Genotype Distribution of Human Papillomavirus (HPV) Types in Oral Gargle Specimens among Men Living with HIV in Mexico, Brazil, and Puerto Rico: a Cross-Sectional Study

Summary

Objectives:

Assess the distribution of oral HPV genotypes and examine their associations with age and HIV-related factors among men living with HIV.

Methods:

This cross-sectional study analyzed baseline data from a randomized, double-blinded, placebo-controlled Phase III trial (‘ULACNet-201’; NCT04255849) assessing the 9vHPV vaccine’s efficacy in PLWH. Participants included men aged 20–50 years from Brazil, Mexico, and Puerto Rico, on antiretroviral therapy for ≥6 months. Oral HPV genotypes were assessed using the SPF₁₀ PCR-DEIA-LIPA₂₅ on oral gargles. Demographics and HIV-related characteristics were collected via questionnaires, and clinical assessments were conducted. ANOVA and chi-square tests assessed associations with age groups and HPV infection.

Results:

Among 700 participants, oral HPV was detected in 27.9%, with HR-HPV detection at 11.0%. 4vHPV and 9vHPV types were detected in 4.9% and 8.9% participants, respectively. The most detected HR-HPV types were HPV 16 (2.4%), HPV 33 (2.0%), and HPV 52 (2.0%). No significant age-specific differences in oral HPV detected were observed. Higher detection rates of any HPV and HR-HPV were observed among participants with baseline CD4 counts below 200 cells/mm³ or a history of AIDS-defining conditions.

Conclusions:

Oral HPV genotyping in men with HIV reveals distinct oncogenic patterns, underscoring the need to monitor long-term OPSCC risk.

Introduction

In 2018, human papillomavirus (HPV) accounted for 690,000 cancer cases worldwide, primarily affecting the cervix uteri, vulva, vagina, penis, anus, and head and neck.¹ HPV is significantly more likely to cause cancer in the oropharynx than in other head and neck regions.² Meta-analyses have shown that HPV is detected in 44.2% (95% CI 38.0–50.6) of oropharyngeal squamous cell carcinoma (OPSCC) cases globally, with regional variations³.

Over the past three decades, the incidence of HPV-related OPSCC has increased worldwide, despite declining smoking habits.² OPSCC cases are expected to grow, especially among older men due to a birth cohort effect.² In 2018, HPV was linked to 42,000 of the 140,000 OPSCC cases globally, disproportionately affecting men.¹ Changes of sexual behavior and increased exposure to oral HPV 16 are key factors contributing to this trend.² Given the absence of effective screening modalities for early OPSCC detection, understanding oral HPV prevalence is crucial for developing effective, evidence-based preventive measures, and guiding cost-effective studies to address gaps in early OPSCC diagnosis.²

In the United States, data from linked population-based HIV and cancer registries showed a higher incidence of HPV-related OPSCC among people living with HIV (PLWH) compared to the general population (SIR 1.64; 95% CI 1.46, 1.84), with an even greater risk observed in those previously diagnosed with acquired immunodeficiency syndrome (AIDS)-defining conditions (SIR 1.84; 95% CI 1.61, 2.09).⁴ This aligns with global studies showing PLWH have a higher proportion of oral high-risk HPV (HR-HPV) infections compared to negative individuals (15.0 vs 3.7%), and a greater six-month persistence of HR-HPV (41.9 vs 27.3%), a recognized obligate precursor to HPV-related OPSCC.^5,6

Overall, oral HPV infection varies significantly by age, sex, geographic region, recruitment setting, and sampling methods.⁶ Studies evaluating age-related associations and patterns of oral HPV infection have only been conducted in immunocompetent men.^7,8 In addition, while CD4 cell count, a marker of immunosuppression, has consistently been associated with oral HPV infection in PLWH, the associations with other HIV-related factors—such as time since diagnosis, AIDS status, nadir CD4 count, and duration of antiretroviral therapy (ART)—remain inconsistent or underexplored.^6,9–13 The primary objective of this study is to assess oral HPV genotype distribution, and their associations with age and HIV-related factors among men living with HIV in Brazil, Mexico, and Puerto Rico enrolled in the ULACNet-201 trial.¹⁴

Materials and Methods

This cross-sectional study used baseline data from the randomized, double-blinded, placebo-controlled Phase III interventional trial evaluating the efficacy of the nine-valent HPV (9vHPV) vaccine aimed in preventing persistent oral HPV infection among men living with HIV in Brazil, Mexico, and Puerto Rico (ULACNet-201, ClinicalTrials.gov Identifier: NCT04255849).¹⁴ The study was conducted in accordance with the strengthening of the reporting of observational studies in epidemiology (STROBE) guidelines. Ethical approval was obtained by the human subject committees of the University of Puerto Rico, School of Medicine in Puerto Rico, the University of São Paulo in Brazil, and the Instituto Nacional de Salud Publica in Mexico. Written informed consent was obtained from all participants.

Study Population

Participants were recruited from ULACNet-201 trial sites from February 2, 2021 to February 27, 2024.¹⁴ Recruitment occurred in Puerto Rico at the UPR-Site Community Engagement Division (CED) and three HIV clinics (CLETS, PRCoNCRA, and HOPE); in Mexico at CAPASITS Cuernavaca and CAPASITS Puebla (STI/HIV prevention centers); and in Brazil at HIV clinics affiliated with the HCFMUSP’s Clinical Research Center and the Centro de Referência e Treinamento DST/AIDS-SP.

The study sample included men living with HIV, aged 20 to 50 years, who had been on ART for at least six months and were sexually active within the past six months. Participants were included only if they agree to adhere to the three-dose 9vHPV vaccine schedule and subsequent follow-up. The subjects were excluded if they had a history of OPSCC or other HPV-related cancer, were vaccinated for any HPV type, had warts/lesions in the oral cavity, were allergic to any vaccine component, were taking immune suppressants, had received any blood products or had an AIDS-defining condition within six months prior to study entry. The protocol of the study has been comprehensively described in a recent publication by Giuliano et colleagues.¹⁴

Study protocol

At enrollment, participants completed a computer-assisted baseline questionnaire collecting a subset of the extensive data.¹⁴ This study focused on demographics (age, marital status, education), HPV-related medical history (oral, genital, perianal, anal warts), and HIV-specific clinical details (time since HIV diagnosis, ART initiation, history of AIDS-defining illness, and CD4 nadir). A CD4 cell count of less than 200 was not considered an AIDS-defining illness for the purposes of this study. A physical examination screened for genital and oral lesions. This study evaluated baseline oral gargle (ORG) specimens for HPV testing and venous blood samples for virological and immunological analyses. ORG specimens were collected using a 30-second rinse and gargle procedure with 15 mL of mouthwash, consisting of 15 seconds of vigorous swishing followed by 15 seconds of gargling. Participants were not required to abstain from sexual intercourse in the days before sample collection. Blood samples were collected for CD4 cell count and plasma HIV-1 RNA, which were processed at local clinical laboratories with CLIA certification or equivalent accreditation.

DNA extraction, and HPV testing

Deoxyribonucleic acid (DNA) was extracted from ORG cell pellets for HPV analysis using the BioRobot MDx (QIAGEN, Inc.). Then, HPV genotyping was performed using the PCR-based method RHA kit HPV SPF₁₀ PCR-DEIA-LIPA₂₅ system (DDL Diagnostic Laboratory, Rijswik, the Netherlands).¹⁵ HPV genotypes were classified into two categories: HR-HPV and low-risk HPV (LR-HPV).¹⁶ The following 13 genotypes were categorized as HR-HPV: 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68. The remaining 12 genotypes were categorized as LR-HPV: 6, 11, 34, 40, 42, 43, 44, 53, 54, 66, 70, and 74. HPV genotypes not assessed in the test were classified as unclassified HPV types.

Categorization

Age

For the purposes of stratifying randomization, we categorized age into three groups: 20–30, 31–40, and 41–50 years.

Oral HPV infection

For the analysis of HPV genotype epidemiology, specific rules were applied. The presence of at least one HR-HPV or LR-HPV type was classified as “any HPV type.” Samples with only one HPV type were categorized as “single HPV type”. Samples with more than one HPV type were categorized as “multiple HPV type.” If only LR-HPV types were found, the sample was categorized as “only LR-HPV type”. If in a single sample only HR-HPV types were found, or if both HR-HPV and LR-HPV types were detected, it was categorized as “any HR-HPV type”. Samples with more than one type of HR-HPV were categorized as “multiple HR-HPV type”. The presence of HPV 16 alone was classified as “HPV 16”. The detection of one or more of the following genotypes: HPV 6, 11, 16, or 18, was categorized as “4-valent (4vHPV) vaccine type”, while positivity for one or more of HPV 6, 11, 16, 18, 31, 33, 45, 52, or 58 was categorized as “9-valent (9vHPV) vaccine type”. The detection of one or more of the following genotypes: HPV 16, 18, 31, 33, 45, 52, or 58, was categorized as “HR-HPV within the 9vHPV vaccine type” (HR-HPV - 9vHPV).

HR-HPV genotypes were classified into phylogenetic species groups based on the high-risk clade: alpha-9 (HPV 16, 31, 33, 35, 52, 58), alpha-7 (HPV 18, 45, 39, 59), alpha-6 (HPV 56, 66), and alpha-5 (HPV 51).¹⁶

Statistical Methods

Demographics, medical history, baseline physical exam findings, HIV clinical features, and the detection of oral HPV infection were summarized for all participants and stratified by age group. Categorical variables were described using frequencies and relative frequencies. Associations between these covariates and the age groups and between HPV genotypes, age, and HIV-related characteristics were analysed using ANOVA test for continuous variables and Chi-square tests for categorical variables. When normality assumptions were not met or sample sizes were small, the Kruskal-Wallis test, Fisher’s exact test, and Barnard's exact test were used. Statistical significance was set at p < 0.05, and analyses were performed using SAS version 9.4.

Results

Participant Characteristics

Among 700 men living with HIV enrolled in the ULACNet-201 trial, 433 (61.9%) participants were from Mexico, 235 (33.6%) from Brazil, and 32 (4.6%) from Puerto Rico.¹⁴ Notably, no participants were excluded from this analysis due to missing values. Demographics, HPV- and HIV-related histories, were stratified by the three age groups (Tab. 1). The median age of men on study was 35 years (interquartile range [IQR], 30, 41). Most participants were single and had completed college or postgraduate education.

Table 1.

Demographics, HPV-Related Medical, and HIV Histories of Men Living with HIV, Overall and by Age Groups, ULACNet-201 Trial (2021–2024)

			Age Groups
Characteristics	Total sample (N=700)		20–30 years (N=193)		31–40 years (N=310)		41–50 years (N=197)
	n	%	n	%	n	%	n	%	p value1
Total participants	700	100	193	27.6	310	44.3	197	28.1
Demographics
Residence
Brazil	235	33.6	15↓↓↓	7.8	107	34.5	113↑↑↑	57.4	<0.001
Mexico	433	61.9	175↑↑↑	90.7	193	62.3	65↓↓↓	33.0
Puerto Rico	32	4.6	3↓	1.6	10	3.2	19↑↑↑	9.6
Marital Status (N=699)
Single	481	68.8	153	79.7	210	67.7	118	59.9	<0.001
Married/Cohabiting/Living together	192	27.5	38↓	19.8	93	30.0	61	31.0
Divorced/Separated/Widowed	26	3.7	1↓	0.5	7	2.3	18↑↑↑	9.1
Education (N=698)
Less than High School	65	9.3	25	13.0	25	8.1	15	7.7	0.001
High School/GED	117	16.8	32	16.7	51	16.5	34	17.3
Vocational School	47	6.7	19	9.9	15	4.8	13	6.6
Some College/Graduated College	378	54.2	107	55.7	175	56.5	96	49.0
Postgraduate/Professional School	91	13.0	9↓↓	4.7	44	14.2	38↑	19.4
HPV-related Medical History
History of Oral Warts
Yes	1	0.1	0	0.0	0	0.0	1	0.5	0.557
No	699	99.9	193	100	310	100	196	99.5
History of Genital Warts
Yes	103	14.7	34	17.6	36	11.6	33	16.7	0.115
No	597	85.3	159	82.4	274	88.4	164	83.3
History of Perianal Warts
Yes	73	10.4	18	9.3	41	13.2	14	7.6	0.075
No	627	89.6	175	90.7	269	86.8	183	92.4
History of Anal Warts
Yes	166	23.7	62↑	32.1	72	23.2	32↓	16.2	0.001
No	534	76.3	131	67.9	238	76.8	165	83.8
HIV History
Length of Time Since HIV Diagnosis, years
Median (IQR)	7	(5, 12)	5	(3, 6)	7	(5, 11)	14	(7, 20)	<0.001
History of AIDS-defining Condition2
Yes	129	18.4	34	17.6	56	18.0	39	19.8	0.836
No	571	81.6	159	82.4	254	82.0	158	80.2
Nadir CD4 Cell Count, cells/mm3 (N=417)
Median (IQR)	326	(196, 500)	316.5	(200, 492)	379.5	(200, 552)	277	(153, 490)	0.105
< 200 cells/mm3	106	25.4	31	24.6	44	23.2	31	30.7	0.361
≥ 200 cells/mm3	311	74.6	95	75.4	146	76.8	70	69.3
Time Since First ART, years
Median (IQR)	7	(4, 11)	4	(3, 6)	7	(5, 10)	12	(7, 16)	<0.001

Abbreviations: AIDS, acquired immunodeficiency syndrome; ART, antiretroviral therapy; GED, General Educational Development; HIV, Human Immunodeficiency Virus; HPV, human papillomavirus; IQR, interquartile range; SD, standard deviation.

Note.

¹The p-value is calculated by ANOVA for numeric variables and Chi-squared test for categorical variables. When the p-value was significant for categorical variables, direction and level of significance was noted by using arrow signs. If the cell count was lower than expected value, then ↓ will be used. Otherwise, ↑ will be used. If the cell chi-square greater than a p-value 0.05 critical value (3.841) then one arrow sign will be noted. If the cell chi-square greater than a p-value 0·01 critical value (6.635) then two arrow signs will be noted. If the cell chi-square greater than a p-value 0.001 critical value (10.828) then three arrow signs will be noted. Kruskal-Wallis and Fisher’s exact tests were used to calculate for p-values when needed;

²before six months prior to study entry.

A history of anal (n=166, 23.7%), genital (n=103, 14.7%), and perianal warts (n=73, 10.4%) were much more common than oral warts - only one participant reported oral warts (Table 1). During the clinical evaluation, 7.3% (n=51) of participants presented with perianal/anal warts and 3.4% (n=24) with genital warts (Table 2). Across the three age groups, the proportion of history of anal warts showed a decreasing trend, declining from 32.1% (n=62) in the 20–30-year group to 23.2% (n=72) in the 31–40-year group and 16.2% (n=32) in the 41–50-year group (p=0.001) (Table 1). A similar pattern was observed for the presence of perianal/anal warts at the enrollment visit, which decreased from 10.4% (n=20) to 7.4% (n=23) and 4.1% (n=8) with increasing age, respectively (p=0.057) (Table 2). In contrast, no trends were observed for the history or baseline presence of oral or genital warts across age groups.

Table 2.

Physical Examination and HIV Status of Men Living with HIV at the Enrollment, Overall and by Age Groups, ULACNet-201 Trial (2021–2024)

			Age Groups
Characteristics	Total sample (N= 700)		20–30 years (N=193)		31–40 years (N=310)		41–50 years (N=197)
	n	%	n	%	n	%	n	%	p value1
Overall	700	100	193	27.6	310	44.3	197	28.1
Physical Examination
Genital Warts
Yes	24	3.4	6	3.1	13	4.2	5	2.5	0.583
No	676	96.6	187	96.9	297	95.8	192	97.5
Perianal/Anal Warts
Yes	51	7.3	20	10.4	23	7.4	8	4.1	0.057
No	649	92.7	173	89.6	287	92.6	189	95.9
HIV Exams
CD4 Cell Count, cells/mm3
Median (IQR)	640.5	(456.5, 867)	605	(436, 793)	648	(449, 871)	686	(506, 921)	0.061
Plasma HIV-1 RNA Levels
< 200	685	98.0	190	98.5	303	97.7	192	98.0	0.941
≥ 200	14	2.0	3	1.5	7	2.3	4	2.0

Abbreviations: HIV, Human Immunodeficiency Virus; HPV, human papillomavirus; IQR, interquartile range; RNA, Ribonucleic Acid; SD, standard deviation.

Note.

¹Pearson's Chi-squared test;

²The limit of detection for the HIV RNA assay is 20 copies/mL in Brazil, 40 copies/mL in Mexico, and 40 copies/mL in the United States.

The median time from HIV diagnosis to baseline visit was 7 years (IQR 5, 12), and differed across age groups (p<0.001)(Table 1). Overall, 18.4% (n=129) reported a previous AIDS-defining condition, with no differences by age group. Among participants with available CD4 T-cell nadir data (n=417), the median nadir was 326 cells/mm³ (IQR 196, 500), with 25.4% (n=106) having a nadir below 200 cell/mm³. As per protocol, all participants were receiving ART, with a median initiation time of 7 years before the visit (IQR 4–11). At the baseline visit, the median CD4 T-cell count was 640.5 cells/mm³ (IQR 456.5, 867), showing an increasing trend across the three age groups: 605 cells/mm³ (IQR 436, 793) in the 20–30-year group, 648 cells/mm³ (IQR 449, 871) in the 31–40-year group, and 686 cells/mm³ (IQR 506, 921) in the 41–50-year group (Table 2). However, this trend was not statistically significant (p=0.061). HIV RNA load < 200 copies/mL was reported for 98% of participants, without differences between age groups. Viral load was undetectable in 91.3% participants, without differences between countries.

HPV Genotypes

Overall, any HPV type in oral samples was detected in 27.9% (n=195) participants, with no significant differences among age groups (Table 3). Twenty-two participants (3.1%) had multiple oral HPV type infections. Genotypes covered by the 4vHPV and 9vHPV vaccines were detected in 4.9% (n=34) and 8.9% (62) participants, respectively. Any HR-HPV type and HR-HPV types included in the 9vHPV vaccine were detected in 11.0% (n=77) and 7.7% (n=54) men, respectively. Specifically, 61 genotypes were alpha-9 (HPV 16, 31, 33, 35, 52, 58), 19 were alpha-7 (HPV 18, 45, 68), 13 were alpha-6 (HPV 56, 66), and 5 were alpha-5 (HPV 51) in the high-risk clade.¹⁶ Ten participants (1.4%) had multiple oral HR-HPV type infections. Additionally, 12.0% (n=84) of the samples tested positive for unclassified HPV types.

Table 3.

Oral HPV Infection and Genotypes in Men Living with HIV, Overall and by Age, Trial ULACNet-201, 2021–2024

			Age Groups
Oral HPV Infection and Genotypes	Total sample (N= 700)		20–30 years (N= 193)		31–40 years (N= 310)		41–50 years (N= 197)
	n	%	n	%	n	%	n	%	p value1
Any Detectable HPV (N=700)									0.873
Single HPV type2	89	12.7	21	10.9	41	13.2	27	13.7
Multiple HPV types3	22	3.1	6	3.1	11	3.5	5	2.5
Negative	589	84.1	166	86.0	258	83.2	165	843.8
Any HPV type4	195	27.9	51	26.4	86	27.7	58	29.4	0.800
Any HR-HPV type5	77	11.0	20	10.4	35	11.3	22	11.2	0.945
Single HR-HPV type	67	9.6	16	8.3	32	10.3	19	9.6	0.752
Multiple HR-HPV type6	10	1.4	4	2.1	3	1.0	3	1.5	0.558
Only LR-HPV type7	34	4.4	7	3.6	17	5.5	10	5.1	0.633
4vHPV type8	34	4.9	9	4.7	19	6.1	6	3.0	0.287
9vHPV type9	62	8.9	18	9.3	28	9.0	16	8.1	0.906
HR-HPV type in 9vHPV10	54	7.7	15	7.8	23	7.4	16	8.1	0.959
unclassified type	84	12.0	24	12.4	34	11.0	26	13.2	0.735
High-risk HPV types
HPV 16	17	2.4	3	1.6	9	2.9	5	2.5	0.671
HPV 18	9	1.3	3	1.6	5	1.6	1	0.5	0.583
HPV 31	6	0.9	2	1.0	1	0.3	3	1.5	0.389
HPV 33	14	2.0	5	2.6	4	1.3	5	2.5	0.428
HPV 35	9	1.3	2	1.0	4	1.3	3	1.5	>0.999
HPV 39	0	0	0	0	0	0	0	0
HPV 45	2	0.3	1	0.5	0	0	1	0.5	0.310
HPV 51	5	0.7	2	1.0	2	0.6	1	0.5	0.735
HPV 52	14	2.0	3	1.6	6	1.9	5	2.5	0.835
HPV 56	7	1.0	4	2.1	3	1.0	0	0	0.099
HPV 58	1	0.1	0	0	0	0	1	0.5	0.557
HPV 59	0	0	0	0	0	0	0	0
HPV 68	8	1.1	2	1.0	4	1.3	2	1.0	>0.999
Low-risk HPV types
HPV 6	7	1.0	3	1.6	4	1.3	0	0	0.216
HPV 11	4	0.6	1	0.5	3	1.0	0	0	0.488
HPV 34	1	0.1	0	0	1	0.3	0	0	>0.999
HPV 40	0	0	0	0	0	0	0	0
HPV 42	0	0	0	0	0	0	0	0
HPV 43	0	0	0	0	0	0	0	0
HPV 44	9	1.3	1	0.5	5	1.6	3	1.5	0.688
HPV 53	4	0.6	1	0.5	3	1.0	0	0	0.488
HPV 54	1	0.1	0	0	1	0.3	0	0	>0.999
HPV 66	6	0.9	0	0	2	0.6	4	2.0	0.080
HPV 70	4	0.6	1	0.5	2	0.6	1	0.5	>0.999
HPV 74	15	2.1	3	1.6	6	1.9	6	3.0	0.633

Abbreviations: LR-HPV, low-risk HPV; HR-HPV, high-risk HPV.

Note:

¹Person’s Chi-squared test; Fisher’s exact test;

²only one HPV type;

³more than one HPV type;

⁴at least one HR-HPV or LR-HPV type;

⁵only HR-HPV types or both HR-HPV and LR-HPV types;

⁶more than one HR-HPV type;

⁷only LR-HPV types;

⁸HPV 6 and/or 11 and/or 16 and/or 18;

⁹HPV 6 and/or 11 and/or 16 and/or 18 and/or 31 and/or 33 and/or 45 and/or 52 and/or 58;

¹⁰HPV 16 and/or 18 and/or 31 and/or 33 and/or 45 and/or 52 and/or 58.

P-values are not presented for cells with counts less than 5 due to statistical limitation

All investigated HR-HPV types were identified, except for HPV 39 and HPV 59, which were not detected in this population (Table 3). Overall, the most detected HR-HPV types were HPV 16 (n=17, 2.4%), HPV 33 (n=14, 2.0%), and HPV 52 (n=14, 2.0%). HPV-66 was found only in six participants (0.9%). The distribution of the most detected HR-HPV varied in specific age groups: HPV 16 reached 2.9% (n=9) in participants aged 31–40 years, HPV 33 reached 2.6% (n=5) in participants aged 20–30 years, and HPV 52 reached 2.5% (n=5) in participants aged 41–50 years. While no statistically significant trends were observed for HR-HPV and LR-HPV types across age groups, the proportion of individuals with HPV 52 and 74 showed a slight increase with age.

Participants testing positive for any HPV type in oral samples had a shorter median time from HIV diagnosis (7 years vs 8 years, p=0.042) and a shorter median time since ART initiation (7 years vs 8 years, p=0.014) compared to those tested HPV negative (Table 4). Table 4 summarizes the proportion of any HPV and HR-HPV in oral samples, stratified by participants' HIV-related history and baseline virologic and immunologic characteristics. Data for HPV 16 are not included due to the small sample size. The proportion of any HPV type was statistically significantly higher in individuals with a history of AIDS-defining condition compared to those without (35.7 vs 26.1%, p=0.029). While the proportion of any HPV type did not differ by CD4 nadir levels reported in the past, it was significantly higher among participants with a CD4 cell count measured at the enrollment of less than 200 cells/mm³ compared to those with more than 500 cells/mm³ (59.1% vs 26.8%, p=0.004) at the enrolment. The median CD4 cell count was lower in participants who tested positive for any HPV type compared to those who tested negative (599 cells/mm³ vs 655 cells/mm³, p=0.042). Proportion of any HPV type was not different by HIV RNA load.

Table 4.

Oral HPV Infection by Genotype and HIV-Related Characteristics in Men Living with HIV (N=700), ULACNet-201 Trial, 2021–2024

HIV Characteristics	Any HPV				HR-HPV
	Positive	Negative	P1 value	Any HPV Infection (95% CI)	Positive	Negative	P1 value	HR-HPV Infection (95% CI)
Overall (N=700),	N=195	N=505		27.9% (24.7%, 31.3%)	N=77	N=623		11.0% (8.9%, 13.5%)
Time Since HIV Diagnosis, Median, years (IQR)	7 (5, 12)	8 (6, 14)	0.042	-	7 (5, 12)	8 (6, 14)	0.099	-
History of AIDS-defining Condition2, n (%)
Yes (N=129)	46 (23.6)	83 (16.4)	0.029	35.7% (27.9%, 44.2%)	21 (27.3)	108 (17.3)	0.034	16.3% (10.9%, 23.6%)
No (N=571)	149 (76.4)	422 (83.6)		26.1% (22.7%, 29.8%)	56 (72.7)	515 (82.7)		9.8% (7.6%, 12.5%)
Nadir CD4 Cell Count, n (%) (N=417)
Median, cells/mm3 (IQR)	300 (167, 535)	337.5 (200, 500)	0.366	-	264 (150, 390)	335.5 (200, 507)	0.056	-
< 100 cells/mm3 (N=46)	15 (11.6)	31 (10.8)	0.539	32.6% (20.9%, 47.0%)	9 (18.4)	37 (10.1)	0.110	19.6% (10.1%, 32.2%)
100–250 cells/mm3 (N=100)	35 (27.1)	65 (22.6)		35.0% (26.4%, 44.7%)	14 (28.6)	86 (23.4)		14.0% (8.5%, 22.1%)
>250 cells/mm3 (N=271)	79 (61.2)	192 (66.7)		29.2% (24.1%, 34.8%)	26 (53.1)	245 (66.6)		9.6% (6.6%, 13.7%)
Time Since First ART, Median, years (IQR)	7 (5, 10)	8 (6, 12)	0.014	-	7 (5, 11)	8 (5, 12)	0.135	-
Current CD4 Cell Count, n (%)
Median, cells/mm3 (IQR)	599 (422, 826)	655 (476, 884)	0.042	-	567 (383, 772)	649 (473, 872)	0.033	-
< 200 cells/mm3 (N=22)	13 (6.7)	9 (1.8)	0.004	59.1% (38.7%, 76.7%)	9 (11.7)	13 (2.1)	<0.001	40.9% (23.2%, 61.3%)
200–500 cells/mm3 (N=189)	51 (26.2)	138 (27.3)		27.0% (21.2%, 33.7%)	21 (27.3)	168 (27.0)		11.1% (7.4%, 16.4%)
>500 cells/mm3 (N=489)	131 (67.2)	358 (70.9)		26.8% (23.1%, 30.9%)	47 (61.0)	442 (71.0)		9.6% (7.3%, 12.5%)
Plasma HIV-1 RNA Levels			0.955				>0.999
< 200 copies/mL (N=685)	191 (98.0)	494 (98.0)		27.9% (24.7%, 31.4%)	76 (98.7)	609 (97.9)		11.1% (9.0%, 13.0%)
≥ 200 copies/mL (N=14)	4 (2.1)	10 (2.0)		28.6% (11.7%, 54.6%)	1 (1.3)	13 (2.1)		7.1% (1.3%, 31.5%)

Note.

¹Pearson’s Chi-squared test and Fisher’s exact test;

²before 6 months prior to study entry.

³The limit of detection for the HIV RNA assay is 20 copies/mL in Brazil, 40 copies/mL in Mexico, and 40 copies/mL in Puerto Rico.

Abbreviations: CI, confidence interval; ART, antiretroviral therapy; HPV, human papillomavirus; IQR, interquartile range; SD, standard deviation.

A comparable pattern was observed for HR-HPV. Participants with a history of AIDS-defining conditions had a statistically significant higher proportion of HR-HPV compared to those without (16.3% vs. 9.8%, p=0.034). Additionally, HR-HPV proportion was higher among participants with a CD4 cell count at the enrollment below 200 cells/mm³ (40.9%), compared to those with 200–500 cells/mm³ (11.1%) and more than 500 cells/mm³ (9.6%, p<0.001). Moreover, participants who tested positive for HR-HPV had a significantly lower median CD4 cell count compared to those who tested negative (567 cells/mm³ vs. 649 cells/mm³, p=0.033). No differences were found among participants according to the viral load. HPV 16, with only 17 positive cases, showed no significant associations with any of the evaluated characteristics.

Discussion

Our analysis of oral specimens from men living with HIV highlights a distinct distribution of oral oncogenic HPV genotypes, consistent with limited studies on PLWH.^6,8,17–25 In our cohort, 27.9% had any oral HPV, and 11.0% had HR-HPV, contrasting with a meta-analysis reporting a pooled prevalence of 9.3% for any oral HPV and 3.9% for HR-HPV in the general population, with South America showing the highest proportion (12.4%).²³ Our results differ also from our previous HPV Infection in Men (HIM) study, which found an oral HR-HPV prevalence of 5.9% among HIV negative men, who generally have a stronger immune system capable of clearing the infection more effectively.⁸ Conversely, our findings align with studies of PLWH, estimating a global prevalence of 26% for any oral HPV and 15% for HR-HPV, with notable variation across countries.^6,17–22 In Mexico City, HR-HPV detection was 9.3% in a 2014–2016 cross-sectional study of men and 14.8% in a 2018 observational study involving MSM.^17,18 Brazilian studies reported HR-HPV infection ranging from 3.5% to 7.3%.^19,20 In Argentina, HPV infection detection among men living with HIV was 22.1%, whereas Puerto Rico reported 20.0% for any HPV type and 10.7% for HR-HPV STI clinic attendees.^21,22 However, the Puerto Rico study did not distinguish between men with and without HIV.²¹ In a 2011 cross-sectional study in Peru among MSM living with HIV, 44.3% had any oral HPV and 24.8% had HR-HPV.²⁶ This elevated detection may be attributed to the extremely high proportion of participants reporting oral sex (95.7%) and history of ever smoking (99.7%).²⁶ HPV 16, the most common HR-HPV type in immunocompetent individuals (1.4%–1.6%), continued to be detected at higher levels among PLWH (4.2%) but other HR-HPV type are also seen to predominate.^{6,8,19,21,23,26} For example, in a cohort of PLWH in São Paulo, HPV 51 (2.1%) was the most common HR-HPV type found.¹⁹ in Puerto Rico, HPV 52 (2.9%) was the most frequently detected HR-HPV types among PLWH, surpassing HPV 16 (2.4%).²¹ Similarly, in Peru, HPV 58 (9.6%) emerged as the most prevalent HR-HPV type, followed by HPV 16 (8.4%) and HPV 45 (7.7%).²⁶ In our study, while HPV 16 remained the most frequently detected HR-HPV type (2.4%), other alpha-9 genotypes, such as HPV 33 and 52 (2.0% each) and HPV 35 (1.3%), were also common, highlighting a broader distribution of genotypes among PLWH. Overall, most of HR-HPV detections belonged to the alpha-9 species group (n = 61), followed by alpha-7 (n = 19), alpha-6 (n = 13), and alpha-5 (n = 5). The proportion of 4vHPV (4.9%), 9vHPV (8.9%), and HR-HPV types covered by the 9vHPV vaccine (7.7%) was significantly higher than in immunocompetent men in the HIM study (3.2%, 4.6%, and 3.5%, respectively).⁸

The ULACNet-201 trial results, along with findings from other studies, confirm the disproportionate burden of oral HPV infection, particularly HR-HPV, among men living with HIV in Latin America and the Caribbean.^6,9,11 These findings underscore the urgent need for strategies to monitor long-term cancer risk in men living with HIV, not only for genital cancers, but also for OPSCC and other head and neck malignancies.^2,6,27 Promising results from existing HPV vaccines, including the 9vHPV vaccine which estimated to prevent 21.3% of HPV-related OPSCC cases globally, underscore the critical importance of expanding vaccination efforts.^27,28 This is especially critical given the significantly higher incidence of HPV-related OPSCC among PLWH, 64.0% higher overall and 84.0% higher in those with a history of AIDS-defining conditions, largely due to the elevated prevalence and persistence of HR-HPV types and additional cancer-related risk factors in this population.^4,9 Oral infection with only LR-HPV types, excluding HR-HPV, was detected in 4.4% of participants and may be clinically relevant because these types are associated with oral warts in immunocompromised individuals. None of the participants had oral warts at baseline, as this was an exclusion criterion for the trial. Nevertheless, longitudinal follow-up is needed not only to assess the long-term risk of OPSCC, but also to monitor the possible short-term development of oral warts.

The ULACNet-201 trial findings are consistent with other studies on PLWH, demonstrating a clear association between oral HPV infection and baseline CD4 cell count. In our study, participants with a baseline CD4 cell count below 200 cells/mm³ at the start of the trial had a significantly higher proportion of HR-HPV compared to those with higher levels. HR-HPV proportion decreased clearly from 40.9% to 11.1% and 9.6% as CD4 levels increased, reflecting a fourfold difference between the lowest and highest CD4 categories. Similarly, men reporting an AIDS-defining condition showed a higher percentage of HR-HPV than those without. These findings underscore the pivotal role of immunological factors in the natural history of HPV and the importance of early treatment.^9,11,13 The lack of association between nadir CD4 count and HPV infection emphasizes the greater significance of current immunological status compared to past conditions. Different from most studies, all ULACNet-201 trial participants were receiving ART, which was likely initiated immediately after HIV diagnosis. Participants with any HPV infection had significantly shorter median durations of ART, emphasizing the potential role of prolonged ART in restoring immunological function and ultimately lowering HPV percentages, regardless of CD4 nadir.²⁹ The lack of association between nadir CD4 and HPV infection suggests that ART's impact extends beyond CD4 recovery, highlighting its essential role in both immune recovery and HPV control. This is in line with findings of the seminal START study, which demonstrated that the immediate initiation of ART after a HIV diagnosis significantly reduces the risk of infection-related cancers, including anal and cervical cancers, more effectively than infection-unrelated cancers.²⁹ However, the risk of oral HPV infection is not fully mitigated by the initiation of ART, as reported by Shiboski et al.³⁰ Their study noted no reduction in oncogenic oral HPV infection among ART-naïve HIV patients after 12–24 weeks of ART, despite achieving high rates of HIV virologic suppression and substantial immune reconstitution.³⁰ Moreover, the risk of new infections persists throughout adulthood for sexually active men, as evidenced by studies showing that genital and oral HPV prevalence remains stable in immunocompetent men across age groups.⁸ Our findings align with these analyses. The absence of significant variations in oral HPV infection by age could be attributed to several factors. First, men living with HIV may have a consistently elevated risk of HPV infection across all age groups due to persistent immunosuppression and ongoing sexual exposure, reducing the expected age-related differences observed in the general population.⁷ Additionally, since all participants were on ART, with a high proportion achieving viral suppression, their immune status may have been relatively stable, which may have reduced differences in how easily they could acquire new HPV infections.

A limitation of this study is recall bias, particularly in participant reporting nadir CD4 count. Given that a median of 7 years (up to 12 years for some) had passed since starting treatment, inaccuracies in recall may affect the reliability of nadir CD4 data and its association with HPV infection. This may lead to misclassification, potentially obscuring associations between CD4 nadir and HPV infection. Sample size is another limitation, which constrained our ability to detect significant associations between the genotype, including HPV 16, the most common HR-HPV type, and explanatory factors. Furthermore, the high proportion of HPV types not assigned to a risk category emphasizes the need for further investigation to improve genotype characterization and enhance our understanding of HPV variety.

In conclusion, this study offers essential insights into the epidemiology of oral HPV infection, particularly oncogenic and 9vHPV types, among men living with HIV, providing a foundation for evidence-based strategies to address HPV-related OPSCC.