First report of transmission of a highly resistant strain of HIV-1 group O

Antiretroviral (ARV) drug resistance is associated with HIV virological escape. WHO recommends surveillance of resistance mutations in ARV-naïve patients to estimate the prevalence and rates of viruses with transmitted drug resistance evolution (TDR) [1]. Prevalence studies on TDR among pandemic HIV-1 group M (HIV-1/M) are common [1–4]; more recently, work focusing on HIV-2 has shown a 5% prevalence of TDR [5]. No data are available on TDR for the divergent HIV-1 group O (HIV-1/O). HIV-1/O infection is endemic in Cameroon [6,7], and found sporadically outside [6,8–11]. In France, the RES-O network that monitors the dynamics of this infection [12] has identified 143 patients since the first case was reported in 1992. As a result of treatment accessibility in Cameroon since 2007, the numerous drugs available in France, and the strong links between these two countries, we can hypothesize that HIV-1/O resistant viruses may circulate in these countries and be transmitted, as previously described for HIV-1/M and HIV-2 [4,5]. Indeed, in Cameroon, first-line combined antiretroviral therapy (cART) contained two nucleosidic reverse transcriptase inhibitor (NRTI) and one non-NRTI (NNRTI). As HIV-1/Os are considered naturally resistant to NNRTI, those patients treated were thereby exposed to two effective drug therapies, which could favour the emergence of resistant viruses. As a result of more limited therapeutic options for HIV-1/O [11], it is essential to identify such viruses for appropriate first-line cART management. The aim of our study was to perform a molecular survey of viruses with TDR in ARV-naive HIV-1/O-infected patients living in France. Resistance sequences for protease and reverse transcriptase regions were obtained, as previously described [12], for 41 patients HIV-1/O positive before cART initiation. According to the lists of mutations described for HIV-1/M by the Agence nationale de Recherche sur le SIDA et les hépatites virales (ANRS), the International Antiviral Society (IAS)-USA, the Stanford database [13,14], and the WHO [15], and to HIV-1/O and HIV-1/M natural polymorphisms previously described [12], we considered a list of six Transmitted Resistance Mutations (TRM) in resistance sequences for protease and four in reverse transcriptase. They were found in three resistance sequences for protease (7.1%) and in only one reverse transcriptase (2.4%) sequence and corresponded to three patients, leading to a TRM prevalence of 7.3% (confidence interval 95%: 2.72–16.5). Two of these patients had viruses with a single mutation, F53Y or A71I, in resistance sequences for protease, and none in reverse transcriptase. The other TRM (resistance mutations for protease: M46V, I54V, I84V, L90M, and reverse transcriptase: D67N, T69N, G190A, K219Q) corresponded to a unique sequence in one patient (RBF218); the complete resistance profile of the strain led to resistance or partial resistance to multiple drugs (Fig. 1a). This man was married to RBF132, who had received 11 different ARV lines before her husband was diagnosed HIV-1/O positive in 08/2010. Her different drug regimens, associated with uncontrolled replication, led to sequential selection of many drug mutations, objectified by resistance genotyping (Fig. 1a and b). Analysis of her sequential genotyping data available before and after the RBF218's HIV-1/O diagnosis, and of phylogenetic relationships between strains of both patients, confirmed transmission of resistant virus to her partner (Fig. 1a–c). As the pattern was identical from April 2004 to August 2007 (data not shown), the resistant strain could have been transmitted between 2004 and 2008, consistent with the high level of replication during this period (Fig. 1b). Presence of numerous mixed populations in RBF132's sample in August 2010 (Fig. 1a), and evolution to wild virus, in the last sample of 2012 (Fig. 1a) are in favour of this hypothesis. At time of diagnosis, RBF218 was classified as Center for Disease Control C AIDS clinical category and had a CD4+ cell count of 415/μl; his virus was highly replicative (4.8 log10 copies/ml with RealTime HIV-1 assay, Abbott), validating the transmissibility potential of even highly mutated HIV-1/O-resistant viruses. Therapeutic management of the recipient patient was complicated because of resistance to different classes. Treatment was initiated immediately in September 2010 with lamivudine + abacavir + darunavir/ritonavir, but the patient left for Cameroon soon after and died in February 2011 of unknown reasons. The question arose of a higher virulence of the strain, but was not possible to prove because of short follow-up.Fig. 1: Genotypic, therapeutic, and phylogenetic data of patient RBF132 and patient RBF218.(a) Transmitted resistance mutations observed in RBF218 sample in August 2010 and mutations found in RBF132 samples in January 2001, September 2001, January 2003, April 2004, September 2007, August 2010, and March 2012; these mutations were defined according to the lists described for HIV-1/M by the ANRS (http://www.hivfrenchresistance.org/2015/Algo-sep-2015.pdf), the IAS-USA (https://www.iasusa.org/tam/article/update-drug-resistance-mutations-hiv-1-march-2013), the Stanford database [13,14], and the WHO [15]. According to the ANRS and Stanford resistance algorithms defined in 2015, the RB218's resistance profile led to resistance or partial resistance against atazanavir, darunavir, fosamprenavir, indinavir, lopinavir, nelfinavir, saquinavir, tipranavir, didanosine, stavudine, zidovudine, efavirenz, etravirine, nevirapine, and rilpivirine. (b) The different ART regimens of patient RBF132 received during her follow-up are represented according to time. The stars represent the date of genotypic resistance tests for the two patients. Viral loads were obtained from samples collected in November 1997, January 2001, December 2001, October 2002, June 2004, January 2006, April 2007, September 2007, June 2008, November 2008, February 2010, August 2010, March 2011, March 2012, and February 2013. (c) Maximum likelihood tree representing the phylogenetic relationships between sequences available for RBF132 and RBF218. The phylogenetic relationships between 39 HIV-1/O pol protease and partial RT sequences (987 nucleotides) and sequences (in bold) obtained in RBF132 samples (January 2001, September 2001, January 2003, April 2004, September 2007, August 2010, and March 2012) and in RBF218 sample (August 2010) were investigated by inferring a maximum likelihood tree with a GTR + Γ + I substitution model using MEGA5 [18]. In total, 500 bootstrap replicates were performed to assess the reliability of each node; bootstrap values are shown when more than 70%. AZT, zidovudine; DDI, didanosine; DDC, zalcitabine; 3TC, lamivudine; D4T, stavudine; RTV, ritonavir; IDV, indinavir; EFV, efavirenz; TDF, tenofovir; LPV/r, lopinavir/ritonavir; FTC, emtricitabine; DRV/r, darunavir/ritonavir; RAL, raltegravir; T20, enfuvirtide).In conclusion, our data showed a prevalence of TRM of 7.3% among the HIV-1/O-positive patients in France; but the very small population here leads to a difficult comparison with data found for HIV-1/M or HIV-2. Among them, we found a treatment-naïve patient presenting a multidrug-resistant strain as described for HIV-1/M and HIV-2 [16,17]. These data support the need of performing pre-cART resistance analysis in HIV-1/O-infected patients, especially in patients whose partner is treated with cART. Acknowledgements We wish to thank all the staff of the Laboratory of Virology at Rouen University Hospital, and Professor Diane Descamps for her helpful insights and comments on this manuscript. We are indebted to all the physicians and biologists involved in the identification and surveillance of HIV-1/O infections for the RES-O network. We are grateful to Nikki Sabourin-Gibbs, Rouen University Hospital, for writing assistance and review of the manuscript in English. Clinical investigators and virological investigators of the French RES-O: Dr J. Chennebault, Dr P. Fialaire, and Dr H. Le Guillou-Guillemette (Angers) Dr J. Gaillat and Dr B. Chanzy (Annecy) Dr P Genet, Dr V Daneluzzi, and Dr L. Courdavault (Argenteuil/Nanterre) Dr R. Fior (Béclère) Dr C. Chirouze and Dr D. Bettinger (Besançon) Professor P. Yeni, Dr R. Landman, Dr U. Colasante, Dr X. Duval, Dr C. Rioux, and Dr F. Damond (Bichat) Dr B. Masquelier (Bordeaux) Dr C. Dupont (Boulogne-Billancourt) Dr G. Otterbein (Bry-sur-Marne) Dr C. Henquell (Clermont-Ferrand) Dr F. Cordonnier, Dr M. Bloch, and Dr H. Ichou (Colombes) Professor Y. Levy, Professor J-D Lelievre, Dr A-S Lascaux-Cametez, and Dr G. Melica-Gregoire, Dr S. Dominguez, Professor J-M Pawlotsky, and Dr M. Bouvier-Alias (Créteil) Dr A. Waldner and Professor P. Pothier (Dijon) Dr J-P. Pathe and Dr B. Olivier (Evreux) Dr D. Zucman, Dr C. Majerholc, and Professor D. Vignon (Foch) Dr M. Wirden (La Pitié Salpetrière) Dr J. Ceccaldi (Libourne) Dr T. Lambolez and Dr P. Barbut (Longjumeau) Dr I. Poizot-Martin and Dr C. Tamalet (Marseille) Dr M. Echard (Montfermeil) Dr L. Rafenne and Dr F-X. Huchet (Montsouris) Dr G. Beck-Wirth and Dr J-M. Delarbre (Mulhouse) Professor V. Ferre and Dr E. Garnier (Nantes) Dr J-M. Descamps, Dr P. Deleplanque (Niort) Dr G. Le Moal and Dr G. Giraudeau (Poitiers) Dr P. Castiel and Dr A. SI-Mohamed (Pompidou), Dr P. Perfezou and Dr I. Dorval (Quimper), Dr J-L Berger, Dr C Rouger, Dr C Strady, and Dr V. Brodard (Reims) Dr J-M. Chappelain, Dr A. Maillard, and Dr A. Ruffault (Rennes) Dr A. Depatureaux and Dr I. Gueit (Rouen), Dr J. Pavie, Professor F. Simon, and Dr C. Delaugerre (Saint Louis), Dr J. Deleuze and Professor F. Rozenberg (Cochin-Saint Vincent de Paul), Dr Y. Leveneur and Dr M-F. Danjoux (Tarbes), Dr F. Bani-Sadr and Professor J-C. Nicolas (Tenon) Dr M-D. Tabone (Trousseau) Dr S. Roussin-Bretagne and Dr M Harzic (Versailles) Dr J-D. Poveda (Laboratoire Cerba) We also thank the CHU de Rouen and the Institut de Veille Sanitaire (InVs) for providing financial support. Conflicts of interest There are no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.