ABSTRACT
The MONOD ANRS 12206 trial was designated to assess simplification of a successful lopinavir (LPV)-based antiretroviral treatment in HIV-infected children younger than 3 years of age using efavirenz (EFV; 25 mg/kg of body weight/day) to preserve the class of protease inhibitors for children in that age group. In this substudy, EFV concentrations were measured to check the consistency of an EFV dose of 25 mg/kg and to compare it with the 2016 FDA recommended dose. Fifty-two children underwent blood sampling for pharmacokinetic study at 6 months and 12 months after switching to EFV. We applied a Bayesian approach to derive EFV pharmacokinetic parameters using the nonlinear mixed-effect modeling (NONMEM) program. The proportion of midinterval concentrations 12 h after drug intake (C12 h) corresponding to the EFV therapeutic pharmacokinetic thresholds (1 to 4 mg/liter) was assessed according to different dose regimens (25 mg/kg in the MONOD study versus the 2016 FDA recommended dose). With both the 25 mg/kg/day dose and the 2016 FDA recommended EFV dose, simulations showed that the majority of C12 h values were within the therapeutic range (62.6% versus 62.8%). However, there were more children underexposed with the 2016 FDA recommended dose (11.6% versus 1.2%). Conversely, there were more concentrations above the threshold of toxicity with the 25 mg/kg dose (36.2% versus 25.6%), with C12 h values of up to 15 mg/liter. Only 1 of 52 children was switched back to LPV because of persistent sleeping disorders, but his C12 h value was within therapeutic ranges. A high EFV dose of 25 mg/kg per day in children under 3 years old achieved satisfactory therapeutic effective levels. However, the 2016 FDA recommended EFV dose appeared to provide more acceptable safe therapeutic profiles. (This study has been registered at ClinicalTrials.gov under identifier NCT01127204.)
INTRODUCTION
Efavirenz (EFV) is a nonnucleoside reverse transcriptase inhibitor approved for the treatment of HIV-1 infection in adults and pediatric patients (1–3). EFV was first approved in the United States, the European Union, and other countries in the late 1990s for children 3 years of age and above and weighing more than 10 kg (1–3) and in 2016 for children <3 years of age (4). The ANRS 12206 MONOD project was a randomized controlled trial conducted in West African children virologically suppressed after 12 months of combined antiretroviral therapy (cART) based on lopinavir (LPV)/ritonavir. Patients were randomized to receive either a simplified once-daily cART regimen based on EFV or to continue the twice-daily regimen based on lopinavir/ritonavir (5). At the start of the trial, no EFV dosing recommendations for children less than 3 years old were available. Therefore, the EFV dosing regimen was selected through simulations based on the results of a trial conducted in Burkina Faso which recommended 25 mg/kg of body weight/day for all children 2 to 4 years of age (6).
Sources of EFV pharmacokinetics (PK) variability have been studied in adults (7–11) and in children (12–16). These data have shown the influence of weight and genetics on the pharmacokinetics of EFV. The recent FDA recommendations for children who are 3 months to 3 years of age, released in 2016, take weight into account (Table 1) (4). Genetic covariates found include the polymorphic nature of cytochrome P450 2B6 (CYP2B6: rs3745274) and other isoforms responsible for EFV metabolism (17, 18). Department of Health and Human Services (DHHS) guidelines include a “research” dose of EFV for children aged 3 months to <3 years according to CYP2B6 genotype (using data generated in IMPAACT P1070: protocol P1070) (17) (Table 1). However, there are still significant gaps in our knowledge of the pharmacokinetics and pharmacodynamics in children (19). Target plasma concentrations of EFV in children are extrapolated to adults and range from 1 mg/liter to 4 mg/liter for the mean concentrations (20).
Recommended dosing of EFV in children, 2016
We developed a pharmacokinetic substudy appended to this project, the aims of which were (i) to estimate the middose EFV concentrations 12 h after drug intake (C12 h) in the cohort of HIV-infected children included in the ANRS 12206 MONOD trial and (ii) to compare the 25 mg/kg dose used in this trial to the FDA recommended dosing regimen regarding the therapeutic effective and safety levels.
RESULTS
Sample selection.Between May 2011 and January 2013, 156 children were initiated on cART at age 13.7 months (median) (21). After 12 to 15 months on cART, 13 infants had died, 2 were lost to follow-up, 3 were withdrawn from the treatment, 32 virologically failed, and 106 (68%) (54 in the LPV arm and 52 in the EFV arm) were randomized for participation in the study.
Among the 52 children randomized in the EFV arm, 49 (94%) (25 females and 24 males) had samples with 105 plasma EFV concentrations available for pharmacokinetic evaluation. The characteristics of those children are summarized in Table 2. The median (interquartile range [IQR]) age was 33.9 months (29.2 to 40.2), and the median body weight was 11.7 kg (10.9 to 12.9). The median (IQR) EFV dose was 24.4 mg/kg of body weight (23.3 to 25.5). A total of 8 (16%), 26 (53%), and 15 (31%) children had 1, 2, and 3 plasma samples available, respectively.
Summary of HIV-1-infected children characteristics included in the PK EFV substudy in pharmacokinetic evaluationsa
Bayesian approach.The observed concentrations matched satisfactorily the simulated concentrations (see Fig. 1), suggesting that the observed data were well described by the selected model. Thus, a maximum a posteriori probability (Bayesian estimation) was made from this model to predict individual pharmacokinetic parameters for EFV. Data with respect to values for Bayesian area under the concentration-time curve (AUC), apparent total body clearance of the drug from plasma (CL/F), and C12 h are presented in Table 3. Bayesian C12 h data are plotted in Fig. 2, where the CYP2B6 polymorphisms are distinguished.
Visual inspection (Visual Predictive Check): comparisons among the 5th (lower dashed line), 50th (solid line), and 95th (upper dashed line) percentiles obtained from the model described by Salem et al. (14) and the observed data (points) for the MONOD trial.
EFV concentrations 12 h after drug intake (derived from Bayesian estimated individual pharmacokinetic parameters) as a function of weight (n = 86)
EFV concentrations 12 h after drug intake (derived from Bayesian estimated individual pharmacokinetic parameters) as a function of weight. Crosses represent extensive metabolizers; circles represent slow metabolizers. The gray lines represent the C12 h thresholds in adults.
In our study, only one C12 h value (0.95 mg/liter) was below the efficiency target (1 mg/liter). Most (61%) of the C12 h values were within the therapeutic range of 1 to 4 mg/liter. However, the C12 h values for 88% of the slow metabolizers (CYP2B6-516-T/T) were above the limit of 4 mg/liter.
We did not notice any clinical seizures in terms of central nervous system (CNS) toxicity among the 52 children switched onto EFV, but 1 child had persistent daytime and nighttime sleeping disorders. EFV treatment of this child was then substituted after 9 months of treatment (18). At the PK evaluation visits, this child was not overexposed with regard to EFV because his C12 h value was estimated at 1.9 mg/liter with the 25 mg/kg EFV dose and at 1.4 mg/liter with the 2016 FDA recommended dose.
Simulations and drug dosage evaluations. (i) Simulations and results that would have been obtained if the children of the MONOD trial had received the FDA recommended doses.Whichever the dosing scheme, i.e., the 25 mg/kg dose or the 2016 FDA recommended EFV dose (200 mg for 7.5 kg to 15 kg), simulations showed that the majority of C12 h values were within the therapeutic range for both dosages (62.6% or 62.8%, respectively) (Fig. 3). However, there were more children underexposed to EFV with 2016 FDA recommended dose than with the 25 mg/kg dose (11.6% versus 1.2%, respectively). Conversely, there were more concentrations above the threshold of toxicity with the dose of 25 mg/kg (36.2% versus 25.6%). Among the concentrations above the toxicity threshold, the majority were those seen with slow metabolizers. Among the 12% of patients whose results were below the target efficiency value for the 2016 FDA recommended EFV dose, the median [range] C12 h value was 0.83 mg/liter [0.76 to 0.88].
Simulations of EFV C12 h Bayesian concentrations as a function of weight for the 2016 FDA recommended dose (200-mg dose regimen for children weighing 7.5 kg to 15 kg) (left) and for the dose regimen (25 mg/kg) (right). Crosses represent extensive metabolizers; circles represent slow metabolizers. The gray lines represent the C12 h thresholds in adults.
(ii) Simulations and results that would have been obtained if children of the MONOD trial had received the protocol P1070 recommended dose based on genotyping.Simulations of C12 h values that would have been obtained if the children of the MONOD trial had received the doses recommended by the P1070 protocol based on the genotype are presented in Fig. 4B. The doses for slow metabolizers would reduce their C12 h values to within or close to the therapeutic range. However, for extensive metabolizers, the recommended dose with regard to CYP2B6 genotype would lead to overexposure. Simulations of a 300-mg EFV dose regimen in all extensive metabolizers allow having the majority of C12 h values in the therapeutic range (Fig. 4C). With the recommended P1070 dose for extensive metabolizers, 66% of C12 h values are in the EFV adult target range, whereas 75% of C12 h values are in the therapeutic range with a 300-mg EFV dose regimen.
Simulations of EFV C12 h Bayesian concentrations as a function of weight. (A) Without regard to CYP2B6 genotyping (FDA recommendations): 200 mg for children weighing 7.5 kg to 15 kg. (B) With regard to CYP2B6 genotyping (protocol P1070 research dose): for extensive metabolizers, 400 mg for children weighing 7 kg to 14 kg and 500 mg for children weighing 14 to 15 kg; for slow metabolizers, 100 mg for children weighing 7 kg to 14 kg and 150 mg for children weighing 14 to 15 kg. (C) Simulation of a more appropriate dose for extensive metabolizers: for extensive metabolizers, 300 mg for children weighing 7 kg to 15 kg; for slow metabolizers, 100 mg for children weighing 7 kg to 14 kg and 150 mg for children weighing 14 to 15 kg. Crosses represent extensive metabolizers; circles represent slow metabolizers. The gray lines represent the C12 h thresholds in adults.
DISCUSSION
The MONOD ANRS 12206 trial assessed a simplified once-daily cART based on EFV versus a twice-daily regimen based on lopinavir/ritonavir in children between 2 and 4 years of age in West Africa. In this subanalysis, we studied EFV pharmacokinetics to assess the appropriateness of the MONOD ANRS 12206 drug dosages (25 mg/kg of body weight/day).
For the study analysis, a Bayesian approach was preferred to the development of a new pharmacokinetic model given the robust EFV population pharmacokinetic models previously published in similar populations. The model selected was developed in 96 HIV-infected children (a total of 3,172 observations) who participated in Pediatric AIDS Clinical Trials Group 382 (14). This model took into account the developmental changes that occur during childhood by inclusion of weight- and age-related effects on the apparent clearance and volume of distribution parameters. This model includes also the CYP2B6-G516T polymorphism, which influences oral clearance. In addition, evaluation of EFV PK parameters with our data provided estimates similar to those reported by Salem et al. (14). We used the therapeutic pharmacokinetic thresholds previously associated with EFV virological efficacy and toxicity in adults (20) to evaluate the protocol drug dosages as well as the 2016 FDA dosing guidelines for HIV-infected children. This target, established by Marzolini et al. (20), relates to the concentrations in the middosing interval. Indeed, it was established on the basis of the results of a study on samples taken between 8 and 20 h after dose intake and is difficult to apply in clinical practices. This target was subsequently taken up as a target for the 12-h time point (22, 23).
The dose of 25 mg/kg was chosen because there were no dosing regimen recommendations in children less than 3 years old (4). The allometric scale described in the model of Salem et al. supports a weight-based dosing recommendation. The dose prescribed by the FDA indicates dosage in milligrams, but these recommendations are, notwithstanding, based on weight because it is provided for different weight bands. Here, we evaluate a restricted class of weights ranging from 8 to 15 kg. According to simulations, in this weight band, the 2016 FDA recommended dose showed a satisfactory percentage of C12 h values in the therapeutic range given the high between-subject variability in the EFV PK.
The simulations showed that the 25 mg/kg dose and the 2016 FDA recommended dose provided similar rates of C12 h values within the therapeutic range but that more children were underexposed with the FDA recommended dose. Conversely, there were more concentrations above the toxicity threshold using the dose of 25 mg/kg. In the MONOD trial, EFV neurological toxicity was very difficult to assess. Only one neurological toxicity event led to treatment interruption. These data may provide useful insights into the current knowledge of EFV PK and toxicity. With the 25 mg/kg/day dose regimen, only one concentration was lower than the efficacy threshold. With the 2016 FDA recommended dose, there was a higher rate of C12 h values lower than 1 mg/liter but all of those C12 h values were greater than 0.75 mg/liter. Whatever the dose, almost all of the slow metabolizers had C12 h values greater than 4 mg/liter. The 2016 FDA recommended dose seems to be a good compromise to avoid overexposure in genotype TT carriers for CYP2B6-G516T polymorphisms without a high risk of underexposure. A supplemental protocol P1070 recommendation based on genotype is also available but appeared very difficult to apply because genotyping is not a common practice, especially in Africa. However, this supplemental dose recommendation would help return doses to the therapeutic range for children who are slow metabolizers but would overexpose a part of the population of extensive metabolizers. A dose of 300 mg would be more appropriate in extensive metabolizers according to our simulations.
EFV is a good candidate for therapeutic drug monitoring (TDM) (24–27). In fact, plasma concentrations are characterized by high interindividual variability. Low concentrations have been linked with viral nonsuppression and high concentrations with toxicity. TDM remains an important tool to detect patients in treatment failure or slow metabolizers, who are at risk of toxicity. In African countries, where TDM is not routinely feasible, methods need to be implemented to prevent the risks associated with the use of efavirenz. Thus, Van de Wijer et al. suggested that, to counter underdiagnosis of neuropsychiatric and neurodevelopmental side effects, screening must be made a high priority in daily practice for children treated with efavirenz and it must be closely linked to adequate care (28).
In conclusion, with both the 25 mg/kg EFV dose used in the MONOD trial and the 2016 FDA recommended EFV dose, the objective was achieved in terms of efficacy and toxicity. The dosage recommended by the FDA could be more appropriate because it causes less overexposure. In addition, the FDA recommendations seem easier to implement.
MATERIALS AND METHODS
Study design.The MONOD-ANRS-12206 project is an international, noninferiority, open-label phase 3 randomized clinical trial conducted after an initial 12 months of cART in a therapeutic cohort in Ouagadougou, Burkina Faso, and in Abidjan, Côte d'Ivoire (ClinicalTrials registration no. NCT01127204) (18). The protocol was approved by the Comité d'Éthique pour la Recherche en Santé du Burkina Faso and by the Comité National d'Éthique et de la Recherche en Côte d'Ivoire. Children were included in an initial prospective cohort that received a lopinavir-based triple therapy twice daily for 12 months. Children with confirmed viral suppression (HIV-1 RNA, <500 copies/ml) after 12 months on cART were randomized at 13 months into two parallel arms, with one arm maintaining the lopinavir-based therapy and one arm switching to a EFV strategy of once-daily administration until the time point of 25 months was reached. Children received EFV in soluble form at a dose of 25 mg/kg of body weight once daily.
Study procedures and analytical methods.Patient data were followed up prospectively from inclusion (at cART initiation) with monthly visits until 25 months. Demographic and clinical variables, including age, sex, body weight, time of administration, and time of sampling, were recorded. Pharmacokinetic samples were taken at 6 months and 12 months postrandomization in the EFV arm. For each sampling occasion, one or two samples per child were taken.
EFV plasma concentrations were measured by the use of a previously published method of high-performance liquid chromatography (HPLC) performed with a UV detector (Waters, Barcelona, Spain) (19).
Pharmacogenetics studies.Genomic DNA was isolated from saliva collected with ORAgen kit swabs (DNA Genotek, Ottawa, Canada), using a PrepIT DNA extraction kit according to the instructions of the manufacturer (DNA Genotek, Ottawa, Canada). Briefly, cells were heated 1 h at 50°C. Proteins were then precipitated with the precipitating solution on ice and centrifuged. Supernatants were harvested, and DNA was precipitated by adding 100% ethanol followed by centrifugation. DNA pellets were washed with 70% ethanol and then dried and were finally dissolved in hydration solution and quantified using a NanoVue spectrophotometer (Nano Drop Technologies, USA).
CYP2B6 G516T (rs3745274) was genotyped using TaqMan genotyping assays (Applied Biosystems, CA, USA) and an ABI 7500 system (Applied Biosystems, Foster City, CA). Quantitative PCR (qPCR) analyses were performed in a 25-μl volume with 20 ng of DNA using TaqMan Universal master mix II. The thermal cycling comprised 50 cycles of 15 s at 92°C and 1 min at 60°C. Analysis was performed using Sequence Detector software (SDSv2.0; Applied Biosystems, CA, USA).
Population pharmacokinetic analysis and modeling.Nonlinear mixed-effect modeling (i.e., NONMEM) was used to compute Bayesian estimates of individual subject parameters using the patient data and assuming previously reported population PK model and parameter values. Population parameters were not reestimated. The first-order conditional estimation with interactions method was used in the estimation step, with the maximum number of function evaluations set to 0. EFV individual predicted concentrations (IPREDs) were derived from the model published by Salem et al. (14). This model was established using a population of children comparable to patients included in the MONOD-ANRS-12206 trial. The authors reported that a one-compartment model, with first-order absorption and elimination rate constants, that used weight-based allometric scaling for oral clearance and apparent volume of distribution adequately described their data. A sigmoid maximum effect (Emax) maturation model demonstrated an increase in oral clearance with age that reached 90% of the mature level by the age of 9 months. Furthermore, the CYP2B6-G516T polymorphism decreased oral clearance. Children with the CYP2B6-516-TT genotype were found to have a CL/F value 51% lower than that shown by patients carrying the GT or TT genotype following a recessive genetic model. An external evaluation of the model published by Salem et al. (14) was first performed using our data. Concentrations measured in the MONOD trial were superimposed on the 5th, 50th, and 95th percentiles of the simulated concentrations obtained with the model of Salem et al. A visual inspection was then performed. Thus, a maximum a posteriori probability (Bayesian estimation) result was obtained from this model to predict individual pharmacokinetic parameters for EFV. IPREDs were determined for each patient in the MONOD data set for the available sampling times, given dosage history, age, body weight, and CYP2B6-G516T covariate values. The middose concentration (C12 h) was derived for each patient.
During the trial follow-up period, CNS toxicity was routinely monitored clinically, looking monthly for sleeping disorders and seizures.
Assessment of MONOD and FDA drug dosages.To assess results of comparisons between regimens, we calculated the percentage of children with (i) EFV middose concentrations (C12 h) below 1 mg/liter and (ii) EFV middose concentrations above 4 mg/liter. These thresholds have been shown to be associated with EFV virological efficacy (7, 8, 11, 20, 29) or EFV toxicity (20, 30) in HIV-infected adults. The individual pharmacokinetic parameters from all participants were used to simulate the 2016 FDA dosing guidelines without regard to CYP2B6 genotype and to simulate the protocol P1070 dose of EFV according to the CYP2B6 genotype (Table 1).
ACKNOWLEDGMENTS
The content is solely the responsibility of the authors and does not necessarily represent the official views of the French INSERM-ANRS, EDCTP, or University of Bordeaux.
The participating sites and members of the ANRS 12206 MONOD Collaboration Study Group (as of 7 July 2015) in Ouagadougou, Burkina Faso, were as follows: at Centre de Recherche International pour la Santé, Malik Coulibaly, Désiré Lucien Dahourou, Nicolas Meda (coinvestigator), Colette Ouédraogo, Mamadou Sawadogo, Wilfried Somé, Désiré Sondo, and Elisabeth Thio; at CHU Charles De Gaulle, Mamadou Barry, William Hiembo, Fla Kouéta, Adama Ouattara, Moussa Ouédraogo, Rasmata Ouédraogo, Sylvie Ouédraogo, Bernadette Congo, Rose Barry, and Diarra Yé; at CHU Yalgado Ouédraogo, Malika Congo, Edouard Minéné, Marie Coulibaly, Pierre Innocent Guissou Angèle Kalmogho, Ludovic Kam, Emile Ouédraogo, Lassana Sangaré, and Caroline Yonaba; at Programme Sectoriel Santé de Lutte contre le SIDA et les IST, Sylvestre Tiendrebeogo; at Programme d'Appui au Monde Associatif et Communautaire (PAMAC), Odette Ky-Zerbo.
Those in Abidjan, Côte d'Ivoire, were as follows: at Programme PACCI, Xavier Anglaret, Clarisse Amani-Bossé, Divine Avit, Christine Danel, Serge Eholié, Didier Ekouévi, Eulalie Kanga, Suzanne Kouadio, Séverin Lennaud, Maxime Aimé Oga, and Thérèse N′Dri-Yoman; at CHU Cocody, Madeleine Amorissani-Folquet, Evelyne Dainguy, Beugre Kouassi, Jean-Claude Kouassi, and Gladys Oka; at CHU Yopougon, Kader Keita, Jean Yves Lambin, François Eboua Tanoh, and Marguerite Timité-Konan (coinvestigator); at Site Abobo-Avocatier, Véronique Mea-Assande; at Site CePReF-enfants, Addi Edmond Aka, Hortense Aka-Dago, Sylvie N′Gbeche, and Eugène Messou; at Laboratory CeDReS, Arlette Emieme, Fatoumata Koné, Hervé Menan, Thomas Toni, and Vincent Yapo; at Programme National de Prise en Charge, Kouamé Abo, Irma Ahoba, and David Aka; at FSU Abobo-Avocatier, Gbaméné Kouassi; at Pharmacie de la Santé Publique, Carine Kodo; implementers, Touré Siaka, Pety Touré (ACONDA), Fassinou Ekouevi (EGPAF), Ida Viho (ICAP), Anthony Richard Tanoh, and Olivier Blé (Fondation ARIEL GLASER); community representatives, Yaya Coulibaly (RIP+) and Philomène Takouo (ONG Bayema); Programme ESTHER, Jean Marie Massumbuko; CIRBA, Kouadio Kouakou; Programme National de Santé Infantile, Dorothée Koumi; Programme Elargi de Vaccination, Berté Koné.
The members at the Methodology and Data Management Center, Inserm U897, Institut de Santé Publique, d'Épidémiologie et de Développement, University of Bordeaux, France, were as follows: Sophie Dattez, Sophie Desmonde, Julie Jesson, Sophie Karcher, Jérôme Le Carrou, Valériane Leroy (coordinating investigator), Karen Malateste, Camille Ndondoki, and Pierre Touret; methodological support, Caroline Bouyssou, Geneviève Chêne, Valérie Conte, Delphine Gabillard, Valérie Journot, and Roger Salamon.
The website for MEREVA, Bordeaux, France, is at http://mereva.isped.u-bordeaux2.fr/monod/Accueil.aspx .
The participating sites and members of the supporting teams are as follows: at Luxembourg Institute of Health, Luxemburg, Vic Arendt (coinvestigator), Carole Devaux, and Jean-Claude Schmit; at Hôpital Universitaire Des Enfants Reine Fabiola, Brussels, Belgium, Philippe Lepage (coinvestigator); at Hôpital Necker-Enfants Malades Assistance Publique-Hopitaux de Paris and EA8, Paris-Descartes, Stéphane Blanche (coinvestigator), Deborah Hirt, Christine Rouzioux, Claire Pressiat, Jean-Marc Treluyer, and Saik Urien; at Commissariat à l'Energie Atomique, Alain Pruvost (CEA); at Laboratoire de virologie, Hopital Saint-Louis, Marie-Laure Chaix-Baudier.
The participant at UMR 1058 “Pathogenesis and Control of Chronic Infections” INSERM—Université Montpellier—EFS, Montpellier, France, is as follows: Philippe Van de Perre (coinvestigator).
The members (locations) of the administrative team are as follows: Elodie Vernoux (Bordeaux, France), Aminata Paré-Karambiri (Ouagadougou, Burkina Faso), Zouma Tinto (Ouagadougou, Burkina Faso), Adoulaye Cisse (Abidjan, Côte d'Ivoire), and Madikona Dosso (Abidjan, Côte d'Ivoire).
The members (locations) of the MONOD ANRS 12206 Scientific Steering Committee are as follows: Roger Salamon (Chair) (Bordeaux, France), Valériane Leroy (coordinating investigator) (Bordeaux, France), Nicolas Meda (coinvestigator) (Ouagadougou, Burkina Faso), Marguerite Timite-Konan (coinvestigator) (Abidjan, Côte d'Ivoire), Vic Arendt (Co-Investigateur, Luxembourg), Stéphane Blanche (coinvestigator) (Paris, France), Philippe Lepage (coinvestigator) (Brussels, Belgium), Philippe Van de Perre (coinvestigator) (Montpellier, France), François Dabis (Bordeaux, France), and Jean-Claude Schmit (CRP-Santé, Luxembourg).
The names (locations) of those who attended the MONOD ANRS 12206 trial independent data monitoring committee meeting are as follows: Dominique Costagliola (Chair) (Paris, France), Mark Cotton (Cape Town, South Africa), Carlo Giaquito (Bologna, Italy), Diana Gibb (London, United Kingdom), and Elisabeth Menu (Paris, France).
The names of those who are associated with the promoter, Inserm-ANRS, France, are as follows: Jean-François Delfraissy (director), Brigitte Bazin, Marie de Solère, and Claire Rekacewicz.
FOOTNOTES
- Received 10 February 2017.
- Returned for modification 22 March 2017.
- Accepted 17 April 2017.
- Accepted manuscript posted online 8 May 2017.
- Copyright © 2017 American Society for Microbiology.
