Plasma HIV-1 Tropism and the Risk of Short-Term Clinical Progression to AIDS Or Death
Total Page:16
File Type:pdf, Size:1020Kb
RESEARCH ARTICLE Plasma HIV-1 Tropism and the Risk of Short- Term Clinical Progression to AIDS or Death Maria Casadellà1,2*, Alessandro Cozzi-Lepri3, Andrew Phillips3, Marc Noguera-Julian1,2,4, Markus Bickel5, Dalibor Sedlacek6, Kai Zilmer7, Bonaventura Clotet1,2,4,8, Jens D. Lundgren9, Roger Paredes1,2,4,8, EuroSIDA in EuroCOORD¶ 1 IrsiCaixa AIDS Research Institute, Badalona, Catalonia, Spain, 2 Universitat Autònoma de Barcelona, Catalonia, Spain, 3 Royal Free Hospital, London, United Kingdom, 4 Universitat de Vic-Universitat Central de Catalunya, Vic, Catalonia, Spain, 5 Goethe University, Frankfurt/Main, Germany, 6 Charles University a1111111111 Hospital, Plzen, Česka Republika, 7 West-Tallinn Central Hospital, Tallinn, Estonia, 8 HIV Unit, Hospital a1111111111 Universitari Germans Trias i Pujol, Badalona, Catalonia, Spain, 9 CHIP, Department of Infectious Diseases, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark a1111111111 a1111111111 ¶ Membership of this author group is listed in the Acknowledgments. a1111111111 * [email protected] Abstract OPEN ACCESS Citation: Casadellà M, Cozzi-Lepri A, Phillips A, Objective Noguera-Julian M, Bickel M, Sedlacek D, et al. (2017) Plasma HIV-1 Tropism and the Risk of To investigate if plasma HIV-1 tropism testing could identify subjects at higher risk for clinical Short-Term Clinical Progression to AIDS or Death. progression and death in routine clinical management. PLoS ONE 12(1): e0166613. doi:10.1371/journal. pone.0166613 Design Editor: Dimitrios Paraskevis, National and Nested case-control study within the EuroSIDA cohort. Kapodistrian University of Athens, GREECE Received: June 11, 2016 Methods Accepted: November 1, 2016 Cases were subjects with AIDS or who died from any cause, with a plasma sample with Published: January 27, 2017 HIV-1 RNA >1000 copies/mL available for tropism testing 3 to 12 months prior to the event. At least 1 control matched for age, HIV-1 RNA and HCV status at the time of sampling were Copyright: © 2017 Casadellà et al. This is an open access article distributed under the terms of the selected per each case. Conditional logistic regression was used to investigate exposures Creative Commons Attribution License, which associated with clinical progression to AIDS or death. A linear mixed model with random permits unrestricted use, distribution, and intercept was used to compare CD4+T-cell slopes by HIV tropism over the 12 months follow- reproduction in any medium, provided the original author and source are credited. ing the date of sampling. Data Availability Statement: Some of the data Results underlying the analyses presented are within the paper and its Supporting Information files. The The study included 266 subjects, 100 cases and 166 controls; one quarter had X4 HIV; 26% remaining data are available upon request because were ART-naïve. Baseline factors independently associated with clinical progression or of ethical and confidentiality constraints and in + accordance with EuroSIDA policy. In accordance death were female gender (OR = 2.13 vs. male, 95CI = 1.04, 4.36), p = 0.038), CD4 T-cell 3 with the ethics approval for the research, count (OR = 0.90 (95CI = 0.80, 1.00) per 100 cells/mm higher, p = 0.058), being on ART researchers can submit research proposals to the (OR = 2.72 vs. being off-ART (95CI = 1.15, 6.41), p = 0.022) and calendar year of sample EuroSIDA Steering Committee by contacting the [OR = 0.84 (95CI = 0.77, 0.91) per more recent year, p<0.001). Baseline tropism was not [email protected]. (For more + information see http://www.chip.dk/Studies/ associated with the risk of clinical progression or death. CD4 T-cell slopes did not differ EuroSIDA/Contacts.) within or between tropism groups. PLOS ONE | DOI:10.1371/journal.pone.0166613 January 27, 2017 1 / 14 HIV Tropism & AIDS Progression Funding: Primary support for EuroSIDA was Conclusions provided by the European Commission BIOMED 1 (CT94-1637), BIOMED 2 (CT97-2713), the 5th The predictive role of plasma tropism determined using 454 sequencing in the context of Framework (QLK2-2000-00773), the 6th people receiving cART with detectable VL is not helpful to identify subjects at higher risk for Framework (LSHP-CT-2006-018632), and the 7th clinical progression to AIDS or death. Framework (FP7/2007-2013, EuroCoord nÊ260694) programmes. Current support also includes unrestricted grants by Janssen R&D (www. janssen.com), Merck and Co. Inc. (www.merck. com), Pfizer Inc. (www.pfizer.com), GlaxoSmithKline LLC (www.gsk.com). The participation of centres from Switzerland was supported by The Swiss National Science Introduction Foundation (Grant 108787). It was also partially Infection with HIV-1 with tropism for lymphocytes expressing the CXCR4 co-receptor (X4 supported by Redes TemaÂticas de InvestigacioÂn en HIV, or CXCR4-tropic HIV) [1,2] has been consistently associated with lower CD4+T-cell SIDA (ISCIII RETIC RD12/0017/0002); AccioÂn counts in antiretroviral treatment (ART)-naïve and -experienced subjects [3,4], and with faster EstrateÂgica en Salud. Plan Nacional de + InvestigacioÂn CientõÂfica, Desarrollo e InnovacioÂn disease progression in untreated patients [5±7], independently of their baseline CD4 T-cell TecnoloÂgica 2008-2011; Instituto de Salud Carlos counts or HIV-1 RNA levels [8,9]. Initially, tropism assessments were based on phenotypic III, Fondos FEDER. The funders had no role in assays, but such assays have limitations for clinical routine diagnostics. They are costly, time study design, data collection and analysis, decision consuming, complex and require a minimum viral load of 1000 c/ml, being unsuitable for sub- to publish, or preparation of the manuscript. jects with low-level viremia or those requiring CCR5 antagonists while being suppressed. Gen- Competing Interests: The authors have declared otyping approaches using V3-loop sequencing were later implemented, being faster and that no competing interests exist. cheaper [10]. Both techniques share the challenge of detecting minority virus populations that may be clinically important. Whereas 60 to 80% of subjects at early stages of HIV infection harbour R5 HIV, X4 viruses emerge in approximately 50% of individuals at later stages of the disease [11,12]. In general, tropism switches occur from R5 to X4, but occasionally X4 to R5 switches can also be seen [13]. It has been much debated whether the emergence of X4 HIV is cause or consequence of immune depression [14]. Using modern ultrasensitive genotyping tools, X4 HIV can be detected in virtually any patient at very low levels shortly after HIV infection, suggesting that X4 HIV are being contained by the immune system. Conversely, in epidemiological studies R5 to X4 tropism switches preceded CD4+ declines [15]. Advances in genotyping techniques and the availability of bioinformatic tools to accurately predict phenotypic tropism from HIV env sequence data enable rapid, reliable and robust assessment of HIV tropism during routine clinical management. Early phenotypic tropism tools used to define the natural history of HIV, like the MT-2 or MT4 assays, were highly spe- cific for X4 viruses but often lacked sensitivity and reproducibility across laboratories due to the difficulty in establishing cell infections. Population sequencing allows streamlined process- ing in low complexity labs and achieves an accuracy to predict the true tropism±defined as HIV tropism determined using the enhanced-sensitivity Trofile1 assay- of around 90%, with a sensitivity to detect X4 HIV in the range of 50-70% and specificity >90% [16]. The sensitivity and specificity of genotypic tropism prediction in plasma are improved by using massively par- allel deep sequencing with either 454, Illumina or any equivalent next-generation sequencing platform [17]. Although tropism testing is mandatory before using a CCR5-inhibitor, it is uncertain whether clinicians should order a tropism test to predict if their patients could have an increased risk of clinical progression or if, instead, they should rely on other more classical pre- dictors such as CD4+ counts, viral load etc. In a previous cohort study [11], subjects with X4 HIV had faster decline in CD4+ T-cell counts and suffered more clinical events over the fol- lowing 12 months remaining off ART than those with R5 HIV. However, such differences were no longer observed in subjects receiving antiretroviral treatment. PLOS ONE | DOI:10.1371/journal.pone.0166613 January 27, 2017 2 / 14 HIV Tropism & AIDS Progression The aim of this study was to determine whether, in people with detectable viral load, pres- ence of X4 tropic HIV in plasma, determined using either population sequencing or 454 deep sequencing, was independently associated with an increased risk of progression to AIDS or death over the following 3 to 12 months. We also sought to evaluate in the same study popula- tion if X4 HIV was associated with steeper declines in CD4+ counts than in people with R5 HIV. The ultimate objective of our analyses was to verify whether plasma tropism testing is useful to predict clinical outcomes in routine HIV clinical management. Methods Subjects This was a nested case-control study within the EuroSIDA cohort (described in [18]). This study underwent an Ethics Comitee in order to assure subjects confidentiality and privacy. The study complies with the regulations for Ethics Committee approval and procedure for obtaining Informed Consent from participants for the EuroSIDA Study, following the Euro- sida Study Group standards. A data collection form is completed by the sites at the time of enrolment and every 6 months hereafter on to a standardized data collection form. From fol- low-up Winter 2014/2015 (DS41), all data collection is in the electronic CRF system, REDCap at https://chip-crf.info/redcap/. For each patient, the date of HIV diagnosis and way of trans- mission is recorded. The date of diagnosis of all AIDS defining diseases are recorded, using the 1993 Centers for Disease Control and Prevention definitions. Further, the date of starting and stopping each antiretroviral drug is recorded, as is the use of drugs for prophylaxis against opportunistic infections.