Once-daily dolutegravir versus darunavir plus cobicistat in adults at the time of primary HIV-1 infection: the OPTIPRIM2-ANRS 169 randomized, open-label, Phase 3 trial

Abstract

Background

Whether integrase strand transfer inhibitors (INSTIs) can decrease HIV-1 DNA levels more rapidly than boosted PIs during primary HIV-1 infection (PHI) is unknown. We hypothesized that once-daily dolutegravir/tenofovir/emtricitabine could reduce the viral reservoir through rapid viral replication control further than once-daily darunavir/cobicistat/tenofovir/emtricitabine.

Methods

The OPTIPRIM2-ANRS 169 study was a randomized (1:1), open-label, multicentre trial in adults with ≤5 or ≤3 HIV antibodies detected, respectively, by western blot or immunoblot in the last 10 days. The primary endpoint was total HIV-1 DNA levels in PBMCs at Week 48 (W48) adjusted for baseline levels. The main secondary endpoint was HIV-1 RNA level decrease.

Results

Between April 2017 and August 2018, 101 patients were included from 31 hospitals. Most patients were men (93%), the median age was 36 years and 17% were Fiebig stage ≤3. The median (IQR) plasma HIV-1 RNA and DNA levels were, respectively, 5.8 (5.0–6.6) and 3.87 (3.52–4.15) log₁₀ copies/million PBMCs. The median (IQR) decreases in HIV-1 DNA levels at W48 were −1.48 (−1.74 to −1.06) and −1.39 (−1.55 to −0.98) log₁₀ copies/million PBMCs in the dolutegravir and darunavir/cobicistat groups, respectively (P = 0.52). Plasma HIV-1 RNA levels were <50 copies/mL in 24% versus 0% of patients in the dolutegravir and darunavir/cobicistat groups at W4, 55% versus 2% at W8, 67% versus 17% at W12, and 94% versus 90% at W48, respectively.

Conclusions

Dolutegravir-based and darunavir-based regimens initiated during PHI strongly and similarly decreased the blood reservoir size. Considering the rapid viral suppression during a period of high HIV-1 transmission risk, dolutegravir-based regimens are a major first-line option.

Introduction

Primary HIV-1 infection (PHI) is a critical phase, during which early treatment rapidly decreases activated and resting infected cell counts,¹ compared with treating patients during the chronic phase.² Such a therapy is also essential to limit the risk of HIV-1 transmission,³ protect the immune system and optimize its recovery,^4,5 and to limit viral genetic diversity.⁶ Patients treated from the time of PHI could be good candidates for HIV cure intervention due to their smaller reservoir and preserved immune functions that are impaired during the chronic phase.

The use of integrase strand transfer inhibitors (INSTIs) such as dolutegravir or bictegravir has been extended by experts and current guidelines^7,8 as an alternative treatment, with the preferred use of dolutegravir in PHI patients,⁷ in line with the recommendations for chronic HIV patients. We hypothesized that by using an INSTI-based combined ART (cART) during PHI, the early and rapid inhibition of HIV-1 RNA replication could limit reservoir seeding. The aim of the randomized OPTIPRIM2-ANRS 169 study (ANRS | Emerging Infectious Diseases) was to assess the effect of dolutegravir-based cART on HIV-1 DNA, a marker of blood HIV reservoir well adapted to clinical trials, compared with darunavir-based cART in patients diagnosed early during PHI. Dolutegravir was chosen for its potent antiviral effect with a proven superiority over darunavir-based regimens during the chronic phase,⁹ and for its higher genetic barrier to HIV drug resistance compared with raltegravir and elvitegravir.¹⁰

Methods

Study design and participants

The OPTIPRIM2-ANRS 169 trial was a randomized, two-arm, open-label, multicentre, Phase 3 study conducted in 40 French hospitals closely involved in the management of HIV patients. Patients were included from April 2017 for an 18 month period.

Inclusion criteria were patients diagnosed during the early PHI phase, symptomatic or not, and meeting the following criteria in the last 10 days: (i) a negative ELISA test and a positive result for HIV-1 RNA, confirmed by a second positive result for HIV-1 RNA; (ii) an ELISA with negative HIV antibodies (Ab⁻) and positive p24⁺ antigen, confirmed by a positive result for HIV-1 RNA; (iii) an ELISA⁺ or ELISA with Ab⁺/p24⁺ signals or ELISA⁻ and detection of 0–5 HIV-1 band(s) or 0–3 HIV-1 band(s) by western blot or immunoblot, respectively, confirmed by a positive result for HIV-1 RNA.

The main exclusion criteria were: pregnant or breastfeeding women; patients with isolated HIV-2 infection; malignancy; cirrhosis; hepatic aminotransferase or total bilirubin concentrations >10 times the upper limit of normal; a prothrombin time >50%; a glomerular filtration rate estimated by the Cockcroft-Gault formula <70 mL/min; and having received post-exposure prophylaxis with cART or pre-exposure prophylaxis in the last 4 weeks.

The study was approved by the French Agency for the Safety of Medicines and Health Products (ANSM) and our institutional Ethics Committee (CPP Ile de France VII). All participants gave their written informed consent.

Randomization and blinding

Minimization randomization was used to randomly assign patients (1:1) to one of the two treatment arms. Randomization was stratified for the four pathophysiological substudies in order to balance the number of patients in each treatment arm for each substudy. Patients and care providers were not blinded to treatment allocation. Investigators analysing the blood samples for HIV-1 DNA and ultrasensitive HIV-1 RNA loads were blinded to treatment allocation.

Procedure

Patients randomized to the experimental arm were treated with dolutegravir-based cART (dolutegravir group) and took one tablet containing 50 mg of dolutegravir and one combination tablet containing 200 mg of emtricitabine and 245 mg of tenofovir once daily. Patients assigned to the comparator darunavir-based cART (darunavir/cobicistat group) took one combination tablet containing 800 mg of darunavir and 150 mg of cobicistat plus one combination tablet containing 200 mg of emtricitabine and 245 mg of tenofovir once daily.

Patients were asked to participate in several pathophysiological substudies including immunovirological, semen, pharmacological and rectal assessments. They were also asked to be co-enrolled in the ANRS C06 PRIMO cohort (NCT03148964) for an extended long-term follow-up after the current trial.

Clinical examinations and laboratory tests were performed at baseline and at Week 2 (W2), W4, W8, W12, W24, W36 and W48. Plasma HIV-1 RNA levels, CD4+ and CD8+ T cell counts, blood cell counts and blood chemistry values were measured locally. Quantification of total cell-associated HIV-1 DNA and plasma HIV-1 RNA levels using ultrasensitive methods was centralized in the virology laboratory of Necker Hospital (Paris, France). HIV-1 DNA was quantified on thawed whole blood using an ultrasensitive real-time PCR method (Generic HIV-1 DNA assay, Biocentric, Bandol, France) with a detection threshold of 5 copies/PCR¹¹ and each DNA extract was tested in two to four replicates. Results are reported as the number of HIV-1 DNA copies/10⁶ PBMCs.¹² Plasma HIV-1 RNA levels were determined using an ultrasensitive method; 1–10 mL of plasma was ultracentrifuged, and a detection threshold of 2–20 copies/mL was set on the Roche Cobas AmpliPrep/Cobas TaqMan v2 apparatus, depending on the plasma volume available. A phylogenetic analysis was performed to determine HIV-1 subtypes. Genotypic resistance tests were performed at baseline and in case of virological failure, using the AC11-ANRS technique. Centralized HIV-1 western blot confirmation assays (BioRad, Marnes-la-Coquette, France) were performed on the frozen plasma sample collected at baseline.

Outcomes

The primary endpoint was the difference in HIV-1 DNA level/10⁶ PBMCs at W48 between the two groups, adjusted for baseline HIV-1 DNA level. The main virological secondary endpoint was the decrease in HIV-1 RNA level in each group. Other pre-specified endpoints were the proportion of patients with plasma HIV-1 RNA level <50 copies/mL, cumulative viraemia, and CD4+ and CD8+ T cell counts, at different timepoints from baseline.

The main safety endpoints were the incidence and severity of adverse events assessed using the ANRS severity scale and the incidence of disease progression and sexually transmitted infections (STIs). Another secondary endpoint was the self-reported treatment adherence (ANRS questionnaire) at W2, W8, W12, W36 and W48. Patients were asked if any tablets were missed in the last 4 days or if at least one tablet was missed in the last weekend, and if they followed their drug prescription in the last 4 weeks.

An unblinded independent data and safety monitoring board reviewed the data once a year.

Statistical analysis

Based on the data from the OPTIPRIM-ANRS 147 study in patients treated from PHI, we calculated that 96 patients were needed to achieve 80% power to detect a difference in HIV-1 DNA viral load of at least 0.25 log₁₀ copies/10⁶ PBMCs between the two groups at W48, with a standard error of 0.43, at a significance level of 5% using a two-sided test. This number was increased to 100 due to an expected dropout rate of 5%.

The primary endpoint was analysed in the modified ITT (mITT) population, defined as all randomized patients who initiated treatment and whose HIV-1 DNA level was measured at baseline and at W48. Only patients who never discontinued treatment for more than a week were included in the per-protocol analysis.

Demographics and clinical characteristics were summarized in each treatment group using the median and IQR for continuous variables, and the frequency and percentage for categorical variables. The Wilcoxon rank sum test was used to compare the distribution of HIV-1 DNA loads (log₁₀ copies/10⁶ PBMCs) at W48 and changes from baseline between the two groups. This test was also used to compare the distribution of levels and changes between baseline and W48 between both groups for CD4+ and CD8+ T cell counts and the CD4/CD8 ratios. The Fisher’s exact test was used to compare the proportion of patients achieving plasma HIV-1 RNA levels <50 copies/mL at W4, W8, W24 and W48. HIV-1 RNA levels measured by local centres were used when centralized data were missing. Ultrasensitive HIV-1 RNA levels allowed calculation of cumulative viraemia up to W48, defined as the area under the HIV-1 RNA level curve.¹³

R software version 3.6.2 was used for all analyses.

This study is registered under the number NCT02987530 on ClinicalTrials.gov.

Results

From 12 April 2017 to 22 August 2018, 101 patients (Figure 1) were included in 31 ANRS clinical sites in France; 51 were randomized to the dolutegravir group and 50 to the darunavir/cobicistat group. The maximum time to initiation of ART from baseline (randomization time) was 4 days, 54% of patients initiated cART on the day of randomization, and 36% on the following day.

Baseline characteristics

Baseline characteristics are presented in Table 1. The median (IQR) age was 36 (28–49)  years, 93% of patients were men, and most of them self-reported being MSM (69%). Symptomatic PHI was found in 90 patients (89%); the most common signs were asthenia (74%), fever (69%), skin rash (46%) and lymphadenopathy (Table S1, available as Supplementary data at JAC Online).

The HIV-1 western blot performed on the baseline samples showed ≤5 bands in 83% of patients. A Fiebig stage ≤3 was found in 14% of patients in the dolutegravir group and in 20% of patients in the darunavir/cobicistat group. Very early PHI (<2 bands detected by western blot) was diagnosed in 20% of patients in the dolutegravir group and in 32% of patients in the darunavir/cobicistat group. The median (IQR) baseline plasma HIV-1 RNA level was, respectively, 5.52 (4.88–6.30) and 6.07 (5.15–6.97) log₁₀ copies/mL. The median (IQR) CD4+ T cell count was, respectively, 442 (343–587) and 434 (305–643) cells/mm³. The median (IQR) HIV-1 DNA level was, respectively, 3.77 (3.42–4.06) and 3.96 (3.62–4.26) log₁₀ copies/10⁶ PBMCs.

The results of the genotypic test performed on the baseline samples showed antiretroviral resistance to dolutegravir (E138K or E138A) in two patients and to tenofovir (K70E) in one patient in the darunavir/cobicistat group, and to emtricitabine (M184V/I) in one patient in the dolutegravir group. The predominant HIV-1 subtypes were B (48%) and CRF02 (24%).

Follow-up data

The median (IQR) follow-up duration was 48 (48–49) weeks. Of the 101 patients, 97 completed the trial up to W48 (49 in the dolutegravir group and 48 in the darunavir/cobicistat group), 2 patients withdrew their consent (one at W2 in the darunavir/cobicistat group and one at W24 in the dolutegravir group), 1 patient died of cytomegalovirus pneumonia after the W4 visit (darunavir/cobicistat group) and 1 patient moved (after the W24 visit) to a city without an ANRS clinical site.

HIV-1 DNA levels in blood cells

At W48, the distribution of HIV-1 DNA levels significantly differed between the two groups, with a median (IQR) of 2.33 (2.01–2.72) log₁₀ copies/10⁶ PBMCs in the dolutegravir group and 2.54 (2.28–2.93) log₁₀ copies/10⁶ PBMCs in the darunavir/cobicistat group (P = 0.025) (Table 2). The baseline HIV-1 DNA level was strongly associated with the HIV-1 DNA level at W48 (Spearman r = 0.58, P < 0.001). After adjustment for the baseline HIV-1 DNA level in a multiple linear regression, the difference at W48 between the two groups was reduced to a mean of 0.12 log₁₀ copies/10⁶ PBMCs (P = 0.15). HIV-1 DNA levels decreased similarly in both groups up to W48 (Figure 2a). The median (IQR) decrease in HIV-1 DNA levels at W48 was −1.48 (−1.74 to −1.06) log₁₀ copies/10⁶ PBMCs in the dolutegravir group and −1.39 (−1.55 to −0.98) log₁₀ copies/10⁶ PBMCs in the darunavir/cobicistat group (P = 0.52).

The strongest decrease in HIV-1 DNA levels was observed during the first 3 months in both groups: at W12, the median (IQR) change from baseline was −0.86 (−1.17 to −0.54) log₁₀ copies/10⁶ PBMCs in the dolutegravir group and −0.94 (−1.15 to −0.59) log₁₀ copies/10⁶ PBMCs in the darunavir/cobicistat group (Figure 2b).

In the per-protocol analysis, the median (IQR) HIV-1 DNA levels at W48 did not significantly differ between the two groups: 2.32 (2.00–2.70) log₁₀ copies/10⁶ PBMCs in the dolutegravir group (46 patients) and 2.43 (2.24–2.76) log₁₀ copies/10⁶ PBMCs in the darunavir/cobicistat group (42 patients) (P = 0.12).

Plasma HIV-1 RNA levels

Plasma HIV-1 RNA levels decreased more rapidly in the dolutegravir group than in the darunavir/cobicistat group. From W4 to W12, the proportion of patients in the dolutegravir group with HIV-1 RNA levels <50 copies/mL was significantly higher than in the darunavir/cobicistat group: 24% versus 0% at W4, 55% versus 2% at W8, and 67% versus 17% at W12 (Figure 3a). This proportion remained higher in the dolutegravir group at W24, without reaching significance, while similar values were achieved at W48: 94% in the dolutegravir group and 90% in the darunavir/cobicistat group (P = 0.48). Using an ultrasensitive quantification method, a continuous decrease in HIV-1 RNA levels was observed, including between W24 and W48 (Figure 3b). Cumulated viraemia was significantly lower in the dolutegravir group than in the darunavir/cobicistat group (P < 0.001) (Table 2).

Only one patient discontinued treatment in the darunavir/cobicistat group at W24 due to suspected virological failure, with no emerging mutation resistance.

CD4+ and CD8+ T cell counts

The CD4+ T cell count mainly increased during the first 6 months in both groups (Table 2, Figure S1a). At W48, the median (IQR) gain from baseline was 208 (105–380) cells/mm³ in the dolutegravir group and 252 (149–362) cells/mm³ in the darunavir/cobicistat group. CD8+ T cell recovery followed the same dynamics in both groups. Nevertheless, the CD4+/CD8+ ratio was higher in the dolutegravir group up to W8, with a higher but non-significant (P = 0.19) proportion of patients rapidly reaching a CD4+/CD8+ ratio >1 at W4 (57% in the dolutegravir group versus 42% in the darunavir/cobicistat group) (Figure S1b).

Adverse events

The mean study treatment duration was similar for both groups: 46 (range: 2–55) weeks in the dolutegravir group and 43 (1–50) weeks in the darunavir/cobicistat group (P = 0.17). Thirteen patients (five in the dolutegravir group and eight in the darunavir/cobicistat group) discontinued study treatment during the follow-up.

Most adverse events were mild to moderate (grade 1 or 2). Grade 3 or 4 clinical adverse events were considered unrelated to cART and were less common in the dolutegravir group (7 events in 6 patients) than in the darunavir/cobicistat group (14 events in 12 patients) (Table S2).

Eighteen serious adverse events (SAEs) occurred and were coded into 24 events: 5 SAEs in the dolutegravir group (5 patients) and 13 SAEs in the darunavir/cobicistat group (12 patients) (Table S3). No SAEs were considered related to cART in the dolutegravir group and two were considered potentially related to cART by the investigator in the darunavir/cobicistat group: one case of myocardial infarction was considered a suspected unexpected serious adverse reaction and was reported to the competent authorities. One case of rhabdomyolysis with creatine phosphokinase levels >10 times the normal value was considered possibly related to darunavir/cobicistat therapy by both the investigator and the sponsor but was expected according to the SmPC (Summary of Product Characteristics). Five SAEs considered unrelated to treatment occurred in the dolutegravir group and 11 in the darunavir/cobicistat group, including the occurrence of fatal cytomegalovirus pneumonia in one patient, a suicide attempt, pulmonary TB and bladder carcinoma in situ. There was no difference in median (IQR) weight gain distribution between the two groups [2 (0–3.5) kg in the dolutegravir group and 2 (0–4.2) kg in the darunavir/cobicistat group; P = 0.59].

An STI (hepatitis C, syphilis, anal, pharyngeal or urethral gonorrhoea, or Chlamydia infection) was diagnosed in 21 patients at baseline (11 in the dolutegravir group, 10 in the darunavir/cobicistat group). During the follow-up, 43 STIs were diagnosed in 21 patients (12 in the dolutegravir group and 9 in the darunavir/cobicistat group) (Figure S2). The incidence of STIs was 20 cases per 100 person-years from W4 to W12, 40 cases per 100 person-years after W12 up to W24, and then 70 cases per 100 person-years until the end of the study.

The treatment adherence rate in the per-protocol population was similar for both groups.

Discussion

The OPTIPRIM2-ANRS 169 study was the first open-label study to compare once-daily treatment with 50 mg of dolutegravir with once-daily combined treatment with 800 mg of darunavir/150 mg of cobicistat, both given in combination with co-formulated tenofovir/emtricitabine initiated as soon as PHI was diagnosed. We assumed that cART based on dolutegravir, a potent second-generation INSTI, initiated during PHI, could rapidly suppress viral replication and thus reduce the HIV-1 blood reservoir size, after 48 weeks of treatment.

We chose to estimate the HIV reservoir size using the total cell-associated HIV-1 DNA level, which is a clinically relevant marker that correlates with the productive reservoir size measured in infectious units/million cells in effectively treated patients.¹² A marked decline in blood HIV reservoir size was found in both groups, with no difference in HIV-1 DNA levels at the study end after adjustment for the baseline HIV-1 DNA level. Of note, a strong decrease in HIV-1 DNA levels by −1.48 log₁₀ copies/10⁶ PBMCs was observed in the dolutegravir group at W48.

Indeed, the earlier that treatment is initiated during PHI, the greater and faster the impact on the reservoir is.^14–16 The reservoir is established early, gradually and continuously during PHI in CD4+ T cell subsets.⁶ Short-lived memory T cells (effector and transitional memory T cells) are the main infected cells during PHI compared with cells with a long half-life (naive and central memory T cells).⁴ Moreover, treatment initiated at the time of PHI could limit the establishment of persistent stable integrated HIV-1 DNA forms and protect naive and central memory T cells.^4,15,17,18 The decay of HIV-1 DNA during treatment initiated at the time of PHI has been characterized using a three-slope curve over three periods: 0–7, 8–32 and >32 months.¹⁴ The first period of rapid HIV-1 DNA decay up to W24 observed in our study could be mainly explained by the effect of cART on the clearance of activated and productive infected T cells with removal of the labile unintegrated HIV-1 DNA forms.^14,17 Thereafter, the effect of treatment was still observed, with a slight but persistent HIV-1 DNA level decrease up to W48, which could mainly be explained by the death of short-lived activated infected cells.^14,19 The continuous decrease in HIV-1 DNA levels up to W48 observed in both treatment groups, and previous results from the literature, suggest that treatment initiated at the time of PHI for more than 48 weeks could prolong this favourable effect on the HIV reservoir through the protection of long-lived cells.^14,15 Nevertheless, despite the virological power of dolutegravir to rapidly limit the production of viral active forms, we did not observe any difference in total HIV-1 DNA levels at W48 between the two treatment arms. Recently, Trémeaux et al.¹⁷ have reported that the proportion of very integrated HIV-1 DNA among total HIV-1 DNA remained very low within the 90 days of PHI in the absence of treatment. The total HIV-1 DNA was then essentially made up of labile forms (linear unintegrated and episomal forms). The low level of integrated HIV-1 DNA persisted during the first weeks of infection, probably because most activated cells carrying integrated HIV-1 DNA and producing new virions died.¹⁷ When treatment is initiated at this very early time of PHI, even if the inhibition of HIV-1 replication is quite slow with boosted PIs compared with INSTIs, the amount of integrated HIV-1 DNA is limited. This could explain why the more rapid inhibition of the viral replication observed with dolutegravir in our patients treated in the earliest stage of infection, i.e. at Fiebig stage 3, was not associated with a difference in total HIV-1 DNA levels at W48. This finding supports the very early initiation of treatment with a potent viral cART at the time of PHI to limit the establishment of persistent proviral DNA in quiescent cells.²⁰

Furthermore, we showed a strong and early effect of dolutegravir-based cART initiated at the time of PHI on viral replication. Indeed, despite a high HIV-1 replication level at baseline, the viral load was suppressed to <50 copies/mL in more than 50% of our patients after 12 weeks of treatment. This percentage was only observed after 24 weeks in the darunavir/cobicistat group, while similar and satisfactory treatment adherence was observed in both groups, and 88 out of the 97 patients (90%) were still treated at W48. This more rapid virological response, well demonstrated in chronic patients initiating dolutegravir-based cART,⁹ is in line with a recent large observational study²¹ and small studies assessing INSTIs initiated at the time of PHI.^18,22–25

The strong effect of dolutegravir on HIV-1 RNA replication, with a low cumulative viraemia, is interesting to reduce immune activation and inflammation levels and preserve the CD4+/CD8+ ratio.^13,18 Indeed, we showed a potential early benefit of dolutegravir-based cART compared with darunavir/cobicistat cART, with a trend towards better CD4+/CD8+ recovery during the first 8 weeks. This immunovirological impact is a clear advantage of dolutegravir-based cART to decrease severity of acute infection,³ restore immunity,⁵ reduce HIV-induced inflammation and limit neurological damage.^4,26

Our patients were included more rapidly than expected, in 16 months, showing that initiating treatment very early was well accepted, even at the sensitive time of the diagnosis of acute/recent infection and even within the framework of a clinical trial. Moreover, dolutegravir-based cART was well tolerated, with no treatment-related SAEs. In addition, the weight gain was similar to that observed in the darunavir/cobicistat group after 1 year of treatment, as recently reported in an observational study in 712 patients with PHI.²¹ PHI patients are a significant potential source of HIV transmission.²⁷ In this sense, the high incidence of STIs in our study should be highlighted, with 22% of patients experiencing at least one STI over a 48 week follow-up, in a study population mainly including MSM, suggesting that prevention messages should be reinforced in these patients. Therefore, a well-tolerated dolutegravir-based regimen that rapidly reduces plasma and thus genital viral loads²⁸ seems to strongly reduce transmission clusters at the time of PHI,^27,28 especially in this very sexually active population of MSM. These results of the randomized OPTIPRIM2 trial, assessing for the first time the benefit of an INSTI-based regimen at the time of PHI on the rapid control of the viral load, confirmed the approach recommended in the current guidelines, in particular the European AIDS Clinical Society (EACS) guidelines, which emphasize a potential benefit of selecting dolutegravir or bictegravir to more rapidly suppress the viral load.^7,8

In this trial, we observed a trend towards higher baseline HIV-1 DNA and RNA levels in the darunavir/cobicistat group compared with the dolutegravir group, especially in the subgroup of patients with 0–1 antibody detected by western blot at baseline. During acute HIV-1 infection, HIV-1 RNA and blood HIV-1 DNA levels vary widely from one patient to another, and with the time since infection.²⁹ We could not exclude a slight imbalance in the earliness of infection between the two groups. Given the potential impact on marker levels of even a slight imbalance in the timing of infection, further larger trials in patients with acute HIV-1 infection are needed to confirm our findings.

Conclusions

In this trial, one of the largest published to date on PHI, dolutegravir-based cART had a strong effect on HIV reservoir, although it was not significantly higher than that of darunavir-based cART. Moreover, the faster and earlier effect on initial viral replication is a clear advantage of dolutegravir during PHI, when viral replication is particularly high, with a major risk of HIV transmission. Among the well-tolerated potent INSTIs available, dolutegravir and probably bictegravir are major first-line options to be used at the time of PHI.

Acknowledgements

We thank the OPTIPRIM2-ANRS 169 study participants and investigators for their valuable contributions. We also thank: Delphine Lebrasseur-Longuet, Marine Saouzanet and Nicolas Leturque for monitoring the data; Arulvani Arulananthan, Carine Lascoux and Valérie Foubert for data management; Emmanuelle Netzer for quality assurance; Guillemette Antoni and Rémonie Seng for statistical support; Adeline Mélard, Marine Fillion and Elise Gardiennet for virological tests; and Ventzislava Petrov-Sanchez and Laura Nailler, from the Clinical Research Department of the ANRS | Emerging Infectious Diseases, and Sophie Pegorier for editing the manuscript. A.C., C.R., L.M., C.L., V.M., V.A.-F., R.B., C.G., Gilles Pialoux and J.R. were part of the OPTIPRIM 2-ANRS-169 scientific committee.

Funding

The study was funded and promoted by the ANRS. ViiV Healthcare and Janssen acted as cofunders through an ANRS contract. The funders had no role in the study design, data collection, data analysis, data interpretation or writing of the manuscript.

Transparency declarations

A.C. reports grants from Gilead, ViiV and Janssen. V.A.-F. reports grants from Gilead and ViiV and financial support to participate in conferences from Gilead, ViiV, Janssen, MSD and Roche. J.R. reports personal fees and financial support for participation in advisory boards, educational programmes and conferences from Gilead, Janssen, MSD, Pfizer, Theratechnologies and ViiV Healthcare. K.L. has received financial support for advisory boards, educational activities and travel grants from ViiV, MSD, GSK, Gilead and Janssen. S.A. reports financial support to participate in conferences from Gilead, ViiV and Janssen. P.M. reports grants from Gilead and MSD and financial support to participate in conferences from Gilead, ViiV, Janssen and MSD. R.C. received grant fees from Bristol-Myers Squibb and consulting fees from Janssen, ViiV Healthcare and Bristol-Myers Squib. F.A., C.R., P.L., C.L., L.M., C.G., R.B. and V.M. had no conflicts of interest to declare.

Author contributions

A.C. was the principal investigator and with L.M., C.R. and V.A.-F., the coinvestigators, designed the trial, and developed the protocol. P.L., C.R. and V.A.-F. performed the virological assessments. A.C., C.L., F.A., K.L., P.M., J.R., R.C. and S.A. enrolled the patients. L.M. and V.M. coordinated data collection and regulatory requirements, and performed, together with R.B., the data analysis. A.C., C.R., V.A.-F., R.B. and L.M. interpreted the data. R.B., V.M., L.M. and A.C. generated the tables and figures. A.C., C.R., R.B. and L.M. wrote the manuscript, and all authors reviewed, revised and approved the final manuscript.

Supplementary data

Tables S1 to S3 and Figures S1 and S2 are available as Supplementary data at JAC Online.

References

Author notes