Prevalence of HIV-1 drug resistance mutations in proviral DNA in the Swiss HIV Cohort Study, a retrospective study from 1995 to 2018
Bashkim Jaha, Corinne D Schenkel, Lisa Jörimann, Michael Huber, Maryam Zaheri, Kathrin Neumann, Christine Leemann, Alexandra Calmy, Matthias Cavassini, Roger D. Kouyos, Huldrych F. Günthard, Karin J. Metzner, Swiss HIV Cohort Study, A Anagnostopoulos, Manuel Battegay, Enos Bernasconi, Jürg Böni, Dominique L. Braun, Heiner C. Bucher, Alexandra Calmy, Matthias Cavassini, Angela Ciuffi, G Dollenmaier, Matthias Egger, Luigia Elzi, Jan Fehr, Jacques Fellay, Hansjakob Furrer, Christoph A. Fux, Huldrych F. Günthard, David Haerry, Barbara Hasse, Hans H. Hirsch, Matthias Hoffmann, Irène Hösli, Michael Huber, Christian R. Kahlert, Laurent Kaiser, Olivia Keiser, Thomas Klimkait, Roger D. Kouyos, Helen Kovari, Katharina Kusejko, Bruno Ledergerber, G Martinetti, Begoña Martínez de Tejada, Catia Marzolini, Karin J. Metzner, N Müller, Dunja Nicca, P Paioni, G Pantaleo, Matthieu Perreau, Andri Rauch, Christoph Rudin, Patrick Schmid, R Speck, M Stöckle, Philip Tarr, Alexandra Trkola, Pietro Vernazza, Gilles Wandeler, Rainer Weber, Sabine Yerly
Abstract
BACKGROUND: Genotypic resistance testing (GRT) is routinely performed upon diagnosis of HIV-1 infection or during virological failure using plasma viral RNA. An alternative source for GRT could be cellular HIV-1 DNA. OBJECTIVES: A substantial number of participants in the Swiss HIV Cohort Study (SHCS) never received GRT. We applied a method that enables access to the near full-length proviral HIV-1 genome without requiring detectable viraemia. METHODS: Nine hundred and sixty-two PBMC specimens were received. Our two-step nested PCR protocol was applied to generate two overlapping long-range amplicons of the HIV-1 genome, sequenced by next-generation sequencing (NGS) and analysed by MinVar, a pipeline to detect drug resistance mutations (DRMs). RESULTS: Six hundred and eighty-one (70.8%) of the samples were successfully amplified, sequenced and analysed by MinVar. Only partial information of the pol gene was contained in 82/681 (12%), probably due to naturally occurring deletions in the proviral sequence. All common HIV-1 subtypes were successfully sequenced. We detected at least one major DRM at high frequency (≥15%) in 331/599 (55.3%) individuals. Excluding APOBEC-signature (G-to-A mutation) DRMs, 145/599 (24.2%) individuals carried at least one major DRM. RT-inhibitor DRMs were most prevalent. The experienced time on ART was significantly longer in DRM carriers (P = 0.001) independent of inclusion or exclusion of APOBEC-signature DRMs. CONCLUSIONS: We successfully applied a reliable and efficient method to analyse near full-length HIV-1 proviral DNA and investigated DRMs in individuals with undetectable or low viraemia. Additionally, our data underscore the need for new computational tools to exclude APOBEC-related hypermutated NGS sequence reads for reporting DRMs.