Keywords
oral HPV, HIV, India
This article is included in the Global Public Health gateway.
This article is included in the Pathogens gateway.
oral HPV, HIV, India
People living with HIV (PLHIV) have a 2–3 times higher odds of oral human papillomavirus (HPV) prevalence compared to HIV-uninfected individuals.1 This is estimated to increase their risk of HPV-associated head and neck cancers (HNCs) 1.2–2.1 times relative to HIV-uninfected individuals.2 However, estimates of oral HPV prevalence among PLHIV are mostly derived from high income countries (HICs) and data from low- and middle-income countries (LMICs) remain scarce, despite the burden of HIV in LMICs being considerably higher than HICs.3
Globally, India has the third highest number of PLHIV (approximately 2.3 million).4 Only one study from the country has previously reported on oral HPV prevalence among PLHIV, restricted to a key population, men who have sex with men (MSM).5 To develop comprehensive strategies for oral HPV prevention that are geographically relevant for PLHIV, it is also important to report on the prevalence of oral HPV among PLHIV that are not key populations. In this manuscript, we estimate the prevalence of oral HPV and assess associated factors among PLHIV in Pune, India.
The Ethics Committee of Byramjee Jeejeebhoy Government Medical College – Sassoon General Hospitals (BJGMC-SGH) (Date of approval: 19 September 2016) and the Institutional Review Board of Johns Hopkins University (IRB00118708, Date of approval: 13 December 2016) approved the study. It should be noted that the Ethics Committee of BJGMC does not issue ethical approval numbers, nor does it have an Ethics Number itself.
We used data from a parent oral cancer screening study conducted between June 2017 and June 2019 at Byramjee Jeejeebhoy Government Medical College – Sassoon General Hospitals (BJGMC-SGH), a tertiary health care centre in Pune, India.6 The parent study enrolled 1,234 participants, including 633 HIV-uninfected individuals and 601 PLHIV. For this analysis, data from HIV-uninfected participants and PLHIV with missing data for oral HPV (n=19) were excluded. Details of the study including enrolment procedures are provided elsewhere.6 This analysis included data of PLHIV≥21 years of age attending the ART centre for care, with no prior history of HNCs who provided a 30 seconds oral rinse and gargle sample, and written informed consent.
Sociodemographic (age, biological sex), behavioural factors (smoking, smokeless tobacco (SLT) and alcohol use), sexual history (men who have sex with men (MSM), oral sex, lifetime sexual partners), clinical history (CD4 counts, duration on antiretroviral therapy, suspected oral potentially malignant disorder (OPMD), HPV results including genotype) data were extracted from the database of the parent study.6 These data were collected by trained non-medical health care professionals of the study team in the parent study.
Assuming that the prevalence of oral HPV was 24%, consistent with the previous study that reported on oral HPV prevalence among Indian MSM living with HIV,5 our sample size of 582 was >95% powered at an alpha of 0.05.
The parent study6 used polymerase chain reaction (PCR) and next-generation sequencing (NGS) to obtain the prevalence of oral HPV. PCR and NGS were performed at a private laboratory (GenePath Diagnostics, Pune). These procedures are described below.
Reagents
Platinum Taq DNA polymerase (Invitrogen, Cat No. 10966034), HotStarTaq DNA polymerase (Qiagen, Cat No. 203205), dNTP (deoxyribonucleotide triphosphate) 25 micromolar (mM) (Lucigen, Cat No. D59104) were used. [Note - Invitrogen PCR buffer and MgCl2 are provided by Invitrogen along with Platinum Taq DNA polymerase].
Thermal cycling conditions, primers used and model of PCR machine
A nested PCR approach targeting the L1 gene was used. The outer PCR was carried out using an equimolar mixture of each PGMY09 and PGMY11 primer pool7 at a concentration of 80 nanomolar (nM) each, which generated a 450 base pairs (bp) product. This product was used as a template for the inner PCR. The inner PCR was carried out using custom tagged GP5+ and GP6+ primers7 at a final concentration 200 nM each, which generated a 150 bp product. The tagged inner PCR products were used for library generation where specific barcodes were incorporated for each sample. The library was sequenced on an Illumina MiSeq or a NextSeq next generation sequencer in a shared run using the Illumina V2 2*250 (MiSeq) or Mid-output 2*150 (NextSeq) paired-end sequencing by synthesis chemistry. The resulting .fastq raw read files were aligned against HPV DNA database using in-house bioinformatics pipeline and read counts generated were analysed for presence/absence and type of HPV DNA in each sample.
The outer PCR was carried out at 20 microlitre (μl) reaction volume, with 1X Invitrogen PCR buffer, 4 mM MgCl2, 0.2 mM dNTP and a blend of two polymerases, 0.5U of Platinum Taq DNA polymerase (Invitrogen) and 0.5U of HotStarTaq DNA polymerase (Qiagen). The outer PCR was carried out on Veriti 96-well thermal cycler (Applied Biosystems) with initial incubation at 37°C, initial denaturation at 95°C for 15 min, followed by 35 cycles of denaturation at 95°C for 20 sec, annealing at 56.5°C for 30 sec, extension at 72°C for 1 min. The amplified outer PCR products were diluted 100 times with ultrapure nuclease free PCR grade water and used as a template for inner PCR.
For the inner PCR, custom tagged GP5+ and GP6+ primers7 were used at a final concentration of 200 nM each. The inner PCR was carried out at 20 μl reaction volume with 1X Invitrogen PCR buffer, 4 mM MgCl2, 0.2 mM dNTP, 0.5 mg/ml BSA, and a blend of two polymerases, 0.5U of Platinum Taq DNA polymerase (Invitrogen) and 0.5U of HotStarTaq DNA polymerase (Qiagen). The inner PCR was carried out on Veriti 96-well thermal cycler (Applied Biosystems) with initial incubation at 37°C, initial denaturation at 95°C for 15 min, followed by eight cycles of denaturation at 95°C for 20 sec, ramp up annealing from 40°C to 72°C for 30 sec, extension at 72°C for 30 sec, followed by 27 cycles of denaturation at 95°C for 20 sec and extension at 72°C for 1 min.
PCR (and NGS) results were treated as the outcome, classified into three groups: 1) no HPV; 2) high risk or oncogenic HPV (hrHPV i.e., if they belonged to any of the following genotypes HPV16, HPV18, HPV31, HPV33, HPV35, HPV39, HPV45, HPV51, HPV52, HPV56, HPV58, HPV59, HPV66, HPV68)8; and 3) low risk HPV (6/11, 7, 13, 27, 32/42, 34, 43/91, 44, 53, 62, 71, 72, 73, 81, 82, 83, 84, 86/87, 89/102, 90/106, 107, 114, and 120).8
Medians and proportions across the outcome were compared using Kruskal Wallis and Fisher’s exact tests, respectively, for different sociodemographic, behavioural, sexual history and clinical predictors (Table 1). We used multinomial logistic regression to obtain prevalence ratios (PRs) comparing lrHPV and hrHPV groups (comparison groups) against the no HPV group (reference group) for different predictors. Multivariable models were adjusted for variables chosen a priori as predictors of oral HPV and that had a minimum of five participants when cross tabulated against the outcome. These were age, sex, CD4 counts and duration on ART.
Characteristics | Total N (%, IQR) | hrHPV* N (%, IQR) | lrHPV* N (%, IQR) | No HPV* N (%, IQR) | PR (95% CI) Ref (No HPV) Comp (hrHPV) | PR (95% CI) Ref (No HPV) Comp (lrHPV) | aPR (95% CI) Ref (No HPV) Comp (hrHPV) | aPR (95% CI) Ref (No HPV) Comp (lrHPV) |
---|---|---|---|---|---|---|---|---|
Total N | 582 | 26 (4.5) | 20 (3.4) | 536 (92.1) | - | - | - | - |
Median age in years (IQR) | 40 (34 – 46) | 38.5 (31 – 48) | 43.5 (40 – 47.5) | 40 (34 – 46) | 0.99 (0.95, 1.04) | 1.05 (1.01, 1.10) | 0.99 (0.95, 1.04) | 1.04 (1.00, 1.09) |
Biological sex | ||||||||
Male | 312 (53.6) | 14 (53.8) | 15 (75.0) | 283 (52.8) | Ref | Ref | Ref | Ref |
Female | 270 (46.4) | 12 (46.2) | 5 (25.0) | 253 (47.2) | 0.96 (0.44, 2.11) | 0.37 (0.13, 1.04) | 0.97 (0.43, 2.21) | 0.50 (0.17, 1.45) |
Ever smoked | ||||||||
No | 475 (81.8) | 20 (76.9) | 14 (70.0) | 441 (82.4) | Ref | Ref | - | - |
Yes | 106 (18.2) | 6 (23.1) | 6 (30.0) | 94 (17.6) | 1.41 (0.55, 3.60) | 2.01 (0.75, 5.37) | ||
Ever consumed alcohol | ||||||||
No | 377 (64.8) | 18 (69.2) | 10 (50.0) | 349 (65.1) | Ref | Ref | - | - |
Yes | 205 (35.2) | 8 (30.8) | 10 (50.0) | 187 (34.9) | 0.83 (0.35, 1.94) | 1.87 (0.76, 4.56) | ||
Ever used SLTa | ||||||||
No | 343 (58.9) | 19 (73.1) | 13 (65.0) | 311 (58.0) | Ref | Ref | - | - |
Yes | 239 (41.1) | 7 (26.9) | 7 (35.0) | 225 (42.0) | 0.51 (0.21, 1.23) | 0.74 (0.29, 1.89) | ||
Ever smoked, used SLT or consumed alcohol | ||||||||
No | 281 (48.3) | 16 (61.5) | 10 (50.0) | 349 (65.1) | Ref | Ref | - | - |
Yes | 301 (51.7) | 10 (38.5) | 10 (50.0) | 187 (34.9) | 0.83 (0.35, 1.94) | 1.87 (0.76,4.56) | ||
Ever given oral sex | ||||||||
No | 521 (89.5) | 24 (92.3) | 16 (80.0) | 481 (89.7) | Ref | Ref | - | - |
Yes | 61 (10.5) | 2 (7.7) | 4 (20.0) | 55 (10.3) | 0.73 (0.17, 3.17) | 2.19 (0.71, 6.77) | ||
Lifetime multiple sexual partners | ||||||||
No | 431 (74.2) | 22 (84.6) | 12 (60.0) | 397 (74.2) | Ref | Ref | - | - |
Yes | 150 (25.8) | 4 (15.4) | 8 (40.0) | 138 (25.8) | 0.52 (0.18, 1.54) | 1.92 (0.77, 4.79) | ||
Self identifies as MSM b | ||||||||
No | 568 (97.6) | 26 (100.0) | 20.0 (100.0) | 522 (97.4) | - | - | - | - |
Yes | 14 (2.4) | 0 (0) | 0 (0) | 14 (2.61) | ||||
Median time- updated CD4 counts (cells/mm3)c | 510 (338 – 700) | 475 (242 – 651) | 467 (311 – 611) | 512 (349 – 708) | 0.97 (0.90, 1.05) | 0.93 (0.85, 1.02) | 0.97 (0.90, 1.05) | 0.96 (0.87, 1.05) |
Median duration on ART (in years)d | 8 (4-12) | 8.5 (2 – 16) | 6.5 (4 – 9.5) | 8 (4-12) | 1.03 (0.97, 1.09) | 0.95 (0.87, 1.03) | 1.05 (0.65, 1.72) | 0.68 (0.38, 1.23) |
Suspected OPMDe | ||||||||
No | 470 (80.8) | 22 (84.6) | 18 (90.0) | 430 (80.2) | Ref | Ref | - | - |
Yes | 112 (19.2) | 4 (15.4) | 2 (10.0) | 106 (19.8) | 0.74 (0.25, 2.18) | 0.45 (0.10, 1.97) |
* There are no significant differences between the medians or proportions across no HPV, hrHPV (high risk HPV) and lrHPV (low risk HPV) groups (p-value associated with Kruskal- Wallis and Fisher’s exact test >0.10 for all characteristics.
a SLT: Smokeless tobacco, which are forms of tobacco that are not burned. Locally available forms available: khaini (tobacco + slaked lime); gutka (tobacco + areca nut + slaked lime); mishri (roasted powdered tobacco); paan (tobacco + areca nut + slaked lime + condiments, wrapped in a betel leaf); paan masala; snuff.
We performed three sensitivity analyses, comparing primary results with: 1) Findings in which CD4 counts were categorized into <200 and ≥200 cells/mm; 2) Findings from models in which PRs were obtained for a binary outcome i.e., no HPV and any HPV (collapsing hrHPV and lrHPV into a single group); 3) Findings from regression models in which imputations using chained equations for those with missing HPV were performed.
Statistical significance was set to a two-sided p-value of 0.05. All analyses were performed in Stata 17.0 (RRID:SCR_012763) (free alternative, RStudio).
A total of 582 PLHIV were enrolled. The median age was 40 years (IQR: 34–46), 46.4% (n=270) were biologically female, 18.2% (n=106) had smoked, 41.1% (n=239) had used smokeless tobacco (SLT) and 35.2% (n=205) had consumed alcohol. Approximately 2.5% (n=14) self-identified as MSM, 10.5% (n=61) had given oral sex and 25.8% (n=150) had multiple sexual partners, in their lifetime. Median time updated CD4 counts were 510 cells/mm3 (IQR: 338–700) and median duration on ART was eight years (IQR: 4–12). None of the participants were vaccinated against HPV.
The prevalence of hrHPV was 4.5% (n=26) and lrHPV was 3.4% (n=20). Of those with hrHPV, 76.9% (n=20) had HPV16. Approximately 40% of those with hrHPV (n=8) and 50% (n=10) of those with lrHPV had never smoked, used SLT or consumed alcohol. The distribution of hrHPV was HPV16: n=20, HPV18: n=2, HPV35: n=1, HPV66: n=2 and HPV68: n=1; and that of lrHPV was HPV7: n=1, HPV13: n=1, HPV27: n=1, HPV32: n=1, HPV42: n=1, HPV44: n=3, HPV72: n=3, HPV81: n=1, HPV84: n=1, HPV90: n=1, HPV107: n=4, HPV120: n=2 (Table 1).
The prevalence of hrHPV and lrHPV were not statistically significantly different (i.e., p>0.05) by age, biological sex, tobacco and alcohol use habits, and sexual and clinical history.
In the univariate analysis, relative to the no HPV group, a unit increase in age was associated with 5% increase in the prevalence of lrHPV (PR: 1.05; 95% CI: 1.01, 1.10). There were no predictors significantly associated with hrHPV in univariate analysis. When adjusted for age, sex, time updated CD4 counts and ART duration, no predictors were significantly associated with either lrLPV or hrHPV (Table 1).
Sensitivity analyses findings were consistent with results presented in Table 1. However, when CD4 counts were categorized into <200 and ≥200 cells/mm3, participants with <200 cells/mm3 had a 3.14- (95% CI: 1.20, 8.22) and 3.12- (95% CI: 1.19, 8.20) fold higher prevalence of hrHPV in univariate and multivariable analyses, respectively, but associations with lrHPV were non-significant. We do not present these as primary findings because there were no female participants with CD4 counts <200 cells/mm3 in the hrHPV group and multivariable results are likely an extrapolation by the model.
Using oral rinse samples, we found the overall prevalence of any HPV to be 7.6%, lrHPV to be 3.4% and hrHPV to be 4.5% among PLHIV. To our knowledge this is the first study from India to report on oral HPV among PLHIV that are not exclusively MSM and the largest study from the country to date to report on the prevalence of oral HPV.9,10
We found combined oral HPV (i.e., any oral HPV: lrHPV and hrHPV) and hrHPV (only) prevalence to be lower than what has been reported among studies in PLHIV from the United States.1 However, the sample size of the present study is larger than most of these studies. Compared to the study conducted among MSM living with HIV in India, our overall prevalence of any oral HPV was lower (7.6% versus 23.7%), hrHPV prevalence was higher (4.5% versus 2.4%) and hrHPV was almost exclusively HPV16 (77% versus 0%).5 These findings indicate the heterogeneity of oral HPV prevalence and genotypes among different subgroups of PLHIV in India and the need for larger studies to better characterize oral HPV in them.
Most previous studies in India have only reported on oral HPV among HIV-uninfected individuals with HNCs. The prevalence of oral HPV in these studies varies between 0–79%, with HPV16 being the most common genotype.9,10 More than three-fourths of those with hrHPV in our study had HPV16. Further, 40% of those with hrHPV reported to have never smoked, used SLT or consumed alcohol, which are traditional risk factors for HNCs. Oral HPV screening is not routinely performed as part of HIV care at present. Given the strong association between HPV16 and HNCs,1 and the increased vulnerability of PLHIV for HNCs,1 oral HPV assessment for PLHIV must be considered.
We found no association with CD4 counts in our primary analysis. This may be due to the high median CD4 counts of the study population. Immunosuppression is an important risk factor for oral HPV,1,8 and our sensitivity analysis indicates that those with CD4 counts <200 cells/mm3 may be more likely to have prevalent hrHPV. A previous publication from our site showed that a large proportion of PLHIV consistently presented with CD4 counts <200 cells/mm3 over a 12-year period (2005–2017).11 This underlines the need to facilitate earlier engagement of PLHIV in HIV care, to reduce their risk of oral HPV acquisition.
There are several limitations in our study that merit discussion. We did not measure HIV viral loads and are unable to comment on the association between viral suppression and oral HPV prevalence. We did not find associations between oral HPV with either multiple sexual partners or oral sex. This is likely due to participants underreporting these behaviours, driven by social desirability. We are committed to improving on this aspect and have organized a series of workshops for sexual history assessment. While this is the largest study till date to report on oral HPV from India, we were not powered to detect significant associations especially in multivariable models. Our findings may also not be generalizable to other PLHIV living elsewhere in India.
PLHIV are a vulnerable group for HPV-associated HNCs and those living in LMICs may be particularly at risk because oral HPV screening is not routinely performed as part of HIV care. We reiterate the need for larger studies to better understand oral HPV prevalence and associated risk factors among PLHIV in India, simultaneously with the scaling-up of oral HPV screening measures among them.
Data cannot be shared publicly because it comes from people living with HIV in India, many of whom have not disclosed their HIV status except to their HIV health care provider and spouses. Data are available from the Ethics Committee - Byramjee Jeejeebhoy Medical College & Sassoon General Hospitals, Pune, India (contact via email @bjmcecirb@gmail.com) for researchers who meet the criteria for access to confidential data.
We would like to thank the non-medical health care workers involved: Suhasini Surwase, Archana Pawar and Kanta Zarekar; the data entry operator: Rohini Kamble; all the ART staff and the study participants without whom this work would not be possible.
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Is the work clearly and accurately presented and does it cite the current literature?
Yes
Is the study design appropriate and is the work technically sound?
No
Are sufficient details of methods and analysis provided to allow replication by others?
Yes
If applicable, is the statistical analysis and its interpretation appropriate?
Yes
Are all the source data underlying the results available to ensure full reproducibility?
Partly
Are the conclusions drawn adequately supported by the results?
Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Viral Immunologist
Is the work clearly and accurately presented and does it cite the current literature?
Yes
Is the study design appropriate and is the work technically sound?
Yes
Are sufficient details of methods and analysis provided to allow replication by others?
Partly
If applicable, is the statistical analysis and its interpretation appropriate?
I cannot comment. A qualified statistician is required.
Are all the source data underlying the results available to ensure full reproducibility?
Partly
Are the conclusions drawn adequately supported by the results?
Partly
References
1. Riddell J, Brouwer AF, Walline HM, Campredon LP, et al.: Oral human papillomavirus prevalence, persistence, and risk-factors in HIV-positive and HIV-negative adults.Tumour Virus Res. 2022; 13: 200237 PubMed Abstract | Publisher Full TextCompeting Interests: No competing interests were disclosed.
Reviewer Expertise: HPV, STIs
Is the work clearly and accurately presented and does it cite the current literature?
Yes
Is the study design appropriate and is the work technically sound?
Yes
Are sufficient details of methods and analysis provided to allow replication by others?
Yes
If applicable, is the statistical analysis and its interpretation appropriate?
Yes
Are all the source data underlying the results available to ensure full reproducibility?
Yes
Are the conclusions drawn adequately supported by the results?
Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Oral cancers, HPV, Minority populations, racism
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | |||
---|---|---|---|
1 | 2 | 3 | |
Version 1 14 Mar 23 |
read | read | read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)