Methods
The PROUD trial was designed to explore the real-world effectiveness of PrEP in which eligible HIV-negative GMSM received open-label tenofovir-emtricitabine (TDF-FTC) either immediately, in a risk reduction package, or after a deferral period of 12 months follow-up. The intention was of implementing this in sexual health clinics and with study procedures as close as possible to routine care in this setting. This paper reports the characteristics of participants recruited during a pilot phase, which aimed to establish the feasibility of a complete trial. However, the unexpectedly large number of HIV infections during the deferral period led to a recommendation from the Trial Steering Committee in October 2014 that all participants be offered PrEP. The findings on the effectiveness of PrEP were published and a larger trial is no longer required [
14]. Trial procedures during the study are described in detail elsewhere [
14].
The anticipated target for an adequately powered trial was 5000 (2500 per arm). This was based on an estimated incidence of 3/100 person-years during the first year in participants who were waiting to access Truvada and a 50 % reduction in incidence in those offered Truvada. An arbitrary 10 % target of 500 for the pilot phase was a pragmatic choice to guide as to whether 5000 participants could be enrolled over 2 years, according to the eligibility criteria in Table
1.
Table 1
Eligibility requirements
1. Born to male gender, age 18 years or older 2. Previously attended the enrolling clinic on at least one occasion 3. Completed a screen for HIV and STIs 4. HIV-negative by a routinely used assay within 4 weeks prior to or on the day of randomisation 5. Reported unprotected anal intercourse (UAI) on more than one occasion within the 90 days prior to randomisation 6. Likely, in the opinion of the volunteer, to have UAI in the next 90 days 7. Willing and able to comply with the visit schedule throughout the follow-up period 8. Willing and able to provide written informed consent | 1. An acute viral illness that could be due to HIV seroconversion 2. Any contraindications to Truvada according to the current package insert 3. Treatment for hepatitis B infection indicated or ongoing 4. Unlikely, in the opinion of the clinician, to comply with the randomised allocation |
Potentially eligible GMSM were identified during routine attendances at 13 sexual health clinics in England, 8 in London and 5 outside (Birmingham, Brighton, Manchester, Sheffield and York). Participants with regular sexual partners (in the opinion of the potential volunteer) who also met eligibility requirements were encouraged to enrol at the same time and both partners were randomly allocated to the same trial arm, minimising the possibility of PrEP being shared. Posters and electronic screens in participating sexual health clinics, as well as advertisements on social media, helped to promote the study. Business cards and leaflets advertising the study were also handed out by community organisations during outreach activities, including efforts to raise awareness of PrEP amongst GMSM. There was no financial payment for participants joining the study, nor were travel costs or other expenses paid for.
A screening visit was not required as the eligibility data on HIV are collected routinely. The Participant Information Sheet was shared with volunteers prior to enrolment. The research team at the clinic determined eligibility through a structured discussion with the volunteer and written informed consent was collected prior to enrolment. Eligible participants were randomised using a web-based tool incorporated within the database at each clinic. At the enrolment visit participants were asked to self-complete, in private, two paper baseline questionnaires (Additional files
1 and
2) on separate booklets collecting information on: demographics, sexual behaviour and lifestyle in the last 30 and 90 days; perception of HIV risk at the last condomless sex act; risk management strategies; past history of STIs; drug and alcohol use; depression severity captured by the Patient Health Questionnaire-9 (PHQ-9) [
15]; motivation for taking part in the study; and perceptions regarding adherence to taking a daily pill. The questionnaires were derived from other studies in MSM populations and were being tested as part of the pilot phase. Questionnaires took less than 30 minutes to complete and were placed in a sealable envelope and sent to the Medical Research Council Clinical Trials Unit (MRC CTU) at University College London (UCL) for data entry without clinic staff seeing the responses.
A screen for STIs was performed, if indicated, according to routine clinical practice. A sample for antibody/antigen HIV testing was collected on the day of enrolment (Table
1).
In NHS sexual health clinics, basic demographic, diagnostic and service data are also returned for each attendance and collated by Public Health England, using the genitourinary medicine clinic activity database version 2 (GUMCADv2). GUMCADv2 is a pseudo-anonymised patient-level electronic dataset collecting information on diagnoses made and services provided by genitourinary medicine (GUM) clinics (level 3) and other commissioned level 2 (non-GUM) sexual health services [
16]. For this analysis, data were extracted from GUMCADv2 on GMSM who were HIV-negative or of unknown status from all clinics in England between January 2013 and December 2013. Data were extracted on key demographics including age, ethnicity and place of birth, STI diagnoses by number of GMSM, as opposed to by GMSM attendances, (pharyngeal, urethral and rectal chlamydia and gonorrhoea, primary secondary and early latent syphilis, hepatitis B and C infections, lymphogranuloma venereum (LGV)), HIV tests (number of HIV tests and average number per GMSM), and episodes of post-exposure HIV prophylaxis (PEP) among GMSM.
The data were analysed using Stata 13.1 [
17]. Comparisons of categorical data were conducted using a χ
2 test or a two-tailed Fisher exact test where numbers were less than five in any group. Participants recruited within London were compared to those recruited outside of London to determine differences in baseline demographics, sexual behaviour and lifestyle. Continuous variables were compared using a Mann-Whitney
U test. Multiple component analysis (MCA) was used to determine whether responses to questions where participants could select multiple answers decomposed into distinct groups [
18].
The PROUD study protocol was approved by London Bridge Research Ethics Committee, the Medicines and Healthcare products Regulatory Agency and each participating hospital trust (Chelsea and Westminster Healthcare NHS Foundation Trust, London, UK; Brighton and Sussex University Hospitals NHS Trust, Brighton, UK; Homerton University Hospital NHS Foundation Trust, London, UK; Central Manchester University Hospitals NHS Foundation Trust, Manchester, UK; Imperial College Healthcare NHS Foundation Trust, London, UK; Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK; York Teaching Hospital and Hull York Medical School, York, UK; Barts Health NHS Trust, London, UK; King’s College Hospital NHS Foundation Trust, London, UK; Guy’s and St Thomas’ NHS Foundation Trust, London, UK; and Heart of England NHS Foundation Trust, Birmingham, UK).
Discussion
The PROUD study is the first randomised entirely open-label study of PrEP with a deferral design that aims to explore its real-world effectiveness, taking account of changes in behaviour that may follow the use of a drug known to reduce the risk of contracting HIV.
The baseline data presented here indicate that the GMSM who enrolled in the PROUD study are at substantially higher risk of acquiring HIV infection sexually than the overall population of GMSM attending sexual health clinics in England. Our data demonstrate that the PROUD cohort had several key indicators to support the observation that they were at high risk of HIV acquisition, particularly those recruited through the London centres. This is consistent with data suggesting that people living in London tend to have poorer sexual health outcomes [
23,
24], are a population with higher rates of partner change and higher use of ChemSex drugs [
25]. These indicators could inform discussions with potential PrEP users in the UK. Further analysis is required to establish whether these indicators are predictive of HIV infection among the participants who seroconverted during the study.
There is a striking excess of reported STIs in the 12 months preceding baseline in the PROUD cohort compared to the number of infections diagnosed in the general HIV-negative GMSM population attending clinics in 2013. This difference is not directly comparable, due to participants’ 12-month history of STIs not all occurring in 2013, and because diagnosed STIs are per clinic whilst reported STIs may span several clinics that a participant visits. Nonetheless, this difference suggests that PROUD participants have particularly high rates of STIs. Of note, a high proportion of the PROUD cohort reported previous rectal infections. Rectal infections, particularly with gonorrhoea and chlamydia, have been associated with higher risk of HIV infection and it has been suggested that a diagnosis of rectal STIs may be used as one of a number of criteria for actively recommending PrEP to GMSM clinic attendees [
26]. Despite this excess of STIs at baseline, only 22 % of participants considered themselves to be at high or very high risk of HIV in general when they last had anal sex without using a condom. This highlights a limitation of this question as perceived HIV risk at the last anal sex act without a condom will vary by partner and may differ from a participant’s overall perceived risk of acquiring HIV.
Participants also reported high numbers of sexual partners, particularly condomless sex partners. The PROUD cohort reported a higher median number of partners in the preceding 90 days, compared to that reported elsewhere. For example, PROUD participants reported 10 partners versus 5 in the European EMIS survey of MSM [
27] and versus 3 in the US National Behavioural Surveillance system survey [
28]. In the Pre-exposure Prophylaxis Initiative (iPrEx) study, eligible GMSM reported an average of 7 sexual partners in the past 3 months at baseline and an average of 18 sexual partners in the past 12 months [
2]. The median number of condomless sex partners reported in PROUD appears high in relation to European and US surveys. In EMIS, 58 % of GMSM reported at least one instance of condomless anal sex in the past 12 months [
27]. In a US survey, white and black GMSM reported a median of two condomless anal sex partners in the past 12 months [
29]. Further work should compare the distribution of the number of partners since the median is of limited value.
There is concern about the role of recreational drugs in facilitating condomless anal sex, leading to an increased number of partners and the possibility of rectal trauma after protracted periods of sexual activity. A survey of ChemSex (sex under the influence of drugs) in south London found that a third of men reported difficulty in negotiating safe sex whilst under the influence of drugs [
25]. Although PROUD participants were not asked specifically about ChemSex, their use of recreational drugs in the past 3 months, particularly drugs associated with ChemSex such as mephedrone, crystal methamphetamine and GHB/GBL, was much higher than previously reported in national Internet surveys of GMSM [
30] and the general population in England and Wales [
31].
The study population reported an eight-fold higher use of PEP for sexual exposure in the past year when compared to the national surveillance data, with 13 % using it more than once. The costs associated with PEP are substantial and include drug costs and frequent clinic visits. Cost-effectiveness analyses need to be conducted to determine if a PrEP programme could reduce costs compared to PEP in a population with this level of PEP usage.
PROUD participants used several other risk reduction strategies such as condoms, serosorting and seropositioning. However, despite active risk management, including using strategic positioning, three participants had a reactive HIV antigen-antibody test at baseline, despite a non-reactive antibody point-of-care test. These recently acquired infections underscore the failure of existing risk reduction strategies to prevent infection. This is not surprising as the range of risk reduction strategies employed prior to enrolment were selective, either selective condom use or selection of partner and type of sex to have without a condom. Follow-up data from PROUD will provide an opportunity to assess how men fit PrEP into their existing strategies.
Previous cost-effectiveness studies for PrEP suggested that with current drug costs a daily PrEP programme will only be of cost-benefit if provided to the population most at risk of HIV [
32‐
35]. Data from PROUD will inform decisions about commissioning of PrEP in England [
36] and UK cost-effectiveness analyses [
37,
38]; these concluded that daily PrEP, in a medium-sized PrEP rollout of 5000 GMSM, would only become cost-effective if current drug prices are substantially reduced. The baseline data suggest that it is feasible to deliver an effective PrEP programme through the sexual health clinic network. PROUD was a pilot study with very limited resources and recruitment depended not only on clinical providers identifying GMSM who would benefit from PrEP, but also GMSM recognising their need for additional options and informing others in their social and sexual networks.
There are some limitations to this analysis. Although data were available for the majority of participants at baseline, a small proportion of baseline data were missing. Efforts were made to contact all of these participants to collect basic demographic data. However, we did not retrospectively collect behavioural data as this would have been subject to recall bias. Recall bias may also explain the difference in 90-day versus 30-day risk assessment findings, with the suggestion that 90-day risk assessment may underestimate more recent risk.
The PROUD population were highly educated; they were older and more likely to have been born outside the UK compared to the general GMSM population attending GUM clinics in England, although were as equally likely to have been born outside the UK as attendees from the sexual health clinics in which they were recruited. The PROUD population has a similar median age to the age of first diagnosis among GMSM in England, which was 34 years in 2012 [
39]. Due to the time lag between infection and diagnosis, this suggests that GMSM accessing PrEP in PROUD are also older than the age of the general GMSM population who seroconvert.
Black, Asian and minority ethnic (BAME) GMSM populations were similarly represented in the PROUD cohort compared to the population attending the same GUM clinics in England but had higher representation than the general GMSM population in England. Data from the US, UK and Canada suggest that this ethnic group is at highest risk of acquiring HIV and would benefit from PrEP. This needs to be explored further in the UK context where there are fewer disparities in access to health care [
40] but where differences in the use of services remain [
41].
PrEP was only offered in the PROUD study in major urban centres, and may not have been accessible to those living in more rural areas and in some parts of the country where there were no PROUD centres. In addition, the process of informed consent, randomisation to wait a year, and participation in a study, in general, could have deterred many eligible GMSM from enrolling. This could have contributed to the initial delay in recruitment to the study and forthcoming quantitative and qualitative work will describe the acceptability of the study in more detail. There were also unanticipated delays getting all 13 recruitment centres operational. Nonetheless, the rate of recruitment increased over time as information about the trial spread via word of mouth and the eventual population recruited were at high risk of HIV, so benefited greatly from the offer of PrEP.
Participants who enrolled in the PROUD study knew that PrEP had proven biological efficacy and these results demonstrate that the population interested in PrEP are likely to be highly self-selective for those at high risk of HIV acquisition. We expect similar effectiveness in populations with a lower risk of sexually acquiring HIV if there are similar levels of adherence, although our findings suggest that populations at low risk of HIV or with low rates of STIs may be less interested in the offer of PrEP if it becomes accessible as part of the NHS. Nonetheless, there were some participants in PROUD who had never reported an STI during their lifetime and further research is ongoing examining any changes in sexual risk behavior and STI diagnosis in this group.
Acknowledgements
Barts: Vanessa Apea, Drew Clark, Paul Davis, James Hand, Machel Hunt, Rebecca Neale, Jackie O’Connell, Margaret Portman, Liat Sarner, John Saunders, Louise Terry, Angelina Twumasi, Salina Tsui, Dayan Vijeratnam, Ryan Whyte, Andy Williams
Birmingham: Sian Gately, Gerry Gilleran, Jill Lyons, Chris McCormack, Katy Moore, Cathy Stretton, Stephen Taylor, David White
Brighton: Alex Acheampong, Michael Bramley, Marion Campbell, Ruby Chowdhry, Amanda Clarke, Stewart Eastwood, Babs Fennell, Martin Fisher, Wendy Hadley, Kerry Hobbs, Sarah Kirk, Nicky Perry, Charlotte Rawlinson, Celia Richardson, Claire Richardson, Mark Roche, Emma Simpkin, Simon Shaw, Elisa Souto, Julia Williams, Elaney Youssef
Chelsea and Westminster: Simone Antonucci, Tristan Barber, Cindy Eliot, Serge Fedele, Chris Higgs, Kathryn McCormick, Sheena McCormack, Alan McOwan, Alexandra Meijer, Sam Pepper, Jane Rowlands, Gurmit Singh, Alfredo Soler-Carracedo, Sonali Sonecha, Ann Sullivan, David Taylor, Lervina Thomas
Gilead Sciences: Rich Clarke, Jim Rooney
Homerton: Frederick Attakora, Marina Bourke, Richard Castles, Rebecca Clark, Anke De-Masi, Veronica Espa, Rumbidzai Hungwe, Martin Lincoln, Sifiso Mguni, Rhianon Nevin-Dolan, Iain Reeves
King’s: Hannah Alexander, Jake Bayley, Michael Brady, Lucy Campbell, Sophie Candfield, Shema Doshi, Olivia Liddle, Larissa Mulka, Priyanka Saigal, James Stevenson
Manchester: James Boateng, Brynn Chappell, Susanna Currie, Carolyn Davies, Dornubari Lebari, Matthew Phillips, Gabriel Schembri, Lisa Southon, Sarah Thorpe, Anna Vas, Chris Ward, Claire Warren, Stephanie Yau
Mortimer Market: Alejandro Arenas-Pinto, Asma Ashraf, Matthew Bolton, Richard Gilson, Lewis Haddow, Sara McNamara, Ana Milinkovic, June Minton, Dianne Morris, Clare Oakland, Steve O’Farrell, Pierre Pellegrino, Sarah Pett, Nina Vora, Carmel Young, Taris Zarko-Flynn
MRC CTU at UCL: Sarah Banbury, Elizabeth Brodnicki, Christina Chung, David Dolling, David Dunn, Keith Fairbrother, Mitzy Gafos, Sajad Khan, Shabana Khan, Sheena McCormack, Brendan Mauger, Yinka Sowunmi, Susan Spencer, Ellen White, Gemma Wood
Public Health England: Monica Desai, Sarika Desai, Nigel Field, Noel Gill, Kate Hyland, Anthony Nardone, Parnam Seyan
Sheffield: Anthony Bains, Gill Bell, Christine Bowman, Terry Cox, Claire Dewsnap, Charlie Hughes, Hannah Loftus, Naomi Sutton, Debbie Talbot, Vince Tucker
Social Science Team: Gill Bell, Caroline Rae, Michael Rayment, Will Nutland, Sonali Wayal
St Marys: Wilbert Ayap, Ling Jun Chen, Adam Croucher, Sarah Fidler, Kristin Kuldanek, Ken Legg, Agathe Leon, Nicola Mackie, Nadia Naous, Killian Quinn, Severine Rey, Judith Zhou
St Thomas’: Margaret-Anne Bevan, Julie Fox, Nina Francia, Eleanor Hamlyn, Lisa Hurley, Helen Iveson, Isabelle Jendrulek, Tammy Murray, Alice Sharp, Andrew Skingsley, Chi Kai Tam, Al Teague, Caroline Thomas, Juan-Manuel Tiraboschi
York: Christine Brewer, Richard Evans, Jan Gravely, Charles Lacey, Gary Lamont, Fabiola Martin, Georgina Morris, Sarah Russell-Sharpe, John Wightman
The PROUD Study Group
Authors’ contributions
SM led the study. DID, MD, MG, DTD, ONG, AN and SM conceived and designed the study. AM, RG, AC, MF, GS, AS, NM, IR, MP, JS, JF, JB, MB, CB, CL, ST, DW, SA and SM recruited patients and collected data. DID and MD led on writing this paper and DID conducted the analyses. All authors read and approved the final manuscript.