BACKGROUND:The aim of this study was to evaluate changes in prevalence of resistance to integrase strand transfer inhibitors (INSTIs) in relation to changes in their use over 16 years in people living with HIV (PLWHIV) treated with antiretrovirals (ARVs) and in virological failure. MATERIALS AND METHODS:We analysed the use of different INSTIs (for ≥3 months during the study year) in ARV-treated PLWHIV in an academic centre and the HIV INSTI resistance profiles in RNA in case of virological failure (2008-24; analysed using the ANRS-MIE algorithm v33). RESULTS:During the studied period (2008-24), INSTI use increased from 3.3% to 71.2%. Raltegravir use peaked at 21.5% in 2013 and then declined to 2.5% in 2024. Elvitegravir followed a similar trend, increasing to 15.7% in 2018 and then decreasing to 1.5% in 2024. Dolutegravir increased steadily from 6.3% in 2014 to 39.3% in 2023 and bictegravir from 2.0% in 2018 to 28.0% in 2024. Cabotegravir, first noted in 2020 (0.3%), will reach 3.8% in 2024. The proportion of patients with INSTI-resistant viruses was initially high for raltegravir (47.8% in 2008), before decreasing. By 2024, resistance was 2.7% for dolutegravir, cabotegravir and bictegravir, but remained higher for raltegravir (8.5%) and elvitegravir (9.8%). CONCLUSIONS:Although the use of INSTIs has increased over time with differences in the duration of use, we did not observe a corresponding increase in the proportion of ARV-treated PLWHIV who experienced virological failure with INSTI-resistant viruses during the study period.

BACKGROUND:Recent studies have shown progress in understanding the evolution of HIV antiretroviral resistance-associated mutations (RAMs) in the reservoir, particularly M184V, but data on non-nucleoside reverse transcriptase (RT) inhibitors RAMs (NNRTI-RAMs) remain limited. This study aimed to describe the evolution of NNRTI-RAMs in the reservoir, with or without M184V. MATERIALS AND METHODS:This single-centre retrospective study included people living with HIV-1 (PLWHIV) who had one or more NNRTI-RAMs detected in plasma genotype, and who had blood samples collected after at least 1 year of virological suppression on antiretroviral therapy. RT NGS was performed at two time points during virological suppression: DNA1 (2019) and DNA2 (2024). RESULTS:Of 49 PLWHIV with NNRTI-RAMs, 40 had an M184V mutation in their previous HIV-1 RNA genotypes. At DNA1, NNRTI-RAMs and M184V were present in 44.9% (n = 22/49) and 65.0% (n = 26/40) of PLWHIV, respectively. In univariate analysis, NNRTI-RAM clearance at DNA1 was associated with shorter duration of replication under an NNRTI regimen at virological failure (VF) (5 versus 37 months, P = 0.013), longer duration between RNA and DNA1 genotypes (14 versus 9 years, P = 0.010) and older age (60 versus 55 years, P = 0.037). Only replication duration remained associated with the persistence of NNRTI-RAMs (P = 0.041) in multivariate analysis. Past M184V in RNA genotypes was not associated with NNRTI-RAM persistence at DNA1 or DNA2 (P = 0.477 and P = 0.711, respectively). No significant evolution of NNRTI-RAMs was observed between DNA1 and DNA2 (P > 0.999). CONCLUSIONS:In virologically suppressed PLWHIV, NNRTI-RAMs and M184V in the HIV blood reservoir decreased without evidence of interaction between their evolution.

BACKGROUND:We aimed to determine how non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance profiles have changed over the last decade in people living with HIV (PLWHIV) experiencing virological failure on all antiretroviral treatments, including different NNRTIs. MATERIALS AND METHODS:We analysed the use of the different NNRTIs in PLWHIV treated with antiretroviral drugs at an academic centre and the HIV NNRTI resistance profiles observed in cases of virological failure over the last 10 years (2014-23). We used the latest ANRS-MIE algorithm (v33; https://hivfrenchresistance.org/) to analyse the resistance mutation profiles of the HIV reverse transcriptase sequences. RESULTS:During this period, the frequency of NNRTI use remained high, fluctuating slightly between 43.5% (n = 1782/4094) and 39.9% (n = 1758/4421). The use of efavirenz (10.8%-1.5%), nevirapine (7.0%-1.2%), and etravirine (11.0%-1.1%) decreased, whereas the use of rilpivirine (14.7%-26.3%) and doravirine (available from 2018, rising to 9.7% in 2023) increased. These trends were statistically significant for etravirine (P = 0.033) and rilpivirine (P < 0.001). Resistance rates for efavirenz, nevirapine and rilpivirine remained above 15% (efavirenz: 17.3%-16.6%, nevirapine: 16.9%-15.4% and rilpivirine: 17.6%-16.1%). This reflects significant cross-resistance between these three NNRTIs. By contrast, resistance rates were lower for etravirine (7.8%-6.0%) and doravirine (4.9%-4.6%), probably due to differences in their resistance profiles and higher genetic barriers to resistance. CONCLUSIONS:The NNRTI class of antiretroviral drugs remains widely used. Changes in the usage of drugs from this class have not altered the ecology of NNRTI resistance in antiretroviral drug-treated PLWHIV with virological failure during the studied period.

BACKGROUND:Doravirine is licensed in patients living with HIV (PWH) harbouring no prior resistance to any NNRTIs. We aimed to evaluate in real life the efficacy of doravirine with prior NNRTI virological failure and NNRTI resistance-associated mutations (RAMs). METHODS:This observational study included PWH switched to a doravirine-containing regimen between 30 September 2019 and 1 May 2022, with an HIV-1 RNA of ≤50 copies/mL and past NNRTI-RAMs. The main outcome was the proportion of participants with virological failure at Week 48 and Week 96. Secondary outcomes evaluated the rate of viral suppression and transient virological blip, RAMs in the case of virological failure and side effects. RESULTS:A total of 102 patients were analysed, mostly men (63%), with a median age of 59 years (IQR 51-63). The median time since HIV-1 diagnosis was 26 years (IQR 16-31), on ART for 22 years (IQR 14-26) and virally suppressed for 7 years (IQR 1-11).Of the patients analysed, 25/102 (25%) had documented historical RAMs to doravirine, 9/25 (36%) showing possible resistance and 16/25 (64%) showing major resistance. The resistance profile primarily (21/23) consisted of the K103N, Y181C and/or G190A/E reverse transcriptase substitutions. Median time since the last detection of NNRTI-RAMs was 12 years (5-17). Over 2 years follow-up, no virological failure occurred, neither at Week 48 (0/87; 0%) nor Week 96 (0/86; 0%). CONCLUSIONS:This is the first real-world study to provide new insight about the use of doravirine-containing regimens as a treatment in long-term suppressed patients whose viruses harboured specific NNRTI-RAMs in their history.

OBJECTIVE:To assess whether the COVID-19 and mpox outbreaks affected hepatitis C virus (HCV) related behaviours among men who have sex with men (MSM) with a cleared HCV infection. DESIGN:Longitudinal analysis from the international ICECREAM trial (2021-2024). METHODS:During the prerandomisation phase (i.e., without any intervention) individuals completed questionnaires on sexual and drug use behaviours and whether the COVID-19 (since start trial) or mpox (shortly after the mpox outbreak in 2022) outbreaks caused changes in these behaviours, all referring to the preceding 6 months. We used mixed-effects logistic regression to model changes in behaviours due to COVID-19 or mpox measures and mixed-effects linear regression to model the average HCV-MOSAIC risk score, as a proxy of HCV-associated risk behaviour, over calendar time. RESULTS:220 MSM ( n  = 117 from the Netherlands, n  = 103 from France) were included. Among 208 that completed the baseline questionnaire, 171 (82.2%) were MSM with HIV. The proportion of individuals reporting any impact of COVID-19 restrictions on risk behaviours, mainly lowering number of partners, decreased from 74.7% in September 2021 to 6.7% in September 2024 ( P  < 0.001) and reporting any impact of mpox from 41.9% in November 2022 to 6.0% in September 2024 ( P  = 0.001). The average HCV-MOSAIC risk score remained constant over time ( P  = 0.59) and was consistently ≥2.0, indicating high reinfection susceptibility. CONCLUSION:HCV-related behaviours decreased when COVID-19 and mpox measures were in place. However, individuals still engaged in behaviours associated with HCV, highlighting the importance of continued sexual health services and prevention efforts during such outbreaks.

Journal Article Ultra-rapid selection of the N74D capsid inhibitor resistance mutation after 3 weeks on lenacapavir Get access Marc Wirden, Marc Wirden INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Virology, AP-HP, Pitié Salpêtrière Hospital, Paris, France Corresponding author. E-mail: marc.wirden@aphp.fr https://orcid.org/0009-0003-3723-4142 Search for other works by this author on: Oxford Academic PubMed Google Scholar Cecile Pouderoux, Cecile Pouderoux Department of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Gilles Peytavin, Gilles Peytavin Pharmacology Department, AP-HP, Bichat Claude-Bernard University Hospital, Paris, France https://orcid.org/0000-0002-4359-537X Search for other works by this author on: Oxford Academic PubMed Google Scholar Basma Abdi, Basma Abdi INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Virology, AP-HP, Pitié Salpêtrière Hospital, Paris, France https://orcid.org/0000-0002-9139-0010 Search for other works by this author on: Oxford Academic PubMed Google Scholar Antoine Fayçal, Antoine Fayçal INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Romain Palich, Romain Palich INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France https://orcid.org/0000-0003-0047-0101 Search for other works by this author on: Oxford Academic PubMed Google Scholar Marc Antoine Valantin, Marc Antoine Valantin INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Sophie Seang, Sophie Seang INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Christine Katlama, Christine Katlama INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Vincent Calvez, Vincent Calvez INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Virology, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar ... Show more Valerie Pourcher, Valerie Pourcher INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Infectious Diseases, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Anne-Geneviève Marcelin Anne-Geneviève Marcelin INSERM, Institut Pierre Louis d'Epidémiologie et de Santé Publique (IPLESP), Sorbonne Université, Paris, F75013, FranceDepartment of Virology, AP-HP, Pitié Salpêtrière Hospital, Paris, France Search for other works by this author on: Oxford Academic PubMed Google Scholar Journal of Antimicrobial Chemotherapy, dkae115, https://doi.org/10.1093/jac/dkae115 Published: 17 April 2024

Objective: To describe the efficacy of intermittent nucleoside analogue-based (NA) regimen to maintain HBV virological suppression in HBV/HIV-1 patients. Methods: Conducted between 2014 and 2023, this observational retrospective study included all HBV (positive AgHbs)/HIV-1 coinfected patients with HIV RNA <= 50 cp/mL and HBV DNA <= 25 UI/mL who were switched to an intermittent (<7/7 days(D)) TDF or TAF-containing antiretroviral (ART) regimen. The primary outcome was the HBV virological success rate (SR) (proportion of patients with HBV pVL < 25 UI/mL) at W48. Results: Among 501 HBV/HIV-1 patients, 19(3.7 %) had switched to an intermittent NA-containing regimen that included TDF/FTC or TDF/3TC or TAF/FTC or TDF alone administered 5D-a-week(n = 7), 4D-a-week(n = 7) or 3D-a-week(n = 5). HBV virological success rates were 100 % [95 %CI 82.3-100] and 100 %[95 %CI 80.5-100] at W48 and W96(n = 17), respectively; with no viral HBV or HIV rebound (61.8 months (32.4-70.3) of follow-up). Conclusion: This case series shows the potential for intermittent NA-containing regimens to maintain long-term control of HBV replication among suppressed HBV/HIV-1 patients.

Introduction:We assessed the kinetics of the clearance of integrase strand transfer inhibitors resistance mutations (INSTIs-RMs) and associated factors from people living with HIV (PWH) displaying suppressed viral replication after virological failure (VF) on an INSTI regimen. Patients and methods:We included PWH with HIV-RNA viral loads ≤20 copies/mL for at least 5 years in whom INSTIs-RM had been identified at least once in a prior RNA resistance genotyping test. HIV DNAs were sequenced by Sanger sequencing (SS) and ultra-deep sequencing (UDS; detection threshold: 5%) every year over the preceding 5 years. Results:We included 39 PWH in the study. Most (95%) had experienced VF on a raltegravir-containing regimen. The past INSTIs-RMs were not detected in the peripheral blood mononuclear cells of 35 of the 39 (90%) PWH by SS at the end of follow-up. In a longitudinal analysis (2017-21) based on UDS, the previously detected INSTIs-RMs were not detected in 29 of the 35 (83%) PWH. In multivariable analysis, the duration of viral replication and the level of HIV-RNA during prior VF were significantly associated with the persistence of INSTIs-RM, with odds ratios of 1.05 per week of replication (95% CI, 1.00-1.11; P = 0.024) and 8.26 per log10 copies/mL (95% CI, 1.46-46.59; P = 0.017). Conclusions:We observed a clear trend towards the clearance of archived INSTIs-RM after a long period of virological control leading to changes in the resistance profile in cellular DNA, raising the possibility of studies assessing the recycling of INSTI classes even in the presence of a history of resistance.

Background and AimsIn France, Bulevirtide (BLV) was available in September 2019 through an early access program to treat patients with hepatitis Delta virus (HDV). The aim of this analysis was to evaluate the efficacy and safety of BLV in HIV patients with HDV coinfection.Patients and methodsPatients received BLV 2 mg +/- pegylated interferon (pegIFNα) according to the physician’s decision. The primary endpoint (per-protocol analysis) was the virological response rate at week 48, defined as the proportion of patients with undetectable serum HDV-RNA or HDV-RNA decline > 2 log10 IU/mL from baseline.ResultsCharacteristics of the 38 patients were as follow: 28 male, mean age 47.7 years, mean baseline HDV-RNA viral load 5.7 ± 1.2 log10 IU/ml. Median HIV viral load and mean CD4 count were 32 (30 - 65) cp/ml and 566 ± 307/mm3, respectively. Eight patients stopped treatment before week 48. At W48, 10 of 19 patients (52.6%) in the 2 mg BLV group, and 5 of 7 patients (71.4%) in the 2 mg BLV + pegIFNα group had reached virological response (no HDV-RNA available in 4 patients). At W48, 7/19 patients in the 2 mg BLV group, and 3/6 patients in the 2 mg BLV + pegIFNα group had combined response (virological response and normal ALT level).ConclusionAdults living with HIV coinfected with HDV can be treated by BLV with a virological response in more than 50% of patients. The combination of BLV and pegIFNα showed a strong virological response.IMPACT AND IMPLICATIONS• Bulevirtide is the only EMA-approved drug for HDV treatment and we showed that it can be used in adults living with HIV, with an overall good tolerability.• Bulevirtide induces a virologic response in more than 50% of patients suggesting that Bulevirtide should be considered as a first-line therapy in this specific population.• Bulevirtide in combination with pegIFNα could be used in patients without pegIFNα contra-indication.• No specific drug-drug interaction is reported.

BACKGROUND & AIMS: Nonalcoholic fatty liver disease (NAFLD) is a growing concern in the aging population with human immunodeficiency virus (HIV). Screening for NAFLD is recommended in patients with metabolic risk factors or unexplained transaminitis. This study aimed to prospectively assess the prevalence and associated factors of liver steatosis and advanced fibrosis (AF) in HIV-monoinfected patients at risk of NAFLD. METHODS: We conducted a multicenter study in HIV-monoinfected patients, nonexcessive drinkers with metabolic syndrome, and/or persistently elevated liver enzymes, and/or clinical lipodystrophy. All participants had magnetic resonance imaging proton density fat fraction (MRI-PDFF), Fibroscan/controlled attenuation parameter (CAP), and cytokine and genetic analysis. RESULTS: From March 2014 to November 2015, we enrolled 442 participants and analyzed 402: male (85%); median age, 55 years (interquartile range [IQR], 50-61 years); body mass index, 27.0 kg/m(2) (IQR, 23.6-28.7 kg/m(2)); metabolic syndrome (67%); and CD4 cell count, 630/mm(3) (IQR, 510-832/mm(3)). Overall 257 of 402 (64%) had NAFLD (MRI-PDFF >= 5%). Among them, 11.3% had a liver stiffness >= 9.6 kPa, suggestive of AF. Multivariable analysis identified 7 factors of steatosis: high CD4-cell count (odds ratio [OR], 4.04; 95% confidence interval [CI], 1.92-8.51), high leptin level (OR, 2.12; 95% CI, 1.14-3.93), non-CC PNPLA3s738409 genetic polymorphism (OR, 1.92; 95% CI, 1.11-3.33), low high-density lipoprotein (OR, 1.83; 95% CI, 1.03-3.27), high triglycerides (OR, 1.48; 95% CI, 1.18-1.84), elevated alanine transaminase (OR, 1.23; 95% CI, 1.16-1.31), and hyper ferritinemia (OR, 1.05; 95% CI, 1.03-1.07). Two factors were associated with AF: high body mass index (OR, 1.23; 95% CI, 1.07-1.42; P = .005, and elevated aspartate aminotransferase (OR, 1.03; 95% CI, 1.01-1.05; P = .001). Using MRI-PDFF as a reference, CAP (best cutoff, 280 dB/m) had good accuracy (area under the receiver operating characteristic curve = 0.86; 95% CI, 0.82-0.90) for the diagnosis of moderate to severe steatosis. CONCLUSIONS: In a large cohort of HIV-moninfected patients at risk of NAFLD, steatosis is present in two-thirds of cases, and around 10% have AF. The CAP technique is accurate for screening steatosis in this population.

Objectives: To assess whether antiretroviral therapy (ART) prescriptions differ between naive and virally suppressed HIV patients born in France (PBFs) and in Sub-Saharan Africa (PBSSAs). Setting: Observational single-center study. Methods: We included all PBFs and PBSSAs who entered into care at Pitié-Salpêtrière Hospital, Paris, France, from 01/01/2000 to 31/12/2018, with plasma HIV-RNA>200 copies/mL. We first compared the initial ART in naive PBFs and PBSSAs. Second, we compared the last-prescribed ART (including drug-reduced ART: daily 2-drug regimens, daily 1-drug regimens and intermittent 3-drug regimens) in virally suppressed PBFs and PBSSAs, by focusing on patients in care in 2018 with HIV-RNA <50 copies for at least 24 months. A univariable and multivariable logistic regression model was used to assess the impact of geographical origin on ART prescriptions. Results: A total of 1944 naive patients were included (915 PBSSAs and 1029 PBFs). PBSSAs were more frequently women, hepatitis B coinfected, with a lower pretherapeutic CD4 T-cell count, and most had tuberculosis at HIV diagnosis. After adjustment for confounders, PBSSAs were more likely to receive a first-line protease inhibitor-based regimen (OR 1.61, 95% CI: 1.31 to 1.98), and less likely to receive an integrase inhibitor-based regimen (OR 0.61, 95% CI: 0.42 to 0.88). Of the 968 virally suppressed patients (431 PBSSAs and 537 PBFs), PBSSAs were less likely to receive drug-reduced ART, including 2-drug regimens and intermittent three-drug regimens (OR 0.48, 95% CI: 0.36 to 0.65). Conclusions: Differences in ART prescriptions between PBSSAs and PBFs were not only explained by different clinical and virologic situations. Personal motivations of doctors in choosing ART according to country of birth need to be explored.

INTRODUCTION Preexposure prophylaxis (PrEP) with tenofovir disoproxil/emtricitabine is a powerful tool to prevent HIV acquisition and provides an opportunity to offer comprehensive prevention services, including assessment for sexually transmitted infections and evaluation of immune status towards vaccinepreventable viral infections such those due hepatitis A virus (HAV), hepatitis B virus (HBV) and human papillomavirus (HPV). In addition to the programme in girls aged 11–19, French guidelines recommended HPV vaccination for men who have sex with men (MSM) ≤26 years old in February 2016 and for all boys aged 11–19 in December 2019. There are no restrictions on HPV vaccine use beyond these age limits, but the cost is not covered by the French Health Service. In reallife settings, suboptimal vaccination coverage against HPV as well as HBV has been reported among European MSM. 4 Our objective was to evaluate HAV, HBV and HPV vaccine needs and coverage in individuals initiating PrEP in a sexual health clinic in Paris. In this observational retrospective singlecentre study, we reviewed all individuals who initiated PrEP between 1 January 2016 and 31 December 2020 with ≥1 year of followup after PrEP initiation. At baseline, we assessed the presence of HAV and HBV antibodies and HPV vaccination status. Immune protection against HAV and HBV was defined as the presence of antiHAV IgG index S/CO >1.00 and antiHBs IgG >10 IU/L, respectively. HPV vaccination status at baseline was assessed through the participants’ recall. Subsequently, we assessed vaccine prescription by physicians for nonimmune and unvaccinated participants, followed by a review of completion of vaccination. Vaccination schedules were considered complete after 2 doses for HAV with a time interval of 0 and 6 months; 3 doses for HBV with an interval of 0, 1 and 6 months; and 3 doses for HPV with an interval of 0, 2 and 6 months. Contrary to HAV and HBV vaccines, HPV vaccine was not accessible in the sexual health centre and had to be purchased from a private pharmacy. Finally, we assessed overall HAV, HBV immune status combining immune protection acquired in the past or by vaccination after PrEP initiation and HPV vaccine coverage. HPV vaccine completion was analysed by age groups ≤26 years old or >26 years old. If any information was missing, individuals were contacted by phone or email to determine whether vaccination had been performed and if not, the reason why. All clinical, biological and prescription data are routinely documented in an electronic health record (NADIS), for which all patients gave consent for the collection and use of their anonymised data after approval by the CNIL (French Data Protection Authority; CNIL authorisation number: 2085881). Statistical data are presented with total numbers and proportions and compared by a χ test. A p value <0.05 was considered statistically significant. A total of 591 PrEP users were analysed. All were MSM with a median age of 33 years (IQR 28–41), including 118 participants (20%) aged ≤26 years. At baseline (table 1), 57.7% (341/591) of PrEP users were immune against HAV and 73.4% (434/591) against HBV. Vaccines were prescribed for 93.2% (233/250) of HAV nonimmune and 87.2% (137/157) of HBV nonimmune participants. Vaccination was completed in 85.8% (200/233) and in 91.2% (125/137) individuals with a HAV and HBV vaccine prescription, respectively. Our results are consistent with other studies where HAV vaccination rates were high, especially among PrEP users. With regards to HPV, only 7 of the 591 (1.2%) individuals had been vaccinated before PrEP initiation, including 4/118 (3.4%) individuals aged ≤26 years. The prescription rate by physicians remained low throughout the study period at 26% (152/584) for all ages and 39.5% (45/114) for those ≤26 years. These results are in agreement with those of other studies which report infrequent HPV vaccination prescription by physicians. 6 Following prescription, the HPV vaccine completion rate was 54.6% (83/152) including 64.4% (29/45) in participants aged ≤26 years. Of 69 individuals who did not complete HPV vaccination despite prescription, 5 (7%) participants did not respond to the questionnaire and 64 (93%) reported the following reasons: forgetting to go to a pharmacy for vaccine delivery (n=29), not feeling at risk (n=20), lost prescription (n=6) and vaccine cost (n=9, all >26 y.o). Several factors may explain our findings: recentness of the French guidelines, vaccine cost and lack of motivation for HPV vaccination, as onethird of the participants described not feeling at risk for this viral oncogenic disease. Finally, combining immunity acquired in the past or by vaccination after PrEP initiation, the overall immune protection rate for these 591 MSM initiating PrEP was 91.5% for HAV, 94.6% for HBV and 15.2% for HPV, including 28% in the ≤26 years age group and 12% in the >26 years age group. Given the high burden of HPVattributable lesions in MSM compared with heterosexual men, a change in Letter