Surveillance of HIV-1 primary infections in France from 2014 to 2016: toward stable resistance, but higher diversity, clustering and virulence?

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.

Integrase strand transfer inhibitor (INSTI)-resistance mutations for the surveillance of transmitted HIV-1 drug resistance.

BACKGROUND: Integrase strand transfer inhibitors (INSTIs) are expected to be widely adopted globally, requiring surveillance of resistance emergence and transmission. OBJECTIVES: We therefore sought to develop a standardized list of INSTI-resistance mutations suitable for the surveillance of transmitted INSTI resistance. METHODS: To characterize the suitability of the INSTI-resistance mutations for transmitted HIV-1 drug resistance (TDR) surveillance, we classified them according to their presence on published expert lists, conservation in INSTI-naive persons, frequency in INSTI-treated persons and contribution to reduced in vitro susceptibility. Mutation prevalences were determined using integrase sequences from 17302 INSTI-naive and 2450 INSTI-treated persons; 53.3% of the INSTI-naive sequences and 20.0% of INSTI-treated sequences were from non-B subtypes. Approximately 10% of sequences were from persons who received dolutegravir alone or a first-generation INSTI followed by dolutegravir. RESULTS: Fifty-nine previously recognized (or established) INSTI-resistance mutations were present on one or more of four published expert lists. They were classified into three main non-overlapping groups: 29 relatively common non-polymorphic mutations, occurring in five or more individuals and significantly selected by INSTI treatment; 8 polymorphic mutations; and 22 rare mutations. Among the 29 relatively common INSTI-selected mutations, 24 emerged as candidates for inclusion on a list of INSTI surveillance drug-resistance mutations: T66A/I/K, E92G/Q, G118R, F121Y, E138A/K/T, G140A/C/S, Y143C/H/R/S, S147G, Q148H/R/K, N155H, S230R and R263K. CONCLUSIONS: A set of 24 non-polymorphic INSTI-selected mutations is likely to be useful for quantifying INSTI-associated TDR. This list may require updating as more sequences become available from INSTI-experienced persons infected with HIV-1 non-subtype B viruses and/or receiving dolutegravir.

Resistance to integrase inhibitors: a national study in HIV-1-infected treatment-naive and -experienced patients.

OBJECTIVES: To describe integrase strand transfer inhibitor (INSTI) resistance profiles and factors associated with resistance in antiretroviral-naive and -experienced patients failing an INSTI-based regimen in clinical practice. METHODS: Data were collected from patients failing an INSTI-containing regimen in a multicentre French study between 2014 and 2017. Failure was defined as two consecutive plasma viral loads (VL) >50 copies/mL. Reverse transcriptase, protease and integrase coding regions were sequenced at baseline and failure. INSTI resistance-associated mutations (RAMs) included in the Agence Nationale de Recherches sur le SIDA genotypic algorithm were investigated. RESULTS: Among the 674 patients, 359 were failing on raltegravir, 154 on elvitegravir and 161 on dolutegravir therapy. Overall, 90% were experienced patients and 389 (58%) patients showed no INSTI RAMs at failure. The strongest factors associated with emergence of at least one INSTI mutation were high VL at failure (OR = 1.2 per 1 log10 copies/mL increase) and low genotypic sensitivity score (GSS) (OR = 0.08 for GSS ≥3 versus GSS = 0-0.5). Patients failing dolutegravir also had significantly fewer INSTI RAMs at failure than patients failing raltegravir (OR = 0.57, P = 0.02) or elvitegravir (OR = 0.45, P = 0.005). Among the 68 patients failing a first-line regimen, 11/41 (27%) patients on raltegravir, 7/18 (39%) on elvitegravir and 0/9 on dolutegravir had viruses with emergent INSTI RAMs at failure. CONCLUSIONS: These results confirmed the robustness of dolutegravir regarding resistance selection in integrase in the case of virological failure in routine clinical care.

Stable prevalence of transmitted drug resistance mutations and increased circulation of non-B subtypes in antiretroviral-naive chronically HIV-infected patients in 2015/2016 in France.

OBJECTIVES: We estimated the prevalence of transmitted-drug-resistance-associated mutations (TDRAMs) in antiretroviral-naive chronically HIV-1-infected patients. PATIENTS AND METHODS: TDRAMs were sought in samples from 660 diagnosed HIV-1-infected individuals in 2015/2016 in 33 HIV clinical centres. Weighted analyses, considering the number of patients followed in each centre, were used to derive representative estimates of the percentage of individuals with TDRAMs. Results were compared with those of the 2010/2011 survey (n = 661) using the same methodology. RESULTS: At inclusion, median CD4 cell counts and plasma HIV-1 RNA were 394 and 350/mm3 (P = 0.056) and 4.6 and 4.6 log10 copies/mL (P = 0.360) in the 2010/2011 survey and the 2015/2016 survey, respectively. The frequency of non-B subtypes increased from 42.9% in 2010/2011 to 54.8% in 2015/2016 (P < 0.001), including 23.4% and 30.6% of CRF02_AG (P = 0.004). The prevalence of virus with protease or reverse-transcriptase TDRAMs was 9.0% (95% CI = 6.8-11.2) in 2010/2011 and 10.8% (95% CI = 8.4-13.2) in 2015/2016 (P = 0.269). No significant increase was observed in integrase inhibitor TDRAMs (6.7% versus 9.2%, P = 0.146). Multivariable analysis showed that men infected with the B subtype were the group with the highest risk of being infected with a resistant virus compared with others (adjusted OR = 2.2, 95% CI = 1.3-3.9). CONCLUSIONS: In France in 2015/2016, the overall prevalence of TDRAMs was 10.8% and stable compared with 9.0% in the 2010/2011 survey. Non-B subtypes dramatically increased after 2010. Men infected with B subtype were the group with the highest risk of being infected with a resistant virus, highlighting the need to re-emphasize safe sex messages.

Anti-gp41 antibody levels reflect HIV viral suppression and cellular reservoir in long-term antiretroviral-treated trial participants.

BACKGROUND: A major challenge to HIV cure strategies is the quantification of persistent reactivation-prone virus in people living with HIV. OBJECTIVES: To determine whether anti-gp41 antibody levels correlate with viral suppression and HIV-1 DNA levels in patients on ART. METHODS: Participants with plasma HIV-1 RNA below 50 copies/mL for >12 months were included from three ANRS cohorts (COPANA, MONOI and APROCO). Antibody levels to gp41 were measured by a low-sensitivity enzyme-linked immunoassay. Correlations with patient and virus characteristics, plasma HIV-1 RNA load (standard and ultrasensitive tests) and cell-associated HIV-1 DNA were assessed. RESULTS: Median age was 41 years and 77.5% of the 683 participants were men. Median CD4+ T cell count was 582 cells/mm3 and median viral suppression duration was 6.6 years (IQR 2.0-9.5). The overall median anti-gp41 antibody titre was 1.3 (IQR 0.6-1.9); median HIV-1 DNA level was 2.6 (IQR 2.1-3.0) log10 copies/106 leucocytes; and HIV-1 RNA was undetectable in 56% of samples. A lower titre of anti-gp41 antibodies correlated with male gender, longer viral suppression and lower HIV-1 DNA burden. Sustained undetectable HIV-1 RNA was associated with lower anti-gp41 levels [median 1.1 (IQR 0.5-1.6) versus 1.4 (IQR 0.7-1.9), P = 0.009]. CONCLUSIONS: Anti-gp41 levels decreased with the duration of antiviral suppression on ART. Lower titres were associated with lower HIV-1 DNA levels and longer duration of viral suppression, reflecting minimal antigen stimulation. Anti-gp41 antibody titration may be a useful biomarker reflecting long-term HIV-1 suppression on ART.

Ultrasensitive Human Immunodeficiency Virus Type 1 Viral Load as a Marker of Treatment Choice for Simplification Strategies.

Background: Using 3 randomized Protease inhibitor (PI) monotherapy studies: Kalesolo, Dream and Monoi, we performed a pooled-analysis. Our objective was to determine in PI monotherapy and standard tritherapy: 1) distribution of ultrasensitive viral load (USVL) at week 96 (W96); 2) factors associated with virological failure (VF) at W96 and 3) factors associated with USVL<1 copy at W96. Methods: VF was defined as 2 consecutive measurements of Human Immunodeficiency Virus Type 1 RNA viral load>50 copies/mL and analysed in Intention-To-Treat. A logistic model was used to investigate which variables were predictive of a VF and Fisher test to investigate differences in USVL at W96. Results: Among 609 patients, 73% were male with median age of 44.4 years (IQR 39.8-52.1), baseline CD4/CD8 ratio was 0.8 (IQR 0.6-1.10), baseline CD4 was 564.5/mm3 (IQR 422-707) and 59% presented a baseline USVL<1 copy/mL. At W96, the proportion of USVL<1 copy/mL was significantly different between PI monotherapy and standard tritherapy in pooled-analysis (65% versus 74%; p=0.04). Overall, baseline USVL<1copy/mL, tritherapy and to be a female were associated with USVL<1 copy/mL at W96 (p<0.0001, p=0.049 and p=0.006). In PI monotherapy receiving DRV/r was associated with USVL<1 copy/mL at W96 (p=0.003). Factors associated to virological succes at W96 were higher baseline CD4 (p=0.034) and baseline USVL<1 copy/mL (p=0.0005). Conclusion: Pooled-analysis of 3 PI monotherapy trials showed better efficacy of tritherapy in terms of USVL at W96. Furthermore regarding USVL at W96, to receive LPV/r seems to be more deleterious than DRV/r. Baseline USVL impacts VF at W96 more specifically in tritherapy arm. Clinical Trials Registration: NCT00421551, NCT00946595, and NCT00140751.

Antiretroviral-treated HIV-1 patients can harbour resistant viruses in CSF despite an undetectable viral load in plasma.

Background: HIV therapy reduces the CSF HIV RNA viral load (VL) and prevents disorders related to HIV encephalitis. However, these brain disorders may persist in some cases. A large population of antiretroviral-treated patients who had a VL > 1.7 log 10 copies/mL in CSF with detectable or undetectable VL in plasma associated with cognitive impairment was studied, in order to characterize discriminatory factors of these two patient populations. Methods: Blood and CSF samples were collected at the time of neurological disorders for 227 patients in 22 centres in France and 1 centre in Switzerland. Genotypic HIV resistance tests were performed on CSF. The genotypic susceptibility score was calculated according to the last Agence Nationale de Recherche sur le Sida et les hépatites virales Action Coordonnée 11 (ANRS AC11) genotype interpretation algorithm. Results: Among the 227 studied patients with VL > 1.7 log 10 copies/mL in CSF, 195 had VL detectable in plasma [median (IQR) HIV RNA was 3.7 (2.7-4.7) log 10 copies/mL] and 32 had discordant VL in plasma (VL < 1.7 log 10 copies/mL). The CSF VL was lower (median 2.8 versus 4.0 log 10 copies/mL; P < 0.001) and the CD4 cell count was higher (median 476 versus 214 cells/mm 3 ; P < 0.001) in the group of patients with VL < 1.7 log 10 copies/mL in plasma compared with patients with plasma VL > 1.7 log 10 copies/mL. Resistance to antiretrovirals was observed in CSF for the two groups of patients. Conclusions: Fourteen percent of this population of patients with cognitive impairment and detectable VL in CSF had well controlled VL in plasma. Thus, it is important to explore CSF HIV (VL and genotype) even if the HIV VL is controlled in plasma because HIV resistance may be observed.

Qualitative and quantitative HIV antibodies and viral reservoir size characterization in vertically infected children with virological suppression.

Background: Absence of detectable viraemia after treatment cessation in some vertically HIV-infected (VHIV) children suggests that early initiation of HAART could lead to functional cure. Objectives: We described the factors associated with HIV antibody levels and the viral reservoir size in HAART-treated VHIV children. Methods: Study included 97 VHIV children with virological suppression, in Bamako, Mali. The anti-gp41 antibody activities and HIV serostatus were assessed. The viral reservoir size was measured by quantifying total cell-associated HIV DNA. Results: Among the children studied, the median total HIV DNA level was 445 copies/10 6 cells (IQR = 187-914) and the median anti-gp41 antibody activity was 0.29 OD (IQR = 0.18-0.75). Low activity of anti-gp41 antibodies was associated with a younger age of HAART initiation ( P = 0.01). Overall, eight HIV-1 seroreversions were identified. Conclusions: Study identified potential candidates with low viral reservoir and low antibody levels or activities for future trials aiming to reduce HIV-1 reservoir to limit HAART duration.

Antiretroviral-naive and -treated HIV-1 patients can harbour more resistant viruses in CSF than in plasma.

OBJECTIVES: The neurological disorders in HIV-1-infected patients remain prevalent. The HIV-1 resistance in plasma and CSF was compared in patients with neurological disorders in a multicentre study. METHODS: Blood and CSF samples were collected at time of neurological disorders for 244 patients. The viral loads were >50 copies/mL in both compartments and bulk genotypic tests were realized. RESULTS: On 244 patients, 89 and 155 were antiretroviral (ARV) naive and ARV treated, respectively. In ARV-naive patients, detection of mutations in CSF and not in plasma were reported for the reverse transcriptase (RT) gene in 2/89 patients (2.2%) and for the protease gene in 1/89 patients (1.1%). In ARV-treated patients, 19/152 (12.5%) patients had HIV-1 mutations only in the CSF for the RT gene and 30/151 (19.8%) for the protease gene. Two mutations appeared statistically more prevalent in the CSF than in plasma: M41L (P=0.0455) and T215Y (P=0.0455). CONCLUSIONS: In most cases, resistance mutations were present and similar in both studied compartments. However, in 3.4% of ARV-naive and 8.8% of ARV-treated patients, the virus was more resistant in CSF than in plasma. These results support the need for genotypic resistance testing when lumbar puncture is performed.