Intravenous ZDV remains an effective tool to reduce transmission in cases of virological failure, even in cART-treated women. However, for the vast majority of women with low viral loads at delivery, in the absence of obstetrical risk factors, systematic intravenous ZDV appears to be unnecessary.

Objectives Patients with primary HIV-1 infection (PHI) are a particular population, giving important insight about ongoing evolution of transmitted drug resistance-associated mutation (TDRAM) prevalence, HIV diversity and clustering patterns. We describe these evolutions of PHI patients diagnosed in France from 2014 to 2016. Methods A total of 1121 PHI patients were included. TDRAMs were characterized using the 2009 Stanford list and the French ANRS algorithm. Viral subtypes and recent transmission clusters (RTCs) were also determined. Results Patients were mainly MSM (70%) living in the Paris area (42%). TDRAMs were identified among 10.8% of patients and rose to 18.6% when including etravirine and rilpivirine TDRAMs. Prevalences of PI-, NRTI-, first-generation NNRTI-, second-generation NNRTI- and integrase inhibitor-associated TDRAMs were 2.9%, 5.0%, 4.0%, 9.4% and 5.4%, respectively. In a multivariable analysis, age >40 years and non-R5 tropic viruses were associated with a >2-fold increased risk of TDRAMs. Regarding HIV diversity, subtype B and CRF02_AG (where CRF stands for circulating recombinant form) were the two main lineages (56% and 20%, respectively). CRF02_AG was associated with higher viral load than subtype B (5.83 versus 5.40 log10 copies/mL, P = 0.004). We identified 138 RTCs ranging from 2 to 14 patients and including overall 41% from the global population. Patients in RTCs were younger, more frequently born in France and more frequently MSM. Conclusions Since 2007, the proportion of TDRAMs has been stable among French PHI patients. Non-B lineages are increasing and may be associated with more virulent CRF02_AG strains. The presence of large RTCs highlights the need for real-time cluster identification to trigger specific prevention action to achieve better control of the epidemic.

Background Safety data about rilpivirine use during pregnancy remain scarce, and rilpivirine plasma concentrations are reduced during second/third trimesters, with a potential risk of viral breakthroughs. Thus, French guidelines recommend switching to rilpivirine-free combinations (RFCs) during pregnancy. Objectives To describe the characteristics of women initiating pregnancy while on rilpivirine and to compare the outcomes for virologically suppressed subjects continuing rilpivirine until delivery versus switching to an RFC. Methods In the ANRS-EPF French Perinatal cohort, we included women on rilpivirine at conception in 2010–18. Pregnancy outcomes were compared between patients continuing versus interrupting rilpivirine. In women with documented viral suppression (<50 copies/mL) before 14 weeks of gestation (WG) while on rilpivirine, we compared the probability of viral rebound (≥50 copies/mL) during pregnancy between subjects continuing rilpivirine versus those switching to RFC. Results Among 247 women included, 88.7% had viral suppression at the beginning of pregnancy. Overall, 184 women (74.5%) switched to an RFC (mostly PI/ritonavir-based regimens) at a median gestational age of 8.0 WG. Plasma HIV-1 RNA nearest delivery was <50 copies/mL in 95.6% of women. Among 69 women with documented viral suppression before 14 WG, the risk of viral rebound was higher when switching to RFCs than when continuing rilpivirine (20.0% versus 0.0%, P = 0.046). Delivery outcomes were similar between groups (overall birth defects, 3.8/100 live births; pregnancy losses, 2.0%; preterm deliveries, 10.6%). No HIV transmission occurred. Conclusions In virologically suppressed women initiating pregnancy, continuing rilpivirine was associated with better virological outcome than changing regimen. We did not observe a higher risk of adverse pregnancy outcomes.

HIV Pre-Exposure Prophylaxis (PrEP) is effective in Men who have Sex with Men (MSM), and is reimbursed by the social security in France. Yet, PrEP is underused due to the difficulty to identify people at risk of HIV infection outside the “sexual health” care path. We developed and validated an automated algorithm that re-uses Electronic Health Record (EHR) data available in eHOP, the Clinical Data Warehouse of Rennes University Hospital (France). Using machine learning methods, we developed five models to predict incident HIV infections with 162 variables that might be exploited to predict HIV risk using EHR data. We divided patients aged 18 or more having at least one hospital admission between 2013 and 2019 in two groups: cases (patients with known HIV infection in the study period) and controls (patients without known HIV infection and no PrEP in the study period, but with at least one HIV risk factor). Among the 624,708 admissions, we selected 156 cases (incident HIV infection) and 761 controls. The best performing model for identifying incident HIV infections was the combined model (LASSO, Random Forest, and Generalized Linear Model): AUC = 0.88 (95% CI: 0.8143-0.9619), specificity = 0.887, and sensitivity = 0.733 using the test dataset. The algorithm seems to efficiently identify patients at risk of HIV infection.

Background Successful 2-drug regimens (2DRs) for HIV were made possible by the availability of drugs combining potency and tolerability with a high genetic barrier to resistance. How these deal with resistance development/re-emergence, compared with 3DRs, is thus of paramount importance. Materials and methods A national survey including patients who were either naive or experienced with any 2DR or 3DR but failing integrase strand transfer inhibitor (INSTI)-containing regimens [two consecutive plasma viral load (VL) values >50 copies/mL] was conducted between 2014 and 2019. Genotypic resistance tests were interpreted with the v28 ANRS algorithm. Results Overall, 1104 patients failing any INSTI-containing regimen (2DRs, n = 207; 3DRs, n = 897) were analysed. Five hundred and seventy-seven (52.3%) patients were infected with a B subtype and 527 (47.3%) with non-B subtypes. Overall, 644 (58%) patients showed no known integrase resistance mutations at failure. In multivariate analysis, factors associated with the emergence of at least one integrase mutation were: high VL at failure (OR = 1.24 per 1 log10 copies/mL increase); non-B versus B subtype (OR = 1.75); low genotypic sensitivity score (GSS) (OR = 0.10 for GSS = 2 versus GSS = 0–0.5); and dolutegravir versus raltegravir (OR = 0.46). Although 3DRs versus 2DRs reached statistical significance in univariate analysis (OR = 0.59, P = 0.007), the variable is not retained in the final model. Conclusions This study is one of the largest studies characterizing integrase resistance in patients failing any INSTI-containing 2DR or 3DR in routine clinical care and reveals factors associated with emergence of integrase resistance that should be taken into consideration in clinical management. No difference was evidenced between patients receiving a 2DR or a 3DR.

Background Two-drug regimens based on integrase strand transfer inhibitors (INSTIs) and boosted PIs have entered recommended ART. However, INSTIs and boosted PIs may not be suitable for all patients. We aimed to report our experience with doravirine/lamivudine as maintenance therapy in people living with HIV (PLWH) followed in French HIV settings. Methods This observational study enrolled all adults who initiated doravirine/lamivudine between 1 September 2019 and 31 October 2021, in French HIV centres participating in the Dat’AIDS cohort. The primary outcome was the rate of virological success (plasma HIV-RNA < 50 copies/mL) at Week (W)48. Secondary outcomes included: rate of treatment discontinuation for non-virological reasons, evolution of CD4 count and CD4/CD8 ratio over follow-up. Results Fifty patients were included, with 34 (68%) men; median age: 58 years (IQR 51–62), ART duration: 20 years (13–23), duration of virological suppression: 14 years (8–19), CD4 count: 784 cells/mm3 (636–889). Prior to switching, all had plasma HIV-RNA < 50 copies/mL. All but three were naive to doravirine, and 36 (72%) came from a three-drug regimen. Median follow-up was 79 weeks (IQR 60–96). Virological success rate at W48 was 98.0% (95% CI 89.4–99.9). One virological failure occurred at W18 (HIV-RNA = 101 copies/mL) in a patient who briefly discontinued doravirine/lamivudine due to intense nightmares; there was no resistance at baseline and no resistance emergence. There were three strategy discontinuations for adverse events (digestive disorders: n = 2; insomnia: n = 1). There was no significant change in CD4/CD8 ratio, while CD4 T cell count significantly increased. Conclusions These preliminary findings suggest that doravirine/lamivudine regimens can maintain high levels of viral suppression in highly ART-experienced PLWH with long-term viral suppression, and good CD4+ T cell count.