Human papillomavirus (HPV) vaccine effectiveness against anal HPV-16 and HPV-18 infections among young men who have sex with men visiting the sexual health center of Amsterdam (HPV4M): study protocol of an observational study

Project status

Recruitment for participants of group 1 began on 1 February 2023 during the Dutch national HPV catch-up campaign; inclusion of 431 participants was completed in June 2024. In February 2025, recruitment of group 2 and 24-month visits of group 1 commenced. The results are expected late 2026.

Objectives

Here we describe the observational study design to estimate the HPV VE for MSM visiting the SHC of Amsterdam (HPV4M study; Clinical Trial Information System registration number 2022-502224-49-00).

The primary aim of the HPV4M study is to evaluate the direct VE of the bivalent HPV vaccine (Cervarix) against anal HPV-16 and HPV-18 DNA positivity among MSM aged 19–26 years, by comparing the prevalence of anal type-specific HPV infections between unvaccinated MSM and vaccinated MSM, 2 years after vaccination.

Our secondary aims are:

1. To assess the anal DNA prevalence, the seroprevalence of HPV-16/18 and the determinants of HPV-16/18 infections.

2. To assess the impact of age and number of lifetime sex partners on VE.

3. To assess the serological response after vaccination and factors affecting serological response.

4. To assess how HPV-16/18 anal positivity 24 months after start of vaccination is affected by HPV DNA and antibody status at the moment of vaccination.

5. To assess the proportion of participants who complete the vaccination schedule and the incidence of adverse events (AE) after administration of Cervarix.

6. Finally, to collect samples and data allowing for later estimations of VE against penile and oral HPV infections.

Study design

The HPV4M study is an observational cross-sectional study among young MSM, with a longitudinal component for the vaccinated group (see figure 1 for schematic representation and timeline).

Study population

In the Netherlands, the bivalent HPV vaccine is routinely offered to preadolescent girls since 2009 and vaccination was extended to boys in 2022. In 2023, a catch-up campaign was launched to offer the bivalent vaccine to all unvaccinated individuals up to the age of 26 years. The vaccination programme uses a 2-dose schedule in accordance with the Dutch national vaccination programme,31 that is, vaccine is given at month 0 and at month 6. People with HIV or other causes of immunosuppression are given an extra dose at month 1.

The HPV4M study was initiated during the national catch-up campaign, but was conducted independently. During the study period, HPV vaccination was available at the same clinic through that national catch-up campaign. The age restriction in the HPV4M study is based on the minimum and maximum age within the national catch-up programme.

Participants are recruited at the SHC of Amsterdam in two phases. During the first recruitment phase, MSM born between 1 January 1996 and 31 December 2003 (aged 19–26 years at the start of the catch-up campaign) attending the SHC and accepting HPV vaccination were invited to participate (group 1). MSM were defined as individuals who were assigned male sex at birth, had not undergone genital gender-affirming surgery, regardless of their current gender identity, and who reported having had sex with partners with a penis in the past 6 months. See table 1 for remaining eligibility criteria. Of note, vaccination leads to seroconversion and very high antibody concentrations in the large majority of people with HIV also.32–36 Therefore, we did not exclude MSM with HIV from participation. For participants with HIV, CD4 cell counts and HIV viral load data are collected. The expected number of MSM with HIV in this age group is low (<5% of eligible men), so stratified analyses by HIV status will not be feasible. In the analysis, we will conduct a sensitivity analysis excluding people with immunocompromising conditions or the use of immunosuppressive medications or HIV infection.

Group 1 will be followed up until 24 months after their first vaccination. In the calendar period that group 1 men return for their 24-month visits, an unvaccinated age-matched control population will be enrolled (group 2) (see figure 1 for schematic representation). Group 2 participants are recruited from the same birth cohorts as group 1 participants, enabling direct comparison of the prevalence of anal HPV-16 and HPV-18 infections between the vaccinated (group 1) and the unvaccinated (group 2) participants. During the recruitment period for group 1, HPV vaccination was available free of charge through the national catch-up campaign for all individuals up to the age of 26 years. Group 2 is recruited after the campaign was concluded, and free vaccination was no longer available. Group 2 participants were asked about their reasons for not having received the HPV vaccine. Participants were eligible for participation in group 2 of the study if they would be willing to get vaccinated if offered the vaccine at enrolment.

Ideally, the study population is representative of all MSM aged 19–26 years attending the SHC at the Public Health Service of Amsterdam, in terms of sexual orientation (eg, sex with only men, sex with both women and men), ethnicity, education, socioeconomic status, number of sexual partners, HPV infection status, presence of coinfections (eg, HIV) and past STIs. However, enrolled participants may differ from this population, as participation depends on willingness and ability to participate in this study. To address this, we will collect information on the above-mentioned variables for both study groups and compare their distribution with that of the reference population from the SHC of Amsterdam.

Recruitment

All MSM born between 1996 and 2003, attending the SHC of Amsterdam, receive an HPV4M-leaflet and are invited to participate in the study. A SHC nurse explains the study procedures and what participation entails. At this point, the person can either decline, participate immediately after providing an informed consent or provide contact details to be contacted later if they are interested on further consideration. If the person declines, the regular SHC visit continues and the decision is noted, so he will not be approached again. Recruitment also takes place through other channels, that is, by distribution of flyers at the SHC and other places frequented by the target population and via social media. The flyers include a quick response code that allows potential participants to register online through a website of the SHC.

Procedures and interventions

When a potential participant (either group 1 or 2) meets with a research nurse, the study is explained and eligibility criteria are assessed. If he decides to participate, both the participant and research nurse sign two copies of an informed consent form, which includes a Patient Information Form; one copy is kept for our records and the other is given to the participant. Since Group 1 is followed up for 24 months, and participation in Group 2 only consists of a single visit, both procedures are explained separately below.

Group 1

During the first visit (month 0), participants are asked to complete a short online questionnaire on an SHC-provided tablet (further explained in the Questionnaires section). Next, the nurse explains the sampling procedures, and the participant provides anal, penile and oral rinse self-samples. For the anal swab, participants insert the swab 3 cm into the anal canal and turn it around for 5–10 s. For the penile swab, participants rub the swab firmly over the skin of the penile shaft, including the outside of the foreskin, for 20 s (ie, 5 s per side). For the oral rinse samples, participants rinse their mouth with a saline solution for 30 s, collecting cells and viral particles from the oral cavity and oropharynx. Finally, the nurse collects 5 mL venous blood. If needed, the visit is combined with a regular STI visit (eg, PrEP consultation, STI tests or treatment). After sample collection, the first dose of the bivalent HPV vaccine, Cervarix, is administered. The next appointment (at month 6 and, if necessary, at month 1) is scheduled during this appointment.

At month 6, participants receive their last vaccine dose. Participants who missed a scheduled dose are contacted by phone and mail to reschedule the appointment as soon as possible. Samples for testing are collected neither at month 1 (for participants with HIV) nor at month 6. At month 15, participants complete a short online questionnaire about sexual behaviour and AEs. During the 24 month visit, participants are asked to complete the same online questionnaire and provide the same anal, penile, oral and blood samples as during the initial appointment. Participants who have missed their 6-month (or 1-month dose in case of immunocompromised individuals and people with HIV) vaccine dose are offered vaccination during the 24-month study visit. Study participation is completed after this visit. When all study visits are completed, participants interested in their serological response and anal HPV status will receive their results by mail, along with careful interpretation.

Group 2

Group 2 will be recruited in the same calendar period when group 1 participants return for their 24-month visit. Men from group 2 will provide the same (anal, penile, oral gargle, venous blood) samples for HPV testing and complete the same questionnaire as group 1. Participation in group 2 consists of one single study visit.

Adverse events

AEs are monitored through an online questionnaire, sent 2 weeks after administration of each vaccine dose (see the Questionnaires section below). Completed questionnaires are checked on a daily basis to identify any AEs that may be serious (SAEs). In the case of possible SAEs, the participant is contacted by telephone to obtain more details, and in some cases, asked to visit the clinic for further examination. SAEs will be reported to the sponsor (ie, Public Health Service (GGD) of Amsterdam) within 24 hours of obtaining knowledge of the events. The sponsor will report SAEs within 7 days to euclinicaltrials.eu, in accordance with international guidance. Symptoms that are severe will be followed up by the study physician with the assistance of the study nurse, until resolution or stability of the symptoms.

Information on AEs after vaccine administrations at 0 months, 1 month and 6 months (group 1) will be analysed as an additional study outcome. We will provide descriptive statistics of AEs.

Data collection

Sample size calculation

The sample size was calculated under the assumption that HPV vaccination of MSM aged 19–26 years will have similar efficacy against newly acquired HPV-16 and HPV-18 infections as among MSM with limited prior exposure to these types. We estimated the expected population-level VE in an MSM population attending the SHC in Amsterdam using assumed ‘true’ prophylactic vaccine efficacy from an existing trial21 and population-level characteristics of the MSM population at the SHC. Next, we determined the required sample size to demonstrate this expected VE at a significance level of 0.05 with 80% power.

For efficacy against newly acquired infections, we used per-protocol (PP) estimates of the quadrivalent HPV vaccine trial in US men.21 The PP population consisted of subjects who were seronegative on day 1 and PCR-negative throughout the 3-dose vaccine course for relevant HPV types. In PP analyses, a prophylactic efficacy of 84.5% (95% CI 63.1% to 94.6%) against anal detection of DNA of HPV-16 and/or HPV-18 at any time was estimated,21 which we took as efficacy against incident HPV-16 and HPV-18 DNA positivity in anal samples for our study. The sample size calculation was further informed by prevalence of anal HPV-16 and HPV-18 infections and seropositivity data from MSM participants in the PASSYON study. PASSYON is a biennial repeated (since 2009) cross-sectional study among SHC attendees aged 16–24 years, enrolled at nine SHCs throughout the Netherlands, among which was the SHC of Amsterdam (online supplemental table S1).39 40 Note that prevalence and seropositivity were higher in the PASSYON study than in the quadrivalent vaccine trial, reflecting a higher prior exposure to HPV. This is in line with the fact that>80% of the PASSYON participants reported>5 sex partners, whereas the preselected population in the quadrivalent vaccine trial was limited to men with ≤5 lifetime sex partners.

In sample size calculation, we made two assumptions. First, we assumed that anyone who tests positive for a vaccine type (either DNA-positive or seropositive) at baseline is not at risk for acquisition of that vaccine type during follow-up and second, that there are no latent or undetectable (DNA-negative) HPV infections at baseline (see figure 2). By powering our study to a scenario without latency, we aim to detect VE that is relevant for achieving substantial population-level impact in a targeted immunisation campaign. We previously concluded, based on a dynamic modelling study, that the efficiency of targeted vaccination would be severely reduced if latency were common among sexually experienced vaccine recipients.28 We deliberately did not power our study to capture diminished VE in such a scenario.

Following these assumptions, we assume a baseline prevalence of 13.9% for HPV-16 and 11.3% for HPV-18 in both groups and calculated an expected 24-month DNA positivity of 6.2% for HPV-16 and 4.6% for HPV-18 in group 1. By using the χ² test to compare the 24-month prevalence in group 1 against the baseline prevalence in group 2, a sample size of 264 for HPV-16 and 284 for HPV-18 is required for 80% statistical power. This corresponds with an OR of 0.41 for HPV-16 and 0.38 for HPV-18, which leads to an anticipated VE against anal HPV-16 and HPV-18 of 59% and 62%, respectively (see online supplemental table S1, figure S1 and appendix II for formula and calculations). The sample size was determined at 300 vaccinated and 300 unvaccinated MSM, to retain power when estimating VE in analyses with adjustment for potential confounders. In order to have complete data for 300 vaccinated men at month 24, with an expected loss to follow-up (LTFU) of 30% among group 1 participants, the target for inclusion was set at 430 participants in group 1. The 30% expected LTFU over 24 months is based on a similar previous study among MSM visiting the SHC, where 20% LTFU was observed.41

Statistical analyses

Patient and public involvement

Patient and/or public were involved in the design of this research. Representatives were selected from the target community to contribute ideas and provide input on the design and recruitment of participants, ensuring that the perspectives and interests of the target population were considered.

Ethics and dissemination

This study has been reviewed and approved by the Medical Ethics Committee of the Amsterdam University Medical Center and registered accordingly in the European Clinical Trials Information System (2022-502224-49-00).

Data collected in this study are pseudonymised at the point of entry (inclusion visit) and stored on a secured folder with access restricted to the principal investigator and research personnel. After data collection, data from the questionnaires, SHC of Amsterdam and laboratory are merged based on unique study numbers. Data are checked for completeness and accuracy, and all analyses are version-controlled. At the end of the study, anonymised datasets and analysis code will be archived for future reference and may be made available to other researchers on reasonable request and ethics approval. The data will be stored for 25 years and the laboratory samples for 10 years.

The study protocol is made public through this protocol paper. Findings from this study will be published in peer-reviewed journals and presented at both national and international conferences. The study design was presented as a poster at the International Papillomavirus Conference in Edinburgh in November 2024. In terms of knowledge translation, the empirical evidence obtained with this study could provide insights for policymakers in the Netherlands and elsewhere. Therefore, results will be presented to stakeholders in immunisation policy and be shared through media aimed at a broad audience.

Patient consent for publication

Not applicable.