Secondary Logo

Institutional members access full text with Ovid®

Caution is needed in interpreting HIV transmission chains by ultradeep sequencing

Todesco, Evea; Wirden, Marca; Calin, Ruxandrab; Simon, Annec; Sayon, Sophiea; Barin, Francisd; Katlama, Christineb; Calvez, Vincenta; Marcelin, Anne-Genevièvea; Hué, Stéphanee

doi: 10.1097/QAD.0000000000002105
Epidemiology and Social

Objectives: Molecular epidemiology is applied to various aspects of HIV transmission analyses. With ultradeep sequencing (UDS), in-depth characterization of transmission episodes involving minority variants is permitted. We explored HIV-1 epidemiological linkage and evaluated characteristics of transmission dynamics and transmitted drug resistance (TDR) detection through the added value of UDS.

Design: HIV pol gene fragments were sequenced by UDS and Sanger sequencing on samples of 70 HIV-1-infected, treatment-naive recently diagnosed MSM.

Methods: Pairwise genetic distances and maximum likelihood phylogenies were computed. Transmission events were identified as clades with branch support at least 70% and intraclade genetic difference less than 4.5%. TDR mutations were recognized from the TDR consensus list. Transmission directionality, directness and inoculum size were inferred from tree topologies.

Results: Both datasets concurred in the identification of seven transmission pairs and one cluster of three patients. With UDS, direction of transmission was inferred in four out of eight chains. Evidence for multiple founder viruses was found in two out of eight chains. No transmission of minority-resistant variants was evidenced. TDR mutations prevalence in protease and reverse transcriptase fragments was 4.3% with Sanger sequencing and 18.6% with UDS.

Conclusion: Although Sanger sequencing and UDS identified the same transmission chains, UDS provided additional information on founder viruses, direction of transmission and levels of TDR. Nevertheless, topology of clusters was not always consistent across gene fragments, calling for a cautious interpretation of the data. Moreover, unobserved intermediary links cannot be excluded. Phylogenetic analysis use as a forensic technique for HIV transmission investigations is risky.

aSorbonne Université, INSERM, Institut Pierre Louis d’Epidémiologie et de Santé Publique (iPLESP), AP-HP, Hôpital Pitié-Salpêtrière, Laboratoire de virologie, Paris

bSorbonne Université, INSERM, Institut Pierre Louis d’Epidémiologie et de Santé Publique (iPLESP), AP-HP, Hôpital Pitié-Salpêtrière, Department of Infectious Diseases, Paris

cDepartment of Internal Medicine, Hôpital Pitié-Salpêtrière, AP-HP, Paris, France

dCentre National de Référence du VIH, laboratoire de Virologie, CHRU de Tours, Inserm U1259, Université François Rabelais, Tours, France

eDepartment of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, UK.

Correspondence to Eve Todesco, Department of Virology, Bât CERVI, Hôpital Pitié-Salpêtrière, 83 Bd de l’Hôpital, Paris 75013, France. Tel: +33 1 42177401; fax: +33 1 42177411; e-mail:

Received 30 September, 2018

Accepted 12 November, 2018

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

Copyright © 2019 Wolters Kluwer Health, Inc.