Drug Resistance Mutations for Surveillance of Transmitted HIV-1 Drug-Resistance: 2009 Update

et al. (2009) Drug Resistance Mutations for Surveillance of Transmitted HIV-1 Drug-Resistance: 2009 Update. PLoS ONE 4(3): e4724. doi:10.1371/journal.pone.0004724 Drug Resistance Mutations for Surveillance of Transmitted HIV-1 Drug-Resistance: 2009 Update Diane E. Bennett Ricardo J. Camacho Dan Otelea Daniel R. Kuritzkes Herve Fleury Mark Kiuchi Walid Heneine Rami Kantor Michael R. Jordan Jonathan M. Schapiro Anne-Mieke Vandamme Paul Sandstrom Charles A. B. Boucher David van de Vijver Soo-Yon Rhee Tommy F. Liu Deenan Pillay Robert W. Shafer Douglas F. Nixon, University of California San Francisco, United States of America Programs that monitor local, national, and regional levels of transmitted HIV-1 drug resistance inform treatment guidelines and provide feedback on the success of HIV-1 treatment and prevention programs. To accurately compare transmitted drug resistance rates across geographic regions and times, the World Health Organization has recommended the adoption of a consensus genotypic definition of transmitted HIV-1 drug resistance. In January 2007, we outlined criteria for developing a list of mutations for drug-resistance surveillance and compiled a list of 80 RT and protease mutations meeting these criteria (surveillance drug resistance mutations; SDRMs). Since January 2007, several new drugs have been approved and several new drug-resistance mutations have been identified. In this paper, we follow the same procedures described previously to develop an updated list of SDRMs that are likely to be useful for ongoing and future studies of transmitted drug resistance. The updated SDRM list has 93 mutations including 34 NRTI-resistance mutations at 15 RT positions, 19 NNRTI-resistance mutations at 10 RT positions, and 40 PI-resistance mutations at 18 protease positions. - Competing Interests: The authors have declared that no competing interests exist. Introduction The worldwide effort to improve treatment outcomes and reduce transmission of HIV through optimal delivery of ART and HIV prevention programmes must be coordinated with and enlightened by ongoing national, regional, and global evaluations of HIV drug resistance. One essential element in the global evaluation is population-based surveillance of transmitted HIV drug resistance in recently infected individuals. As HIV drug resistance surveillance programs are underway in many countries and regions, it has become essential to develop a standard list of mutations to characterize the epidemiology of transmitted drug resistance [1,2,3,4,5]. Only with a standard list of mutations is it possible to compare the prevalence of transmitted resistance from different times and regions and facilitate meta-analyses of surveillance data collected by different groups at different times. Compiling such a standard list, however, is not simple because of the rapidly changing field of ARV therapy and the large numbers of mutations associated with ARV drug resistance [6,7]. In 2007, we outlined four criteria for identifying surveillance drug-resistance mutations (SDRMs) and used these criteria to create a provisional list of SDRMs [6]. The first criterion was that SDRMs should be recognized as causing or contributing to drug resistance defined as being present on three or more of five expert lists of drug resistance mutations. The second criterion was that mutations should be non-polymorphic and should not occur at highly polymorphic positions. The third criterion was that the mutation list had to be applicable to the eight most common HIV1 subtypes. The fourth criterion was that the list should be parsimonious, excluding mutations resulting exceedingly rarely from drug pressure. Since the 2007 list was published, new drug-resistance mutations have been identified including mutations arising from the increased use of non-thymidine-analog containing regimens, the expanded use of two new protease inhibitors (PIs), and the recent approval of a new non-nucleoside RT inhibitor (NNRTI). The number of sequences from ARV-nave persons infected with subtype B and non-B HIV-1 viruses in our analysis dataset has approximately doubled since the 2007 publication, increasing the confidence with which nonpolymorphic mutations can be identified. In this paper, we followed the same steps used to create the 2007 mutation list. Identification of mutations causing or contributing to drug resistance Mutations that were present on three or more of the following five expert lists ANRS drug resistance interpretation algorithm (2008.07), HIVdb drug resistance interpretation algorithm (4.3.7), IAS-USA Mutations Associated With Drug Resistance (March/ April 2008), Los Alamos National Laboratories HIV Sequence database (2007), or Rega Institute Drug Resistance Interpretation Algorithm (7.1.1) were considered to be recognized as causing or contributing to drug resistance. The complete list of mutations associated with each of these lists can be found on the Surveillance Drug Resistance Mutation (SDRM) worksheet (http://hivdb. stanford.edu/cgi-bin/AgMutPrev.cgi). Identification of nonpolymorphic mutations and mutations not occurring at highly polymorphic positions Some drug resistance mutations occur commonly in the absence of drug selective pressure, these polymorphic drug-resistance mutations should not be used for surveillance of transmitted drug resistance because they could lead to falsely elevated estimates of transmitted resistance. For the purposes of generating a nonpolymorphic list of drug resistance mutations, we defined nonpolymorphic mutations to be mutations present at a frequency #0.5% in ARV-nave individuals infected with subtypes for which .1,000 sequences were available in our dataset and at levels .0.5% in no more than one subtype for which fewer than 1,000 sequences were available. Nonpolymorphic mutations occurring at polymorphic positions, defined as positions with mutations occurring at .1% in any subtype, were generally excluded. Exceptions were made for major mutations that directly contribute to causing resistance. Assignment of HIV-1 subtype A set of 100 reference sequences was compiled by combining 65 representative group M sequences curated by the Los Alamos Sequence Database and an additional 35 samples added so that the dataset would include three or more divergent reference sequences for each pure subtype and many of the most common CRFs. Neighbor joining trees were created from an alignment of each sequence with the 100 reference sequences. Sequences clustering within clades formed by subtypes A, B, C, D, F, G, H, J, and K, and CRF01_AE and CRF02_AG sequences were assigned to that clade. Sequences grouping within clades CRF_03 to CRF_19 were assigned to that clade unless the region spanned by the CRF mapped onto one of the pure subtypes or CRF01_AE or CRF02_AG, in which case the sequence was assigned to one of these. Sequences that were not within a clade were assigned to the subtype or CRF of the closes (...truncated)