run among those with HPV16 DNA only. For the multivariate result, HPV16 copy number was adjusted for gender, age, HIV status and infection type. Other multivariate results in the top of table were not adjusted for HPV16 copy number because the primary model in this table includes all oncogenic oral HPV infections, not just HPV16.
DISCUSSION
With 692 oncHPV infections among 435 individuals followed for up to ten years, this analysis is the longest and largest existing natural history study of oncHPV, to our knowledge. The majority of oncHPV cleared within two years when using less stringent definition of clearance. Notably, we identified a subset of persons with long-term persistent oncHPV DNA detected for ≥8 years, who may be at increased risk of HPV-OPC. This included almost half of those with prevalent oncHPV and oHPV16, representing 6.5% and 1.6% of participants eligible for this study respectively. When oHPV16 is detected at 2 consecutive visits ~72% persist, therefore representing an at-risk group for HPV-OPC. These findings inform future screening strategies of visit intervals, dynamics of infections over time, and factors affecting clearance.
Our findings imply that in a large-scale HPV-OPC screening scenario, discernment between one-time versus continued detection of oncHPV infection is crucial. The definitions explored demonstrate that more stringent definitions are less likely to be impacted by intermittently negative tests and may more accurately capture true risk. When considering populations to screen, age, sex and HIV serostatus may enrich for individuals at increased risk for development of HPV-OPC. (15,16)
Our analysis underscores the importance of detection of prevalent infection (at first visit). While most incident oncHPVs cleared quickly, clearance was significantly reduced for prevalent oncHPV, especially by conservative definitions. This suggests adults with prevalent oncHPV represent a population at increased risk for long-term persistence and HPV-OPC particularly for those with persistent oHPV16, which is responsible for most HPV-OPC in the U.S..(17) Overall, 63.1% of oncHPV infections cleared within 5 years, suggesting low specificity for HPV-OPC (many false-positives) and a low positive predictive rate for oncHPV persistence among persons with a single detection of oncHPV DNA. It may be feasible to identify individuals at greatest risk for HPV-OPC by screening for oHPV16 and following only those individuals with prevalent oHPV16. Not all cases would be identified as there is a broad range of oral rinse sensitivity(18,19) using PCR-based method and, among persistent infections, 28.6% of visits involved a negative HPV test, suggesting some false-negative testing. However, screening for prevalent infections among older adult men (given the role of sex and age) would capture most individuals at risk and potentially have a low false-negative rate. In a screening program, oncHPV testing could be done for oHPV16 DNA alone due to its higher specificity and positive predictive value compared to non-16 oncHPV.(17) Recent research suggests the 10-year risk of OPC following HPV16 E6 seropositivity is high in some groups (e.g., 17% among 50 year old men)(20), therefore combination of oHPV16 and E6 seropositivity may be especially predictive of OPC. Oral rinse and gargle assess the mucosal compartment (i.e., represents the oral and oropharyngeal mucosa), while plasma HPV DNA detection may reflect locoregional or systemic sites. Next generation sequencing, ddPCR, and/or methylation markers(21) may offer improved performance characteristics. Acknowledging limitations of current testing and sampling paradigms, it is clear that persistent oHPV16 is a precursor to HPV-OPC.
The large sample size of this study permitted the calculation of an incidence rate among individuals with HPV-OPC precursor (persistent infection). While previous work used different population-based studies to estimate risk,(22) this is the first study to provide an estimate of HPV-OPC incidence among persons with oHPV16 DNA. Since this estimate is based upon two cases of HPV-OPC and 107 subjects with oHPV16 detected, it warrants caution and further validation. Nonetheless, this represents an important step towards understanding level of risk and natural history among those with oHPV16.
Previous studies of oncHPV natural history involved short follow-up (≤3 years), and fewer infections (<200 oncHPV infections studied for clearance).(23–29) For example, a study of healthy young adults showed 24% of infections persisted one year.(27) Longer-term persistence in our study may be explained by the increased biologic risk due to older age and HIV-related immunosuppression. A shorter study of 114 prevalent oncHPV infections among a similar population (adult men who have sex with men, 40% of whom were living with HIV) used a clear2 definition and reported a similar estimate of 1-year clearance (53%).(29)
oncHPV clearance estimates were similar to previous anogenital HPV studies in comparable populations which reported that most infections clear within 4 years(30–33), with faster median clearance <2 years in some.(31,34) Slower clearance of oral compared to genital HPV might explain later onset (after HPV infection) of HPV-OPC compared to anogenital cancers. Similar to our findings, anogenital HPV clearance is lower in people living with HIV and older individuals.(33–36) While smoking is associated with HPV persistence in several studies, no association was observed here. Higher HPV viral load was less likely to clear, similar to several anogenital HPV studies(37–39), implying that higher or increasing viral load over time may be an important marker for HPV-OPC risk.
Our analysis has several limitations. oncHPV detection was PCR-based which has moderate sensitivity.(18) Newer more costly technologies including genetic sequencing or ddPCR(21), may ultimately improve performance characteristics. Our study population included many people living with HIV and was not representative of the general population but was chosen to enrich for HPV infections and to allow sufficient power for studying long-term persistence of oncHPV and OPC risk, rare events in the general population despite increasing overall OPC incidence.
The opportunity to study oncHPV natural history as represented in this report is unique since HPV vaccination efforts should prevent oncHPV infection in younger age cohorts and thereby preclude such study in the future. However, our findings support the assertion that persistent oncHPV may represent a condition justifying surveillance, where no oropharyngeal precancer has yet been identified. These data underscore that prevalent infections have lower clearance and are of potential concern, especially infections that persist longer than 2 years which may represent a group who may benefit from screening if such a surveillance strategy is developed.
This analysis used a stringent definition of clearance (clear3) to better reflect the dynamic nature of oncHPV DNA detection. Many type-specific infections are intermittently undetectable despite an overall pattern of positivity. Whether an intermittent detection of HPV (“flickering”) represents variation in immunologic control, sampling limitations or intermittent fluctuations at the limits of detection is uncertain. Additionally, a false negative HPV at first visit which was flickering could have misclassified an infection as incident, so some of the persistent incident infections may have been these intermittently detected infections, suggesting clearance estimates for truly incident infections may be underestimated. Using an overly lenient definition of clearance could therefore underestimate risk. It is possible that intermittent HPV presence within an individual may represent clearance and new acquisition of the same HPV type, however, data from this and previous studies suggests new infection of the same type is less likely given the “flickering” patterns were apparent even among subjects who reported no new sexual partners. Older clearance definitions used less sensitive PCR methods, but with assay sensitivity advancement, flickering results may be more common. Hence, clearance definitions may need revisions to account for that. In sum, this study, the longest and largest existing natural history study of oncHPV, revealed important risk factors for OPC development and estimated OPC incidence rate among those at risk. This highlights the importance of persistent oHPV16 infection and the possible role for oHPV16 for future targeted screening programs.
Supplementary Material
ACKNOWLEDGEMENTS
Funding Sources:
This work was supported by the National Institute of Dental and Craniofacial Research [R35DE026631, U01HL146193, R01DE021395; to G D’Souza]
The contents of this publication are solely the responsibility of the authors and do not represent the official views of the National Institutes of Health (NIH). MWCCS (Principal Investigators): Atlanta CRS (Ighovwerha Ofotokun, Anandi Sheth, and Gina Wingood), U01-HL146241; Baltimore CRS (Todd Brown and Joseph Margolick), U01-HL146201; Bronx CRS (Kathryn Anastos, David Hanna, and Anjali Sharma), U01-HL146204; Brooklyn CRS (Deborah Gustafson and Tracey Wilson), U01-HL146202; Data Analysis and Coordination Center (Gypsyamber D’Souza, Stephen Gange and Elizabeth Topper), U01-HL146193; Chicago-Cook County CRS (Mardge Cohen and Audrey French), U01-HL146245; Chicago-Northwestern CRS (Steven Wolinsky, Frank Palella, and Valentina Stosor), U01-HL146240; Northern California CRS (Bradley Aouizerat, Jennifer Price, and Phyllis Tien), U01-HL146242; Los Angeles CRS (Roger Detels and Matthew Mimiaga), U01-HL146333; Metropolitan Washington CRS (Seble Kassaye and Daniel Merenstein), U01-HL146205; Miami CRS (Maria Alcaide, Margaret Fischl, and Deborah Jones), U01-HL146203; Pittsburgh CRS (Jeremy Martinson and Charles Rinaldo), U01-HL146208; UAB-MS CRS (Mirjam-Colette Kempf, Jodie Dionne-Odom, Deborah Konkle-Parker, and James B. Brock), U01-HL146192; UNC CRS (Adaora Adimora and Michelle Floris-Moore), U01-HL146194. The MWCCS is funded primarily by the National Heart, Lung, and Blood Institute (NHLBI), with additional co-funding from the Eunice Kennedy Shriver, National Institute Of Child Health & Human Development (NICHD), National Institute On Aging (NIA), National Institute Of Dental & Craniofacial Research (NIDCR), National Institute Of Allergy And Infectious Diseases (NIAID), National Institute Of Neurological Disorders And Stroke (NINDS), National Institute Of Mental Health (NIMH), National Institute On Drug Abuse (NIDA), National Institute Of Nursing Research (NINR), National Cancer Institute (NCI), National Institute on Alcohol Abuse and Alcoholism (NIAAA), National Institute on Deafness and Other Communication Disorders (NIDCD), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute on Minority Health and Health Disparities (NIMHD), and in coordination and alignment with the research priorities of the National Institutes of Health, Office of AIDS Research (OAR). MWCCS data collection is also supported by UL1-TR000004 (UCSF CTSA), UL1-TR003098 (JHU ICTR), UL1-TR001881 (UCLA CTSI), P30-AI-050409 (Atlanta CFAR), P30-AI-073961 (Miami CFAR), P30-AI-050410 (UNC CFAR), P30-AI-027767 (UAB CFAR), P30-MH-116867 (Miami CHARM), UL1-TR001409 (DC CTSA), KL2-TR001432 (DC CTSA), and TL1-TR001431 (DC CTSA).
The authors gratefully acknowledge the contributions of the study participants and dedication of the staff at the MWCCS sites. We would like to also acknowledge the National Program of Cancer Registries of the Centers for Disease Control and Prevention (CDC) for the funds that helped support the collection and availability of the cancer registry data and thank the following state cancer registries for their help: AL, CA, FL, GA, IL, MD, MS, NY, NC, PA, and VA. The authors assume full responsibility for analyses and interpretations of these data.
Conflict of Interest:
TW serves on advisory boards for MSD (Merck) Sharp & Dohme. CDL serves on an advisory board for Theratechnologies, LLC and receives grant funding from Merck. DW serves as External Advisory Board Member in UCLA Clinical Translational Science Institute, and as Member of California HPV Vaccination Roundtable and National HPV Roundtable. All other authors have no conflicts.
References
- 1.Agalliu I, Gapstur S, Chen Z, Wang T, Anderson RL, Teras L, et al. Associations of Oral α-, β-, and γ-Human Papillomavirus Types With Risk of Incident Head and Neck Cancer. JAMA Oncol. 2016;2:599–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.D’Souza G, Clemens G, Strickler HD, Wiley DJ, Troy T, Struijk L, et al. Long-term Persistence of Oral HPV Over 7 Years of Follow-up. JNCI Cancer Spectr. 2020;4:pkaa047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Pierce Campbell CM, Kreimer AR, Lin H-Y, Fulp W, O’Keefe MT, Ingles DJ, et al. Long-term persistence of oral human papillomavirus type 16: the HPV Infection in Men (HIM) study. Cancer Prev Res Phila Pa. 2015;8:190–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lydiatt WM, Patel SG, O’Sullivan B, Brandwein MS, Ridge JA, Migliacci JC, et al. Head and Neck cancers-major changes in the American Joint Committee on cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67:122–37. [DOI] [PubMed] [Google Scholar]
- 5.Herrero R, Quint W, Hildesheim A, Gonzalez P, Struijk L, Katki HA, et al. Reduced prevalence of oral human papillomavirus (HPV) 4 years after bivalent HPV vaccination in a randomized clinical trial in Costa Rica. PloS One. 2013;8:e68329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fakhry C, Blackford AL, Neuner G, Xiao W, Jiang B, Agrawal A, et al. Association of Oral Human Papillomavirus DNA Persistence With Cancer Progression After Primary Treatment for Oral Cavity and Oropharyngeal Squamous Cell Carcinoma. JAMA Oncol. 2019;5:985–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rettig EM, Wentz A, Posner MR, Gross ND, Haddad RI, Gillison ML, et al. Prognostic Implication of Persistent Human Papillomavirus Type 16 DNA Detection in Oral Rinses for Human Papillomavirus-Related Oropharyngeal Carcinoma. JAMA Oncol. 2015;1:907–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fakhry C, Sugar E, D’Souza G, Gillison M. Two-week versus six-month sampling interval in a short-term natural history study of oral HPV infection in an HIV-positive cohort. PloS One. 2010;5:e11918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schiffman M, Kjaer SK. Chapter 2: Natural history of anogenital human papillomavirus infection and neoplasia. J Natl Cancer Inst Monogr. 2003;14–9. [DOI] [PubMed] [Google Scholar]
- 10.Beachler DC, Weber KM, Margolick JB, Strickler HD, Cranston RD, Burk RD, et al. Risk factors for oral HPV infection among a high prevalence population of HIV-positive and at-risk HIV-negative adults. Cancer Epidemiol Biomark Prev Publ Am Assoc Cancer Res Cosponsored Am Soc Prev Oncol. 2012;21:122–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.MOUTH Study - Featured Clinical Trial: Johns Hopkins Kimmel Cancer Center [Internet]. Men Women Offer. Underst. Throat HPV MOUTH Study Support. NIDCR R35 DE026631. [cited 2022 Aug 31]. Available from: http://bit.ly/MOUTHStudy [Google Scholar]
- 12.D’Souza G, Tewari SR, Troy T, Waterboer T, Struijk L, Castillo R, et al. Prevalence of oral and blood oncogenic human papillomavirus biomarkers among an enriched screening population: Baseline results of the MOUTH study. Cancer. 2023;129:2372–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Waterboer T, Sehr P, Michael KM, Franceschi S, Nieland JD, Joos TO, et al. Multiplex human papillomavirus serology based on in situ-purified glutathione s-transferase fusion proteins. Clin Chem. 2005;51:1845–53. [DOI] [PubMed] [Google Scholar]
- 14.Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA. 1983;249:1743–5. [PubMed] [Google Scholar]
- 15.Lorenzoni V, Chaturvedi AK, Vignat J, Laversanne M, Bray F, Vaccarella S. The Current Burden of Oropharyngeal Cancer: A Global Assessment Based on GLOBOCAN 2020. Cancer Epidemiol Biomark Prev Publ Am Assoc Cancer Res Cosponsored Am Soc Prev Oncol. 2022;31:2054–62. [DOI] [PubMed] [Google Scholar]
- 16.Chaturvedi AK, Anderson WF, Lortet-Tieulent J, Curado MP, Ferlay J, Franceschi S, et al. Worldwide trends in incidence rates for oral cavity and oropharyngeal cancers. J Clin Oncol Off J Am Soc Clin Oncol. 2013;31:4550–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fakhry C, Fung N, Tewari SR, D’Souza G. Unique role of HPV16 in predicting oropharyngeal cancer risk more than other oncogenic oral HPV infections. Oral Oncol. 2020;111:104981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gipson BJ, Robbins HA, Fakhry C, D’Souza G. Sensitivity and specificity of oral HPV detection for HPV-positive head and neck cancer. Oral Oncol. 2018;77:52–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Xie DX, Kut C, Quon H, Seiwert TY, D’Souza G, Fakhry C. Clinical Uncertainties of Circulating Tumor DNA in Human Papillomavirus-Related Oropharyngeal Squamous Cell Carcinoma in the Absence of National Comprehensive Cancer Network Guidelines. J Clin Oncol Off J Am Soc Clin Oncol. 2023;41:2483–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Robbins HA, Ferreiro-Iglesias A, Waterboer T, Brenner N, Nygard M, Bender N, et al. Absolute Risk of Oropharyngeal Cancer After an HPV16-E6 Serology Test and Potential Implications for Screening: Results From the Human Papillomavirus Cancer Cohort Consortium. J Clin Oncol. 2022;40:3613–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mattox AK, D’Souza G, Khan Z, Allen H, Henson S, Seiwert TY, et al. Comparison of next generation sequencing, droplet digital PCR, and quantitative real-time PCR for the earlier detection and quantification of HPV in HPV-positive oropharyngeal cancer. Oral Oncol. 2022;128:105805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.D’Souza G, McNeel TS, Fakhry C. Understanding personal risk of oropharyngeal cancer: risk-groups for oncogenic oral HPV infection and oropharyngeal cancer. Ann Oncol Off J Eur Soc Med Oncol. 2017;28:3065–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Antonsson A, de Souza M, Wood ZC, Carroll A, Van K, Paterson L, et al. Natural history of oral HPV infection: Longitudinal analyses in prospective cohorts from Australia. Int J Cancer. 2021;148:1964–72. [DOI] [PubMed] [Google Scholar]
- 24.Rintala M, Grénman S, Puranen M, Syrjänen S. Natural history of oral papillomavirus infections in spouses: a prospective Finnish HPV Family Study. J Clin Virol Off Publ Pan Am Soc Clin Virol. 2006;35:89–94. [DOI] [PubMed] [Google Scholar]
- 25.Beachler DC, D’Souza G, Sugar EA, Xiao W, Gillison ML. Natural history of anal vs oral HPV infection in HIV-infected men and women. J Infect Dis. 2013;208:330–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Beachler DC, Lang Kuhs KA, Struijk L, Schussler J, Herrero R, Porras C, et al. The Natural History of Oral Human Papillomavirus in Young Costa Rican Women. Sex Transm Dis. 2017;44:442–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.D’Souza G, Wentz A, Kluz N, Zhang Y, Sugar E, Youngfellow RM, et al. Sex Differences in Risk Factors and Natural History of Oral Human Papillomavirus Infection. J Infect Dis. 2016;213:1893–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Riddell J, Brouwer AF, Walline HM, Campredon LP, Meza R, Eisenberg MC, et al. Oral human papillomavirus prevalence, persistence, and risk-factors in HIV-positive and HIV-negative adults. Tumour Virus Res. 2022;13:200237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.van Aar F, Mooij SH, van der Sande MAB, Meijer CJLM, King AJ, Verhagen DWM, et al. Twelve-month incidence and clearance of oral HPV infection in HIV-negative and HIV-infected men who have sex with men: the H2M cohort study. BMC Infect Dis. 2014;14:668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Patel P, Bush T, Conley L, Unger ER, Darragh TM, Henry K, et al. Prevalence, Incidence, and Clearance of Human Papillomavirus Types Covered by Current Vaccines in Men With Human Immunodeficiency Virus in the SUN Study. J Infect Dis. 2020;222:234–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Koshiol JE, Schroeder JC, Jamieson DJ, Marshall SW, Duerr A, Heilig CM, et al. Time to clearance of human papillomavirus infection by type and human immunodeficiency virus serostatus. Int J Cancer. 2006;119:1623–9. [DOI] [PubMed] [Google Scholar]
- 32.Darwich L, Cañadas M-P, Videla S, Coll J, Molina-López RA, Sirera G, et al. Prevalence, clearance, and incidence of human papillomavirus type-specific infection at the anal and penile site of HIV-infected men. Sex Transm Dis. 2013;40:611–8. [DOI] [PubMed] [Google Scholar]
- 33.Rowhani-Rahbar A, Hawes SE, Sow PS, Toure P, Feng Q, Dem A, et al. The Impact of HIV Status and Type on the Clearance of Human Papillomavirus Infection among Senegalese Women. J Infect Dis. 2007;196:887–94. [DOI] [PubMed] [Google Scholar]
- 34.Wei F, Goodman MT, Xia N, Zhang J, Giuliano AR, D’Souza G, et al. Incidence and Clearance of Anal Human Papillomavirus Infection in 16 164 Individuals, According to Human Immunodeficiency Virus Status, Sex, and Male Sexuality: An International Pooled Analysis of 34 Longitudinal Studies. Clin Infect Dis Off Publ Infect Dis Soc Am. 2023;76:e692–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Taylor S, Bunge E, Bakker M, Castellsagué X. The incidence, clearance and persistence of non-cervical human papillomavirus infections: a systematic review of the literature. BMC Infect Dis. 2016;16:293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Branca M, Garbuglia AR, Benedetto A, Cappiello T, Leoncini L, Migliore G, et al. Factors predicting the persistence of genital human papillomavirus infections and PAP smear abnormality in HIV-positive and HIV-negative women during prospective follow-up. Int J STD AIDS. 2003;14:417–25. [DOI] [PubMed] [Google Scholar]
- 37.Wissing MD, Louvanto K, Comète E, Burchell AN, El-Zein M, Rodrigues A, et al. Human Papillomavirus Viral Load and Transmission in Young, Recently Formed Heterosexual Couples. J Infect Dis. 2019;220:1152–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Senkomago V, Backes DM, Hudgens MG, Poole C, Meshnick SR, Agot K, et al. Higher HPV16 and HPV18 Penile Viral Loads Are Associated With Decreased Human Papillomavirus Clearance in Uncircumcised Kenyan Men. Sex Transm Dis. Lippincott Williams & Wilkins; 2016;43:572–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kim J, Song S, Jin C, Lee J, Lee N, Lee K. Factors Affecting the Clearance of High-Risk Human Papillomavirus Infection and the Progression of Cervical Intraepithelial Neoplasia. J Int Med Res. SAGE Publications Ltd; 2012;40:486–96. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data generated in this study are available upon request from the corresponding author.