Could the effect of antimicrobials on antimicrobial resistance be saturated at high-antimicrobial consumption? A comparison of the MORDOR and ResistAZM studies.

OBJECTIVES: Antimicrobial resistance poses a considerable threat in high-antimicrobial-consumption populations, such as men who have sex with men (MSM) taking HIV pre-exposure prophylaxis. While the ResistAZM trial found no increase in macrolide resistance genes in MSM with gonorrhea after azithromycin treatment, the MORDOR trial observed an increase in these genes after mass azithromycin distribution. We hypothesized that this could be due to saturation of the resistome. To test this hypothesis, we compared the abundance of macrolide resistance determinants in anorectal samples between the baselines of the two trials. METHODS: Shotgun metagenome reads from the anorectal baseline samples from the ResistAZM (n = 42) and MORDOR (n = 30) trials were analyzed using AMRPlusPlus. Nonhost reads were mapped to the MEGARes database to detect antibiotic resistance genes (ARG). Antimicrobial resistance (AMR) was normalized using cumulative sum scaling, and ARG abundance was estimated. RESULTS: Macrolide, lincosamides, and streptogramins determinants were approximately 10-fold more abundant in the ResistAZM than the MORDOR samples (P ≤ 0.001). CONCLUSION: The findings are compatible with our hypothesis. Thus, in populations with high-antimicrobial use, the relationship between antimicrobial consumption and AMR may be diminished due to saturation. These findings are vital for future studies investigating the resistogencity of novel interventions, such as doxycycline post-exposure prophylaxis, in populations with high preceding consumption of antimicrobials.

Oral streptococcal susceptibility to azithromycin may be associated with doxycycline use: A post-hoc analysis with implications for doxycycline post exposure prophylaxis.

OBJECTIVE: Tetracycline and macrolide resistance are frequently linked in streptococci and other species. We aimed to assess the association between doxycycline use and azithromycin MICs in oral streptococci. METHODS: Linear regression was used to assess the association between doxycycline use in the prior year and the median MIC per participant of oral streptococcal colonies isolated at the baseline visit of the ResistAZM study. The analysis controlled for receipt of other antimicrobials as well as time since antimicrobial consumption. RESULTS: Fifty-six individual colonies confirmed to be streptococci were isolated from 19 individuals at baseline. The azithromycin MICs of these isolates varied considerably between 0.25 mg/L and >256 mg/L (median 28 mg/L; IQR 1-192 mg/L). The consumption of doxycycline in the preceding 12 months was positively associated with median streptococcal azithromycin MIC (coef. 151.6 [95% CI 10.6-292.7]; p = .037). CONCLUSION: This post-hoc analysis found that doxycycline use was associated with streptococcal azithromycin susceptibility. Numerous limitations of the study design mean that this study is best considered hypothesis generating. Prospective studies are required to assess if the use of doxycycline could select for macrolide resistance in oral streptococci.

45 years of tetracycline post exposure prophylaxis for STIs and the risk of tetracycline resistance: a systematic review and meta-analysis.

There is considerable interest in the use of doxycycline post exposure prophylaxis (PEP) to reduce the incidence of bacterial sexually transmitted infections (STIs). An important concern is that this could select for tetracycline resistance in these STIs and other species. We searched PubMed and Google Scholar, (1948-2023) for randomized controlled trials comparing tetracycline PEP with non-tetracycline controls. The primary outcome was antimicrobial resistance (AMR) to tetracyclines in all bacterial species with available data. Our search yielded 140 studies, of which three met the inclusion criteria. Tetracycline PEP was associated with an increasedprevalence of tetracycline resistance in Neisseria gonorrhoeae, but this effect was not statistically significant (Pooled OR 2.3, 95% CI 0.9-3.4). PEP had a marked effect on the N. gonorrhoeae tetracycline MIC distribution in the one study where this was assessed. Prophylactic efficacy was 100% at low MICs and 0% at high MICs. In the one study where this was assessed, PEP resulted in a significant increase in tetracycline resistance in commensal Neisseria species compared to the control group (OR 2.9, 95% CI 1.5-5.5) but no significant effect on the prevalence of tetracycline resistance in Staphylococcus aureus. The available evidence suggests that PEP with tetracyclines could be associated with selecting tetracycline resistance in N. gonorrhoeae and commensal Neisseria species.

Four recent insights suggest the need for more refined methods to assess the resistogenicity of doxycycline post exposure prophylaxis.

Two recently published randomized trials of doxycycline post exposure prophylaxis (PEP) have concluded that this intervention is highly effective at reducing the incidence of bacterial sexually transmitted infections (STIs) and has little or no risk of promoting the spread of antimicrobial resistance (AMR). In this perspective piece, we review four types of evidence that suggest that the risk of promoting AMR has been inadequately assessed in these studies. 1) The studies have all used proportion resistant as the outcome measure. This is a less sensitive measure of resistogenicity than MIC distribution. 2) These RCTs have not considered population-level pathways of AMR selection. 3) In populations with very high antimicrobial consumption such as PrEP cohorts, the relationship between antimicrobial consumption and resistance may be saturated. 4) Genetic linkage of AMR means that increased tetracycline use may select for AMR to not only tetracyclines but also other antimicrobials in STIs and other bacterial species. We recommend novel study designs to more adequately assess the AMR-inducing risk of doxycycline PEP.

Self-sampling with oral rinse to detect oropharyngeal Neisseria gonorrhoeae among men who have sex with men: results from an exploratory study in Belgium (the SSONG Study).

OBJECTIVES: We aimed to assess whether a self-collected oral rinse was non-inferior to clinician-collected oropharyngeal swabs to detect Neisseria gonorrhoeae (Ng) using culture and nucleic acid amplification tests (NAAT) among men who have sex with men (MSM), and whether Ng may still be detected in oral rinses for a minimum of 5 days after collection. METHODS: MSM with a positive Ng result in an oropharyngeal or pooled sample (oropharynx, urethra and anorectum) were approached. Clinician-collected oropharyngeal swabs and oral rinses (15 mL sterile water) were taken. Ng culture and NAAT (Abbott 2000m RealTime System CT/NG assay and in-house PCR) were performed. Diagnostic accuracy was assessed using sensitivity and specificity, and agreement between both techniques using Cohen's kappa statistic. Aliquots of positive oral rinses were left at room temperature for a minimum of 5 days and reanalysed using NAAT. Lastly, participants filled in a questionnaire to explore perceptions of both methods. RESULTS: We included 100 participants between June 2022 and October 2023. 45 individuals (45 of 100) had a positive Ng result in either the oral rinses (42 of 45, 93%) or the swabs (36 of 45, 80%). Sensitivity was higher for oral rinses than swabs (sensitivity=0.93/0.80, specificity=1.0/1.0, respectively) and agreement between both techniques was good (kappa=0.75, p<0.001). Of the 42 positive oral rinses, 37 remained positive after a minimum of 5 days (88.1%). Using culture, 18 individuals had a positive Ng result in either the oral rinses (8 of 18, 44%) or the swabs (16 of 18, 88%). Most participants found the oral rinse easy or very easy to use and would be willing to use the oral rinse for home-based sampling. CONCLUSION: We detected more oropharyngeal Ng infections via NAAT using oral rinses than swab samples. However, swabs were better than oral rinses for culturing Ng. Oral rinses might allow for home-based self-sampling to detect oropharyngeal Ng.

Effect of mass treatment on the long-term prevalence of gonorrhoea, chlamydia and syphilis-a systematic review.

BACKGROUND: Selective mass treatment of STIs may lead to a durable reduction in the prevalence of STIs or a temporary reduction associated with an increased probability of antimicrobial resistance emerging. METHODS: We searched PubMed and Google Scholar for studies evaluating the impact of mass STI treatment on the long-term prevalence of chlamydia, gonorrhoea, syphilis and chancroid. The primary outcomes were the long term (≥3 months post the intervention) impact of the intervention on prevalence/incidence of the STI and on antimicrobial resistance. RESULTS: Our search yielded 269 studies, of which 4 met the inclusion criteria. With the exception of the Carletonville study, where this was not assessed, three of the four studies found that intensive STI treatment was associated with a reduced prevalence of the targeted STI during or immediately after the intervention. In all four studies, there was no evidence that the intense treatment had a long-term effect on prevalence. In the only study where this was assessed, the intensive use of penicillin to reduce gonococcal prevalence was associated with the emergence of reduced susceptibility to penicillin in N. gonorrhoeae. CONCLUSION: The available evidence suggests that mass treatment of chlamydia, gonorrhoea and syphilis in high prevalence populations is only associated with a temporary reduction in the prevalence of these infections and may select for antimicrobial resistance.

Lack of Association between Antimicrobial Consumption and Antimicrobial Resistance in a HIV Preexposure Prophylaxis Population: A Cross-Sectional Study.

BACKGROUND: In antibiotic naïve populations, there is a strong association between the use of an antimicrobial and resistance to this antimicrobial. Less evidence is available as to whether this relationship is weakened in populations highly exposed to antimicrobials. Individuals taking HIV preexposure prophylaxis (PrEP) have a high intake of antimicrobials. We previously found that there was no difference in the prevalence of pheno- and genotypic antimicrobial resistance between two groups of PrEP clients who had, and had not, taken antimicrobials in the prior 6 months. Both groups did, however, have a higher prevalence of resistance than a sample of the general population. METHODS: In the current study, we used zero-inflated negative binomial regression models to evaluate if there was an individual level association between the consumption of antimicrobials and 1. the minimum inhibitory susceptibilities of oral Neisseria subflava and 2. the abundance of antimicrobial resistance genes in the oropharynges of these individuals. RESULTS: We found no evidence of an association between the consumption of antimicrobials and the minimum inhibitory susceptibilities of oral Neisseria subflava or the abundance of antimicrobial resistance genes in these individuals. CONCLUSIONS: We conclude that in high-antimicrobial-consumption populations, the association between antimicrobial consumption and resistance may be attenuated. This conclusion would not apply to lower-consumption populations.

Effect of screening for Neisseria gonorrhoeae and Chlamydia trachomatis on incidence of these infections in men who have sex with men and transgender women taking HIV pre-exposure prophylaxis (the Gonoscreen study): results from a randomised, multicentre, controlled trial.

BACKGROUND: Guidelines recommend screening for Neisseria gonorrhoeae and Chlamydia trachomatis at three anatomical sites (urethra, anus, and pharynx) every 3 months (3 × 3) in men who have sex with men (MSM) and transgender women taking HIV pre-exposure prophylaxis (PrEP). We present the first randomised controlled trial to compare the effect of screening versus non-screening for N gonorrhoeae and C trachomatis on the incidence of these infections in MSM and transgender women taking PrEP. METHODS: A multicentre, randomised, controlled trial of 3 × 3 screening for N gonorrhoeae and C trachomatis versus non-screening was done among MSM and transgender women taking PrEP in five HIV reference centers in Belgium. Participants attended the PrEP clinics quarterly for 12 months. N gonorrhoeae and C trachomatis was tested at each visit in both arms, but results were not provided to the non-screening arm, if asymptomatic. The primary outcome was incidence rate of N gonorrhoeae and C trachomatis infections in each arm, assessed in the per-protocol population. Non-inferiority of the non-screening arm was proven if the upper limit of the 95% CI of the incidence rate ratio (IRR) was lower than 1·25. This trial is registered with ClinicalTrials.gov, NCT04269434, and is completed. FINDINGS: Between Sept 21, 2020, and June 4, 2021, 506 participants were randomly assigned to the 3 × 3 screening arm and 508 to the non-screening arm. The overall incidence rate of N gonorrhoeae and C trachomatis was 0·155 cases per 100 person-days (95% CI 0·128-0·186) in the 3 × 3 screening arm and 0·205 (95% CI 0·171-0·246) in the non-screening arm. The incidence rate was significantly higher in the non-screening arm (IRR 1·318, 95% CI 1·068-1·627). Participants in the non-screening arm had a higher incidence of C trachomatis infections and symptomatic C trachomatis infections. There were no significant differences in N gonorrhoeae infections. Participants in the non-screening arm consumed significantly fewer antimicrobial drugs. No serious adverse events were reported. INTERPRETATION: We failed to show that non-screening for N gonorrhoeae and C trachomatis is non-inferior to 3 × 3 screening in MSM and transgender women taking PrEP in Belgium. However, screening was associated with higher antibiotic consumption and had no effect on the incidence of N gonorrhoeae. Further research is needed to assess the benefits and harms of N gonorrhoeae and C trachomatis screening in this population. FUNDING: Belgian Health Care Knowledge Centre.

HIV Pre-exposure Prophylaxis (PrEP) Care in Belgium: A mixed-methods Study on PrEP Users' Experiences and Service Delivery Preferences.

In Belgium, HIV pre-exposure prophylaxis (PrEP) services are mainly provided through specialised HIV clinics. To optimise PrEP uptake and retention in care, we require insights into users' perspectives on PrEP care. We aimed to elicit experiences with, and preferences for, PrEP service delivery among PrEP users in Belgium, including willingness to involve their family physician (FP) in PrEP care. We adopted a sequential mixed-methods design. We used a web-based longitudinal study among 326 PrEP users that consisted of two questionnaires at six-month intervals, and complemented this with 21 semi-structured interviews (September 2020-January 2022). We conducted descriptive analyses and logistic regression to examine factors associated with willingness to involve their FP in PrEP care. Interviews were analysed using thematic analysis. Survey respondents reported high satisfaction with care received in HIV clinics [median score 9 (IQR 8-10), 10='very satisfied']. Interviews revealed the importance of regular HIV/STI screening, and the expertise and stigma-free environment of HIV clinics. Yet, they also contextualised service delivery barriers reported in the questionnaire, including the burden of cost and challenges integrating PrEP visits into their private and professional lives. Although 63.8% (n = 208/326) of baseline respondents preferred attending an HIV clinic for PrEP follow-up, 51.9% (n = 108/208) of participants in the follow-up questionnaire reported to be willing to have their FP involved in PrEP care. Participants reporting trust in FPs' PrEP and sexual health expertise, or who didn't feel judged by their FP, were more likely to be willing to involve them in PrEP care. Therefore, we recommend a differentiated PrEP service delivery approach, including involving FPs, to make PrEP care more client-centred.

Doxycycline PEP can induce doxycycline resistance in Klebsiella pneumoniae in a Galleria mellonella model of PEP.

Background: Four randomized controlled trials have now established that doxycycline post exposure (sex) prophylaxis (PEP) can reduce the incidence of chlamydia and syphilis in men who have sex with men. These studies have concluded that the risk of selecting for antimicrobial resistance is low. We evaluated this risk in vitro and in vivo using a Galleria mellonella infection model. Methods: We evaluated how long it took for doxycycline resistance to emerge during passage on doxycycline containing agar plates in 4 species - Escherichia coli, Klebsiella pneumoniae, Neisseria gonorrhoeae and Neisseria subflava. We then assessed if K. pneumoniae could acquire resistance to doxycycline (and cross resistance to other antimicrobials) during intermittent exposure to doxycycline in a Galleria mellonella model of doxycycline PEP. Results: In our passage experiments, we found that resistance first emerged in K. pneumoniae. By day 7 the K. pneumoniae MIC had increased from 2 mg/L to a median of 96 mg/L (IQR 64-96). Under various simulations of doxycycline PEP in the G. mellonella model, the doxycycline MIC of K. pneumoniae increased from 2 mg/L to 48 mg/L (IQR 48-84). Ceftriaxone and ciprofloxacin MICs increased over ten-fold. Whole genome sequencing revealed acquired mutations in ramR which regulates the expression of the AcrAB-TolC efflux pump. Conclusion: Doxycycline PEP can select for doxycycline, ceftriaxone and ciprofloxacin resistance in K. pneumoniae in a G. mellonella model. The emergent ramR mutations were similar to those seen in circulating strains of K. pneumoniae. These findings suggest that we need to assess the effect of doxycycline PEP on resistance induction on a broader range of bacterial species than has hitherto been the case.

Characteristics of confirmed mpox cases among clinical suspects: A prospective single-centre study in Belgium during the 2022 outbreak.

Background: The presentation of mpox clade IIb during the 2022 outbreak overlaps with a range of other diseases. Understanding the factors associated with mpox is important for clinical decision making. Methods: We described the characteristics of mpox patients who sought care at Belgian sexual health clinic. Furthermore we compared their characteristics to those of patients with a clinical suspicion of mpox but who tested negative on polymerase chain reaction. Results: Between May 23 and September 20, 2022, 155 patients were diagnosed with mpox, and 51 patients with suspected symptoms tested negative. All mpox patients self-identified as men and 148/155 (95.5%) as gay or bisexual MSM. Systemic symptoms were present in 116/155 (74.8%) patients. All but 10 patients (145/155, 93.5%) presented with skin lesions. Other manifestations were lymphadenopathy (72/155, 46.5%), proctitis (50/155, 32.3%), urethritis (12/155, 7.7%), tonsillitis (2/155, 1.3%). Complications involved bacterial skin infection (13/155, 8.4%) and penile oedema with or without paraphimosis (4/155, 2.6%). In multivariable logistic regression models, the presence of lymphadenopathy (OR 3.79 95% CI 1.44-11.49), skin lesions (OR 4.35 95% CI 1.15-17.57) and proctitis (OR 9.41 95% CI 2.72-47.07) were associated with the diagnosis of mpox. There were no associations with age, HIV status, childhood smallpox vaccination, number of sexual partners and international travel. Conclusions: The presence of proctitis, lymphadenopathies and skin lesions should increase clinical suspicion of mpox in patients with compatible symptoms.

Doxycycline Postexposure Prophylaxis Could Induce Cross-Resistance to Other Classes of Antimicrobials in Neisseria gonorrhoeae : An In Silico Analysis.

ABSTRACT: We found that tetracycline resistance-associated mutations and genes in Neisseria gonorrhoeae are linked to mutations causing resistance to other antimicrobials. Therefore, the use of doxycycline postexposure prophylaxis may select for resistance to other antimicrobials.

Alternative sampling specimens for the molecular detection of mpox (formerly monkeypox) virus.

BACKGROUND: Mpox (formerly monkeypox) is a viral disease caused by the mpox virus (MPXV), endemic in Central and West Africa and currently causing a global outbreak of international concern. Much remains unknown about sample types most suited for mpox laboratory diagnosis. While it is established that high viral loads can be found in active skin lesions (currently the recommended mpox laboratory confirmation specimen type), WHO mpox testing guidelines encourage the use of oropharyngeal swabs as an additional sample type for mpox diagnosis and suggest investigating the value of other specimens like blood samples. OBJECTIVE: In this study, we verified the value of select alternative specimen types for mpox laboratory confirmation. METHODS: We included 25 patients with MPXV-confirmed skin lesions to compare diagnostic sensitivity of MPXV PCR testing on EDTA plasma and two upper respiratory specimens: oropharyngeal swabs and saliva. RESULTS: In our patient cohort with MPXV-confirmed skin lesions, diagnostic sensitivity of MPXV PCR was 80% in EDTA plasma, 64% in oropharyngeal swabs, and 88% in saliva. MPXV viral loads were significantly higher in saliva compared to oropharyngeal swabs and EDTA plasma. DISCUSSION: The WHO recommendation to collect oropharyngeal swabs as an additional specimen for mpox diagnosis might need to be revised to include saliva wherever feasible. We suggest investigating saliva as a diagnostic specimen in the absence of active skin lesions or during the phase preceding skin manifestations. Moreover, the relatively high MPXV DNA content of saliva warrants elucidating its potential role in disease transmission.

Understanding sexual transmission dynamics and transmission contexts of monkeypox virus: a mixed-methods study of the early outbreak in Belgium (May-June 2022).

OBJECTIVE: The available epidemiological and clinical evidence from the currently ongoing monkeypox (MPX) outbreak in non-endemic areas suggests an important factor of sexual transmission. However, limited information on the behaviour and experiences of individuals with an MPX infection has to date been provided. We aimed to describe the initial phase of the MPX outbreak in Belgium, and to provide a more in-depth description of sexual behaviour and transmission contexts. METHODS: We used routine national surveillance data of 139 confirmed MPX cases with date of symptom onset until 19 June 2022, complemented with 12 semistructured interviews conducted with a subsample of these cases. RESULTS: Sexualised environments, including large festivals and cruising venues for gay men, were the suspected exposure setting for the majority of the cases in the early outbreak phase. In-depth narratives of sexual behaviour support the hypothesis of MPX transmission through close physical contact during sex. Despite awareness of the ongoing MPX outbreak, low self-perceived risk of MPX acquisition and confusing initial signs and symptoms for other STIs or skin conditions delayed early detection of an MPX infection. In addition, we describe relevant contextual factors beyond individual behaviour, related to sexual networks, interpersonal interactions and health systems. Some of these factors may complicate early MPX detection and control efforts. CONCLUSION: Our results highlight the role of sexual contact and networks in the transmission of MPX during the early phase of the outbreak in Belgium. Risk communication messages should consistently and transparently state the predominant sexual transmission potential of MPX virus, and prevention and control measures must be adapted to reflect multilevel factors contributing to MPX transmission risk.

Putting 2-1-1 into Practice: PrEP Users' Knowledge of Effectively Starting and Stopping Oral PrEP Use.

Starting and stopping oral HIV pre-exposure prophylaxis (PrEP) in a way that compromises its effectiveness should be avoided. Between September 2020 and June 2021, we assessed self-perceived and actual knowledge of effectively starting and stopping oral PrEP through an online survey among 206 PrEP users assigned male at birth in Belgium. We examined associations between incorrect start-and-stop knowledge and socio-demographics, sexual behaviour and PrEP use, using bi- and multi-variable logistic regression. The majority of men (84.9%) perceived their start-and-stop knowledge as 'very good', but only 62.1% of all men correctly indicated how to effectively start and stop with PrEP. Using PrEP daily [adjusted OR 2.12, 95% CI (1.06-4.28), p = 0.034] was significantly associated with incorrect start-and-stop knowledge. To enable PrEP users to effectively use PrEP, they need to be better informed about how to start and stop use, irrespective of the dosing regimen.

Pre-Exposure Prophylaxis Users' Attitudes About Sexually Transmitted Infections and Its Influence on Condom Use: A Mixed-Method Study in Belgium.

Incidence rates of sexually transmitted infections (STIs) are rising among men who have sex with men (MSM). Since the rollout of HIV pre-exposure prophylaxis (PrEP), promoting condom use to prevent the spread of STIs has become more challenging. Using a mixed-method design, we explored MSM PrEP users' attitudes toward STIs, condoms, and condom use with nonsteady partners to prevent STIs. We triangulated data from 22 in-depth interviews conducted at a large HIV/STI clinic between August 2021 and January 2022 and an online survey among 326 PrEP users between September 2020 and January 2022. Interviews were analyzed iteratively using a thematic analysis approach. We used bivariate and multi-variate ordered logistic regression to analyze the online survey data. Themes identified in the qualitative data influencing condom use decisions to prevent STIs were as follows: (1) awareness (i.e., perceived severity of and susceptibility to STIs, condom counseling), (2) motivation (i.e., concerns about STIs, sexual pleasure and protection of own health), and (3) perceived social norms and practices (e.g., reduced condom use at community level). Overall, 10.7% of survey respondents consistently used condoms with nonsteady partners. Survey respondents who reported high or moderate levels of willingness to use condoms to prevent acquiring STIs were significantly more likely to use condoms for anal sex with nonsteady partners; those who initiated PrEP 6-12 months ago were less likely to use condoms. We found a wide variation in attitudes toward condom use for the prevention of STIs among MSM using PrEP. We recommend client-centered approaches, taking into account PrEP users' values and priorities toward STI prevention to help reduce the spread of STIs.

PrEP user profiles, dynamics of PrEP use and follow-up: a cohort analysis at a Belgian HIV centre (2017-2020).

INTRODUCTION: The number of individuals initiating antiretroviral pre-exposure prophylaxis (PrEP) is increasing, but we do not fully understand who is coming forward for PrEP, how they use it and how they are followed-up. The objective of this study was to examine PrEP user profiles, dynamics in PrEP use and follow-up over time. METHODS: We conducted a cohort analysis of longitudinally collected clinical record and questionnaire data among PrEP users at an HIV centre in Antwerp, Belgium, between June 2017 and March 2020. PrEP follow-up and user profiles were examined using descriptive analyses and bivariate logistic regression. We compared early adopting PrEP users (started before June 2018) with late users. We also calculated the probabilities of switching between daily and on-demand PrEP, and interruption, using a naïve estimator. RESULTS AND DISCUSSION: We included 1347 PrEP users in the analysis. After 12 months, retention in care was 72.3%. Median time between PrEP visits was 98 days (IQR 85-119 days). At screening visit, early adopting PrEP users (starting June 2017-May 2018) were significantly more likely to report one or more sexually transmitted infection in the prior 12 months, having used drugs during sex, a higher number of sexual partners and a history of paid sex and PrEP use prior to initiation, compared with PrEP users who initiated later (starting June 2018-February 2020). When taking PrEP daily, the probability of staying on daily PrEP at the next visit was 76%, while this was 73% when taking PrEP on-demand. Those using on-demand PrEP had a higher probability (13%) of interrupting PrEP care than daily PrEP users (7%), whereas those returning to PrEP care would mostly re-start with on-demand (35% vs. 13% for daily). CONCLUSIONS: The majority of PrEP users in this sample remained in care after 12 months. The probability of remaining on the same PrEP regimen at the subsequent visit was high. Though, we observed a diversity of transitions between regimens and interruptions in between visits. Our findings reaffirm the need to provide tailored PrEP services, counselling PrEP users across their life course.