Factors Associated With Persistence of Plasma HIV-1 RNA During Long-term Continuously Suppressive Firstline Antiretroviral Therapy

Factors Associated With Persistence of Plasma HIV-1 RNA During Long-term Continuously Suppressive... Open Forum Infectious Diseases MAJOR ARTICLE Factors Associated With Persistence of Plasma HIV-1 RNA During Long-term Continuously Suppressive Firstline Antiretroviral Therapy 1 2 1 3 4 2 1 Alessandra Ruggiero, Alessandro Cozzi-Lepri, Apostolos Beloukas, Douglas Richman, Saye Khoo, Andrew Phillips, and Anna Maria Geretti ; on behalf of the ERAS Study Group 1 2 Institute of Infection and Global Health, University of Liverpool, Liverpool, United Kingdom; Department of Infection and Population Health, University College London, London, United Kingdom; 3 4 VA San Diego Healthcare System and Center for AIDS Research, University of California San Diego, La Jolla, California; Institute of Translational Medicine, University of Liverpool, Liverpool, United Kingdom Background. Persistence of plasma HIV-1 RNA during seemingly effective antiretroviral thereapy (ART) is incompletely under - stood. Using an ultrasensitive assay, this cross-sectional study investigated residual plasma HIV-1 RNA in subjects maintained on firstline ART with continuous viral load suppression <50 copies/mL for ≤15 years without recognized viral load blips or treatment interruptions and explored its relationship with the duration of suppressive ART, efavirenz concentrations in plasma, 2-LTR circular HIV-1 DNA (2-LTRc DNA) in peripheral blood mononuclear cells, and cellular (CD4 plus CD26/CD38/CD69; CD8 plus CD38/ HLA-DR/DP/DQ) and soluble (sCD14, sCD27, sCD30, IL-6) markers of immune activation in peripheral blood.  Methods. Residual plasma HIV-1 RNA, total HIV-1 DNA and 2-LTRc DNA were quantified by real-time and digital droplet PCR. Cellular (CD4 plus CD26/CD38/CD69; CD8 plus CD38/HLA-DR/DP/DQ) and soluble (sCD14, sCD27, sCD30, IL-6) mark- ers of immune activation were measured by flow cytometry and ELISA. Results. Residual plasma HIV-1 RNA and 2-LTRc DNA were detected in 52/104 (50%) and 24/104 (23%) subjects, respectively. Among subjects with detectable HIV-1 RNA, 50/52 showed levels ≤11 copies/mL. In adjusted analyses, HIV-1 RNA levels were 0.37 log copies/mL higher with each log U/mL increase in sCD27 (95% confidence interval, 0.01–0.73; P = .02). No significant associ- 10 10 ation was found between residual plasma HIV-1 RNA and other explored parameters.  Conclusions. es Th e findings point to an ongoing relationship between plasma HIV-1 RNA and selected markers of immune activation during continuously suppressive ART. The novel direct association with levels of sCD27 warrants further investigation. Keywords. 2-LTR circular DNA; activation; drug concentration; sCD27; viral load. Maintaining plasma HIV-1 RNA suppression below the limit of heterogeneous treatment histories and varying definitions of detection of commercial viral load assays defines clinical suc- suppression, which may in part explain conflicting findings cess of antiretroviral therapy (ART) [1, 2]. Ultrasensitive assays [11–13]. can detect traces of HIV-1 RNA in the plasma of many success- Identifying the determinants of HIV-1 RNA persistence fully treated patients [3], but the source remains controversial during seemingly effective ART is central to our understand- [4, 5]. On the one hand, it is proposed that detection of residual ing of how the virus may be eradicated [14, 15]. Such studies plasma HIV-1 RNA reflects short-lived bursts of virus produc- must also consider the interplay between virus replication and tion from latently infected memory CD4 T cells, but with ongo- host immunity, and the role of chronic immune activation and ing virus replication and genetic evolution inhibited by ART [6]. inflammation in disease pathogenesis [16–18]. Effective ART An alternative model derives from evidence of ongoing virus reduces but does not always normalize inflammatory mark- replication in compartments where drug penetration or activ- ers, raising the possibility that residual immune activation may ity is suboptimal [7–10]. While the 2 models are not mutually both sustain and be sustained by residual virus replication [18]. exclusive, published studies have oen in ft cluded patients with Available data are not conclusive, however, and they rather point to virologic and immunologic events that take place before the initiation of ART as key determinants of the HIV-1 reservoirs Received 6 December 2017; editorial decision 25 January 2018; accepted 2 February 2018. and immune activation that can persist despite ART [19]. Correspondence: A. M. Geretti, MD, PhD, Department of Clinical Infection, Microbiology and We previously reported that treated patients with plasma Immunology, Institute of Infection and Global Health, University of Liverpool, 8 West Derby Street, Liverpool L69 7BE, UK (geretti@liverpool.ac.uk). HIV-1 RNA detected at levels ranging between approximately Open Forum Infectious Diseases 10 and 49 copies/mL by a commercial viral load assay were more © The Author(s) 2018. Published by Oxford University Press on behalf of Infectious Diseases likely to experience viral load rebound above 50 copies/mL and Society of America. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits above 400 copies/mL during 12  months of follow-up when unrestricted reuse, distribution, and reproduction in any medium, provided the original work compared with treated patients with undetectable HIV-1 RNA is properly cited. [20]. e Th aim of this study was to take the converse approach DOI: 10.1093/ofid/ofy032 Residual HIV-1 RNA During ART • OFID • 1 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 and assess a population that experienced continuous viral load Following centrifugation, cell pellets were lysed by 10-minute suppression <50 copies/mL for up to 15 years of firstline ART incubation at room temperature in 1× Hoffman’s lysis buffer and use ultrasensitive testing to detect HIV-1 RNA below the and incubated with FITC-, Cy5- or PE-labeled antibodies tar- 10-copies/mL cuto. Th ff e cohort was selected according to pre- geting CD4 plus CD26, CD38, or CD69, and CD8 plus CD38 or cise inclusion criteria, and residual plasma HIV-1 RNA was HLA-DR/DP/DQ (Becton Dickinson, Swindon, UK). Labeled investigated cross-sectionally in relation to the duration of sup- cells were incubated in 1% paraformaldehyde for 30 minutes pressive ART, plasma concentrations of efavirenz, 2-LTRc DNA prior to acquisition with BD FACSCalibur and analysis by in peripheral blood mononuclear cells (PBMCs), and cellular CellQuest. Soluble markers were measured in frozen plasma and soluble markers of immune activation in peripheral blood. by commercial Elisa assays (EIAs) for soluble CD14 (sCD14, −6 R&D system, London, UK; assay sensitivity ≥125 ×  10 µg/ METHODS mL), soluble CD27 (sCD27, EBiosciences; ThermoFisher, Loughborough, UK; assay sensitivity ≥0.2 U/mL), soluble CD30 Study Population (sCD30, EBiosciences, UK; assay sensitivity ≥0.3  ng/mL), and In this cross-sectional study, eligible patients had started firstline IL6 (EBiosciences, UK, assay sensitivity ≥0.9 pg/mL). ART with 2 nucleoside/nucleotide reverse transcriptase inhibi- tors (NRTIs) plus 1 non-nucleoside reverse transcriptase inhib- Plasma Drug Concentrations itor (NNRTI), had achieved viral load suppression <50 copies/ Efavirenz (EFV) and nevirapine (NVP) concentrations were mL within 6 months of starting ART, and during subsequent fol- measured in untimed plasma samples using high-performance low-up had remained on the initial NNRTI, undergone ≥2 viral liquid chromatography–tandem mass spectrometry, as previ- load measurements per year, and maintained continuous viral ously described [10]. The time of sample collection relative to load suppression <50 copies/mL without any recorded viral load last dosing was recorded. elevation >50 copies/mL or treatment interruption. To ensure balance in terms of duration of suppressive ART, recruitment was Analysis stratified to span from 1 year of treatment to 10 years or longer. Duration of suppressive ART was defined as the length of time Changes of the initial NNRTI were not allowed. Changes of following the first viral load measurement <50 copies/mL, which the initial NRTIs were allowed to reflect evolving practice, pro- in all subjects was recorded within 6  months of starting ART. vided they did not coincide with a viral load >50 copies/mL or a When analyzing residual plasma HIV-1 RNA and 2-LTRc DNA treatment interruption. The study took place at multiple centers as continuous variables, undetectable values were replaced with across the United Kingdom. It was approved by the National the midpoint value between 0 and the lower limit of detection Research Ethics Service (London-Dulwich) and included in the of the assays. EFV and NVP concentrations were categorized as National Institute for Health Research Clinical Research Network above or below the recommended minimal effective concentra- Portfolio. Patients provided written informed consent. tion (MEC) for wild-type virus: 1000  ng/mL for EFV [24] and 3400 ng/mL for NVP [25]. Participants’ characteristics were com- Residual Plasma HIV-1 RNA pared by analysis of variance (ANOVA), the Kruskal-Wallis test, Plasma was separated from venous blood in EDTA within 2 and the Fisher exact test for continuous and categorical variables, hours of collection and stored at −80°. HIV-1 RNA was quan- as appropriate. Data distribution was assessed through descrip- tified using a modified version of the RealTime HIV-1 assay tive scatter plots. The correlation between EFV concentrations (Abbott, Maidenhead, UK). Validation studies previously and levels of residual plasma HIV-1 RNA and 2-LTRc DNA was showed that the assay 50% and 95% detection rates were 1 and 3 measured by Spearman’s rank test. Given that all but 2 HIV-1 HIV-1 RNA copies/mL, respectively, with high reproducibility RNA values were ≤11 copies/mL, in order to reduce heterogene- across diverse HIV-1 subtypes [10, 20, 21]. ity, the main statistical models excluded 2 subjects with residual Total HIV-1 DNA and 2-LTRc DNA plasma HIV-1 RNA >11 copies/mL (referred to as outliers). This PBMCs were isolated by Ficoll-Hypaque gradient centrifuga- was based on our previous data indicating that patients on stable tion and cryopreserved. Total HIV-1 DNA was quantified by ART with a viral load above approximately 10 copies/mL are a real-time polymerase chain reaction (PCR) as described [22]; distinct population at risk of viral load rebound [20]. Three main the assay 50% and 95% detection rates were 20 and 40 HIV-1 models were run. The mean difference (with 95% confidence DNA copies/10 PBMC, respectively. 2-LTRc DNA was meas- interval [CI]) in residual plasma HIV-1 RNA, total HIV-1 DNA, ured by droplet digital PCR as described [23]; the assay’s lower 2-LTRc DNA, and markers of immune activation according to limit of detection was a median (interquartile range [IQR]) of 5 duration of suppressive ART (normalized over 1 year) was ana- (4–6) copies/10 PBMC.  lyzed by univariable linear regression analysis after log-transfor- Cellular and Soluble Markers of Immune Activation mation of the variables. Factors associated with levels of residual Flow cytometry was performed at the Immunology Service of plasma HIV-1 RNA or 2-LTRc DNA were investigated by uni- the Royal Free Hospital in London using freshly collected blood. variable and multivariable linear regression analysis. Variables 2 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 associated (P ≤ .20) with the outcome of interest in the univaria- <50 copies/mL for a median (IQR) of 5 (3–8) years while receiv- ble analysis were included in the multivariable models, with the ing firstline ART with 2 NRTIs plus either EFV (n = 86, 83%) following modifications: (a) when modeling 2-LTRc DNA, total or NVP (Table 1). Reflecting evolving guidelines about when to HIV-1 DNA was excluded due to high colinearity (P  <  .0001); start ART and which antiretrovirals to use, patients with longer (b) age and duration of ART were excluded in favor of duration duration of suppressive ART had a lower nadir CD4 cell count of suppressive ART. Separate models explored: (a) the effect of and a higher pre-ART viral load, were more often receiving forcing the nadir CD4 cell count and pre-ART viral load into the NVP, and were more likely to have experienced changes of the multivariable analyses and (b) the association between residual NRTI backbone after first starting ART: 48/104 (48%) patients plasma HIV-1 RNA or 2-LTRc DNA and plasma concentrations changed ≥1 NRTIs prior to recruitment, with a median (IQR) of of EFV (as a continuous variable) in EFV recipients, after adjust- 1 (1–2) NRTI change per subject. At recruitment, NRTI back- ment for the time of sampling in relation to the last EFV dose. bones comprised tenofovir disoproxil fumarate (78/104, 75%), IBM SPSS v. 22.0 (Armonk, NY, USA) and SAS v9.4 (Cary, NC, abacavir (22/104, 21%) or zidovudine (4/104, 4%), plus emtric- USA) were used for the analyses. itabine or lamivudine. Residual plasma HIV-1 RNA was detected in 52/104 (50%) RESULTS patients (Table  1). Levels ranged between 1 and 35 copies/ mL and were ≤11 copies/mL in 50/52 (96%) patients; levels in Study Population the 2 outliers were 35 and 24 copies/mL aer 2 a ft nd 3 years of The population comprised 104 patients that at the time of suppressive ART, respectively (Figure 1). 2-LTRc HIV-1 DNA recruitment had experienced continuous viral load suppression Table 1. Characteristics of the Study Population Overall and by Duration of Suppressive ART (n = 104) Duration of Suppressive ART, y Total (n = 104) <4 (n = 34) 4–7 (n = 35) >7 (n = 35) Duration of suppressive ART, median (IQR), y 5 (3–8) 2 (2–3) 5 (5–6) 9 (8–10) Age, median (IQR), y 47 (40–53) 44 (36–49) 49 (39–53) 48 (43–55) Male, n (%) 81 (78) 29 (85) 29 (83) 23 (66) Ethnicity, n (%) White 59 (57) 20 (59) 22 (63) 17 (49) Black 41 (39) 13 (38) 11 (31) 17 (49) Asian 4 (4) 1 (3) 2 (6) 1 (3) Pre-ART viral load, median (IQR), log cps/mL 4.9 (4.6–5.3) 4.7 (4.1–5.1) 4.9 (4.6–5.3) 5.1 (4.8–5.6) Nadir CD4 count, median (IQR), cells/mm 201 (110–270) 279 (216–365) 166 (126–238) 118 (79–215) EFV use, n (%) 86 (83) 30 (88) 32 (89) 24 (69) NNRTI concentration, median (IQR), ng/mL EFV 1531 (846–2235) 1492 (735–2130) 1496 (866–2154) 1966 (855–2626) NVP 5889 (3096–7419) 4545 (2798–6147) 2734 (2539–5743) 6311 (4390–7603) NNRTI concentration above target MEC , n (%) 71 (66) 22 (65) 24 (69) 25 (71) Changed initial NRTIs, n (%) 48 (46) 4 (12) 13 (37) 31 (89) Detectable residual plasma HIV-1 RNA, n (%) 52 (50) 21 (62) 18 (51) 13 (37) Residual plasma HIV-1 RNA, median (IQR), cps/mL 2 (2–4) 2 (2–4) 2 (2–4) 2 (2–2) Detectable 2-LTRc DNA, n (%) 24 (23) 12 (35) 7 (20) 5 (14) d 6 2-LTRc DNA, median (IQR), cps/10 PBMC 3 (3–3) 3 (3–7) 3 (3–3) 3 (3–3) Total HIV-1 DNA, median (IQR), log cps/10 PBMC 2.5 (2.1–2.8) 2.6 (2.3–3.0) 2.4 (1.7–2.8) 2.5 (2.1–2.8) CD4 count, median (IQR), cells/mm 581 (474–722) 548 (427–629) 580 (509–720) 632 (503–796) CD8 count, median (IQR), cells/mm 798 (656–1047) 808 (596–1060) 818 (685–1024) 679 (496–929) + + CD4 CD38 , median (IQR), % 24 (16–34) 32 (23–36) 21 (15–31) 22 (15–28) + + CD4 CD26 , median (IQR), % 58 (46–64) 56 (47–67) 59 (49–63) 58 (44–64) + + CD4 CD69 , median (IQR), % 1 (1–2) 2 (1–3) 1 (1–2) 2 (1–2) + + CD8 CD38 , median (IQR), % 5 (3–7) 6 (4–8) 4 (3–6) 4 (3–6) + + CD8 HLA-DR/DP/DQ , median (IQR), % 35 (24–42) 37 (27–47) 35 (26–40) 30 (20–39) 2.2 (1.8–2.5) 2.2 (1.8–2.4) 2.0 (1.8–2.5) 2.3 (1.9–2.5) sCD14, median (IQR), µg/mL sCD27, median (IQR), Ug/mL 62 (49–76) 64 (52–80) 61 (54–76) 60 (46–71) sCD30, median (IQR), ng/mL 11 (9–15) 11 (9–15) 12(10–17) 11 (9–13) IL6 median (IQR), pg/mL 0.8 (0.5–2.1) 0.5 (0.5–1.2) 1.1 (0.5–2.3) 0.8 (0.5–2.1) Abbreviations: 2-LTRc DNA, 2-LTR circular HIV-1 DNA; ART, antiretroviral therapy; cps, copies; EFV, efavirenz; IQR, interquartile range; MEC, minimum effective concentration; NNRTI, non-nu- cleoside reverse transcriptase inhibitor; NRTI, nucleoside/nucleotide reverse transcriptase inhibitor; NVP, nevirapine; PBMC, peripheral blood mononuclear cell. Defined as the length of time between the first viral load <50 copies/mL measured after starting ART and the date of recruitment. Minimum effective concentration recommended for wild-type virus (1000 ng/mL for EFV and 3400 ng/mL for NVP). Samples with undetectable HIV-1 RNA were assigned an arbitrary value of 1.5 copies/mL.  d 6 Samples with undetectable 2-LTRc DNA were assigned an arbitrary value of 2.5 copies/10 PBMC. Residual HIV-1 RNA During ART • OFID • 3 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 A Residual plasma HIV-1 RNA B 2-LTRc DNA 0 0 01 23 45 67 89 10 111213 1415 01 2 3456789 101112131415 Duration of suppressive ART, years Duration of suppressive ART, years C D Total HIV-1 DNA sCD27 0 12345678 9101112131415 05 10 15 Duration of suppressive ART, years Duration of suppressive ART, years Figure 1. Virologic and immunologic parameters in patients with up to 15 years of consistently suppressive antiretroviral therapy (ART; n = 104), comprising (A) residual plasma HIV-1 RNA (copies/mL), (B) 2-LTR circular HIV-1 DNA (2-LTRc DNA; copies per 10 peripheral blood mononuclear cell [PBMC]), (C) total HIV-1 DNA (log copies per 10 PBMC), and (D) sCD27 (U/mL) in relation to the duration of suppressive ART (years). In (D), the dotted line indicates the mean sCD27 levels reported in HIV-negative volunteers in published studies (164 U/mL) [25]; the solid lines indicate the median values measured in the whole study population. Each circle represents 1 participant; the 2 solid squares represent the 2 patients with residual plasma HIV-1 RNA >11 copies/mL. Virologic and Immunologic Parameters in Relation to the Duration of was detected in 24/104 (23%) patients with median levels of 3 6 6 Suppressive ART copies/10 PBMC that ranged between 4 and 119 copies/10 Differences in virologic and immunologic parameters PBMC (Figure 1). Cellular (Supplementary Figure 1) and sol- according to the duration of suppressive ART, analyzed after uble (Figure  1; Supplementary Figure  2) markers of immune exclusion of the 2 outliers, are shown in Table 2. Patients with activation were detected in all patients. Levels of soluble mark- longer duration of suppressive ART had higher CD4 counts ers were generally not dissimilar from those of HIV-negative (+19 cells/mm /y) and lower expression of activation mark- volunteers reported in published studies [26–29]. When ana- ers on CD4 cells (CD38) and CD8 cells (CD38, HLA-DR/ lyzing subjects whose levels of soluble markers fell above the DP/DQ). Levels of sCD14, sCD30, and IL-6 did not differ by upper quartile of the whole study population, a few patients duration of suppressive ART, whereas levels of sCD27 were presented concomitant elevations, but with little evidence significantly lower with longer duration of suppressive ART. of consistent patterns. Overall 5/104 (5%) patients showed A  trend was observed for lower levels of 2-LTRc DNA with concomitant elevations of >2 soluble markers, comprising longer duration of suppressive ART. There was less evidence sCD14 + sCD27 + sCD30 (3/104, 3%); CD27 + sCD30 + IL-6 of a difference in residual plasma HIV-1 RNA, and no sig- (1/104, 1%); and sCD14 + sCD27 + sCD30 + IL-6 (1/104, 1%) nificant difference was seen in total HIV-1 DNA. A  linear (Supplementary Table  1). CD8 cells counts and frequency of + + regression analysis confirmed an inverse association between CD8 HLA-DR/DP/DQ cells were oen, a ft lbeit not always, 2-LTRc DNA and duration of suppressive ART, with an increased in these patients, but there were no apparent patterns adjusted mean difference of 0.03 log copies/10 PBMC per in other parameters. 4 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Total HIV-1 DNA, HIV-1 RNA, copies/mL log copies/10 PBMC 2-LTRc HIV-1 DNA, sCD27, U/mL copies/10 PBMC Table  2. Univariable Analysis of the Mean Difference in Virologic and Factors Associated With Residual Plasma HIV-1 RNA Immunologic Parameters per 1 Year of Suppressive ART (n = 102) The univariable linear regression analysis indicated a signif- icant direct association between residual plasma HIV-1 RNA Mean b and sCD27 among the 102 subjects with HIV-1 RNA ≤11 cop- Difference 95% CI P ies/mL (Table  3). There was a weaker direct association with Residual plasma HIV-1 RNA, –0.013 –0.030 to 0.004 .12 cps/mL sCD30, whereas no association was evident with other markers 2-LTRc DNA, cps/10 PBMC –0.018 –0.039 to 0.003 .09 of immune activation. There was also no association between Total HIV-1 DNA, cps/10 PBMC 0.004 –0.033 to 0.042 .82 residual plasma HIV-1 RNA and total HIV-1 DNA and 2-LTRc CD4 count, cells/mm 0.015 0.004 to 0.026 .01 DNA. The direct association between residual plasma HIV-1 CD8 count, cells/mm –0.004 –0.018 to –0.009 .52 + + RNA and sCD27 persisted after adjusting for the duration of CD4 CD38 , % –0.022 –0.034 to –0.009 .001 + + + + suppressive ART, frequency of CD8 CD38 cells, and levels of CD4 CD26 , % –0.003 –0.012 to 0.005 .44 + + CD4 CD69 , % –0.008 –0.021 to 0.005 .21 sCD30 and IL-6 (Table  3). There was no apparent association + + CD8 CD38 , % –0.014 –0.032 to 0.004 .05 between residual plasma HIV-1 RNA and nadir CD4 cell count + + CD8 HLA-DR/DP/DQ , % –0.014 –0.027 to 0.001 .04 and pre-ART viral load in the univariable analysis. Forcing the sCD14, µg/mL 0.002 –0.004 to 0.009 .50 nadir CD4 cell count and pre-ART viral load into the multivar- sCD27, U/mL –0.010 –0.019 to 0.000 .05 iable model did not change the findings: residual plasma HIV-1 sCD30, ng/mL –0.005 –0.019 to 0.009 .44 RNA was 0.44 log copies/mL higher per each log increase IL-6, pg/mL 0.003 –0.015 to 0.022 .72 10 10 in sCD27 (95% CI, 0.09–0.79; P  =  .02; not shown). The asso- Abbreviations: 2-LTRc DNA,  2-LTR circular HIV-1 DNA; CI,  confidence interval; PBMC, peripheral blood mononuclear cell. ciation similarly persisted in analyses including the 2 outliers The analysis excluded the 2 outliers with residual plasma HIV-1 RNA >11 copies/mL.  (not shown). Variables were log-transformed prior to the analysis. Relationship Between Virologic Parameters and Plasma Drug Concentrations year longer (95% CI, –0.05 to –0.00; P = .02) (Supplementary We first explored the association between plasma EFV concen- Table 2). Forcing the nadir CD4 cell count and pre-ART viral trations and virologic parameters in 76 EFV recipients (includ- load into the multivariable model did not change the find- ing the 2 outliers) who had taken the last EFV dose a median ings: 2-LTRc DNA levels were 0.03 log copies/10 PBMC (IQR) of 12 (10–13) hours prior to sampling, and were there- lower per year of suppressive ART (95% CI, –0.05 to –0.00; fore in the range of the “mid-dose” interval, where targets have P = .02; not shown). been previously defined [30]. Although EFV concentrations Table 3. Univariable and Multivariable Linear Regression Analysis of Factors Associated With Mean Differences in Levels of Residual Plasma HIV-1 RNA (n = 102) Univariate Multivariable Mean Difference 95% CI P Mean Difference 95% CI P Nadir CD4 count per 100 cell/mm higher 0.01 –0.06 to 0.05 .58 Pre-ART viral load per log cps/mL higher 0.00 –0.08 to 0.08 .99 Duration of suppressive ART per 1 y longer –0.01 –0.03 to 0.00 .12 –0.01 –0.02 to 0.01 .39 CD4 count per 100 cells/mm higher –0.01 –0.02 to 0.02 .92 CD4/CD8 ratio per 1 unit higher 0.04 –0.11 to 0.17 .62 2-LTRc DNA per log cps/10 PBMC higher 0.05 –0.11 to 0.20 .57 Total HIV-1 DNA per log cps/10 PBMC higher –0.01 –0.09 to 0.08 .87 + + CD4 CD38 per 50% higher 0.08 –0.14 to 0.30 .47 + + CD4 CD26 per 50% higher 0.08 –0.10 to 0.27 .37 + + CD4 CD69 per 50% higher –0.61 –2.97 to 1.75 .60 CD8 count per 100 cells/mm higher 0.02 –0.01 to 0.01 .70 + + CD8 CD38 per 50% higher 0.32 –0.11 to 0.75 .16 0.26 –0.17 to 0.68 .24 + + CD8 HLA-DR/DP/DQ per 50% higher –0.03 –0.21 to 0.15 .76 sCD14 per log µg /mL higher 0.26 –0.24 to 0.75 .31 sCD27 per log U/mL higher 0.49 0.17 to 0.81 .003 0.37 0.01 to 0.73 .02 sCD30 per log ng/mL higher 0.21 –0.02 to 0.45 .07 0.02 –0.25 to 0.28 .88 IL6 per log pg/mL higher 0.13 –0.05 to 0.30 .15 0.12 –0.06 to 0.29 .19 Abbreviations: 2-LTRc DNA, 2-LTR circular HIV-1 DNA; ART, antiretroviral therapy; CI, confidence interval; cps, copies; PBMC, peripheral blood mononuclear cell. The analysis excluded the 2 outliers with residual plasma HIV-1 RNA >11 copies/mL.  Mean difference in log copies/mL.  Variables with P < .2 in the univariable analysis were included in the multivariable analysis. Residual HIV-1 RNA During ART • OFID • 5 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 varied across patients, there was no clear correlation with levels measurements made in peripheral blood. In this context, the of residual plasma HIV-1 RNA and 2-LTRc DNA (Spearman’s findings may provide support to the notion that during effect- rank test) (Figure  2). Two separate linear regression analy- ive treatment, detection of a few HIV-1 RNA copies in plasma ses explored the association between EFV concentrations and reflects transient bursts of virus production from latent reser - residual plasma HIV-1 RNA and 2-LTRc DNA among the 74 voirs that are inhibited from ongoing replication by ART [8, 32]. patients with residual plasma HIV-1 RNA ≤11 copies/mL. In In this model, a direct correlation may be anticipated between models that included both EFV concentration and timing of the likelihood of HIV-1 RNA persistence and the size of the sampling relative to last EFV dose, there was no evidence of an viral reservoir. We did not detect an association between resid- association between plasma HIV-1 RNA (P  =  .30) or 2-LTRc ual HIV-1 RNA in plasma and total HIV-1 DNA in PBMC, and DNA (P  =  .47) and EFV concentrations (Supplementary we similarly did not previously observe an association with inte- Table 3). grated HIV-1 DNA in PBMC [21]. Findings from other studies are conflicting, possibly reflecting differences in the compos- DISCUSSION ition of the study populations (eg, proportion of patients on This study demonstrated that half of subjects who had expe- protease inhibitors) and the method used for measuring HIV-1 rienced viral load suppression <50 copies/mL continuously RNA [33]. We also measured 2-LTRc DNA in PBMC as a pos- for up to 15 years while maintained on the initial NNRTI had sible indicator of viral gene expression and virus replication detectable HIV-1 RNA in plasma when tested with an ultra- and found no correlation with residual plasma HIV-1 RNA. In sensitive assay, at levels that rarely exceeded 11 copies/mL. We a previous study of 30 subjects who had received various ART previously reported that in treated patients, plasma HIV-1 RNA regimens for 7–12  years, 2-LTRc DNA was detected in 8 sub- levels above approximately 10 copies/mL were associated with jects (27%) [33], which is highly comparable to our detection a risk of viral load rebound above 50 copies/mL and above 400 rate of 23%. In our study, 2-LTRc DNA levels were lower, with copies/mL over 12  months of follow-up [20]. Taken together, longer duration of suppressive ART, with a reduction of 0.02 our findings suggest a cutoff level for plasma HIV-1 RNA below log 2-LTRc DNA copies/10 PBMC per year. This apparent, which a risk of viral load rebound is not observed. However, slow decay in 2-LTRc DNA is consistent with observations from an independent, direct association persisted between residual Bushman [34] and Siliciano [35], and potentially in line with plasma HIV-1 RNA and the immune activation marker sCD27 the hypothesis that episomal HIV-1 DNA survives for the life in this stably treated population. of the cell, rather than being a short-lived product of ongoing es Th e data do not elucidate the debated source of residual virus replication [36]. plasma HIV-1 RNA [15, 31]. The lack of a correlation with lev- Previous studies did not detect an association between els of 2-LTRc DNA, drug concentrations, or multiple markers residual plasma HIV-1 RNA and markers of immune activa- of immune activation may indicate that ongoing virus replica- tion on CD4 or CD8 cells (including CD38 and HLA-DR) in tion was an unlikely source, as far as it can be extrapolated from patients treated for up to 4 years [14, 37], and also detected no Residual plasma HIV-1 RNA 2-LTRc concentrations A B vs plasma EFV concentrations vs plasma EFV concentrations 40 120 rho = –0.01 rho = –0.07 P = .94 P = .54 0 0 0 1000 5000 10 000 15 000 0 1000 5000 10 000 15 000 EFV concentration, ng/mL EFV concentration, ng/mL Figure 2. Relationship between efavirenz (EFV) plasma concentrations and (A) residual plasma HIV-1 RNA and (B) 2-LTR circular HIV-1 DNA (2-LTRc DNA) among 76 patients who had taken the last EFV dose a median of 12 hours (interquartile range, 10–13) prior to sampling. Cutoffs in the x-axes represent the minimum effective concentration (MEC) recommended for wild-type virus (1000 ng/mL). The dotted black lines indicate the assay limit of detection (HIV-1 RNA = 3 copies/mL; 2-LTR DNA = 5 copies/10 peripheral blood mononuclear cell). Each circle represents 1 participant; the 2 solid squares represent the 2 patients with residual plasma HIV-1 RNA >11 copies/mL. 6 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 HIV-1 RNA, copies/mL 2-LTRc DNA, copies/10 PBMC association with levels of sCD14 and IL-6 [19]. Our data con- based on the infection stage at the start of ART, which could firm these observations over the longer term. However, we have introduced a bias. We observed a lower nadir CD4 cell found that levels of sCD27 were independently associated with count and a higher pre-ART viral load in subjects with longer levels of residual plasma HIV-1 RNA.  sCD27 is a 32-kD pro- duration of suppressive ART; however, the analyses did not tein identical to the extracellular domain of CD27, a transmem- identify an effect of nadir CD4 cell count or pre-ART viral load brane protein and TNF receptor that is expressed on subsets of on the measured parameters in our highly selected population. T, B, NK, and hematopoietic progenitor cells. sCD27 is released Of note, we excluded 2 outliers from the main analyses with primarily from CD4 T cells and provides a costimulatory func- residual plasma HIV-1 RNA levels >11 copies/mL. Whereas the tion that promotes activation and proliferation of T and B lym- main conclusions were not ae ff cted by their inclusion, it would phocytes [38–41]. sCD27 can be detected in healthy donors, have been of interest to determine any risk of viral load rebound and increased plasma levels have been observed in conditions with prospective follow-up in these 2 subjects. Further, we characterized by immune activation, especially auto-immune reported total HIV-1 DNA rather than integrated HIV-1 DNA disorders [42–44]. There are limited data in the setting of HIV as the indicator of the viral reservoir in this study. However, we infection. Levels of sCD27 have been shown to be generally previously described integrated HIV-1 DNA levels in a subset higher than in HIV-negative controls and found to be predic- of the same population and demonstrated that they were highly tive of the risk of HIV-related renal dysfunction and lympho- correlated with the levels of total HIV-1 DNA, not influenced mas [42, 45]. Levels have been shown to decline significantly in by the duration of suppressive ART, and not associated with the first 12 to 24 months of ART and to increase again with viral residual HIV-1 RNA [21]. We collected large volumes of blood load rebound or upon treatment discontinuation [29, 46, 47]. In from patients in order to test for multiple parameters (eg, 8 mL our study, consistent with the reported effect of treatment, levels of plasma was required for the ultrasensitive HIV-1 RNA assay of sCD27 were overall low and reduced further with longer dur- alone). Limited sample volumes meant that not all potentially ation of suppressive ART. Taken in the context of these previous relevant immune parameters were measured. When selecting observations, our findings illustrate a close association between soluble markers of immune activation, we elected to include 2 sCD27 and potential indicators of HIV replication during sup- markers that have been extensively explored in the published pressive ART that warrants further investigation. literature (sCD14 and IL-6) to provide a reference and added er Th e are scarce data on the effect of ART on sCD30 [26]. 2 markers for which there is high plausibility but scarce pub- Levels in our population were considerably lower than those lished evidence (sCD27 and sCD30). In this context, levels of measured in the first 4 weeks of ART [48], and closer to levels soluble markers of immune activation were found to be sim- published from HIV-negative volunteers [28]; however, sCD30 ilar to those historically reported in HIV-negative volunteers, showed no independent association with virologic parameters. which is consistent with the established beneficial effects of Levels of sCD14, IL-6, and cellular markers of immune activa- suppressive ART. However, published data from healthy volun- tion in our study population were comparable to those previ- teers were limited, especially for sCD27 and sCD30, and ideally, ously measured in HIV-1-positive patients receiving ART for matched HIV-negative controls should be sampled to confirm up to 6  years [19, 49]. Consistent with previous observations reference values. As an additional point, we measured plasma (reviewed in [16]), markers of immune activation on CD4 NNRTI concentrations in untimed specimens and for practical and CD8 cells (CD38 and HLA-DR/DP/DQ) were lower with reasons did not seek to obtain a measure of C . The patients trough longer duration of suppressive ART, and this apparent decline analyzed in the models had taken the last EFV dose at a consist- was paralleled by a predictable gain in CD4 cell counts. Overall, ent interval relative to sampling, and time of sampling would as also reported by Gandhi et al. [19], sCD14, IL6, and cellular be expected to have little impact overall. Finally, we recognize markers of immune activation showed no apparent association that both drug concentration and antiviral activity measured in with levels of residual plasma HIV-1 RNA and 2-LTRc DNA. peripheral blood do not necessarily reflect the situation in lym- Our study has limitations. First, findings obtained in a phoid tissue [8, 9]. cross-sectional study should be confirmed with prospectively In conclusion, this study provides evidence that residual collected data, which is challenging when aiming to study long- plasma HIV-1 RNA and PBMC-associated 2-LTRc DNA can term outcomes in populations maintained on the same regi- be detected in a subset of patients receiving long-term, con- men. To date, prospectively determined virologic indicators tinuously suppressive firstline ART, without evidence of a clear of HIV-1 persistence during long-term suppressive ART have association between the two, and without evidence of an asso- been poorly documented. Compared with published studies ciation of either parameter with plasma concentrations of EFV. addressing similar questions, one strength of our study is the While residual plasma HIV-1 RNA was not associated with relatively large population combined with stringent eligibility commonly reported markers of immune activation, there was criteria and a consistent definition of virologic suppression. a novel, independent direct association with levels of sCD27. It could be argued that our study population was not selected Further studies are warranted to confirm the findings, extend Residual HIV-1 RNA During ART • OFID • 7 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 5. Martinez-Picado J, Deeks SG. Persistent HIV-1 replication during antiretroviral the immune activation profiling, and determine the underlying therapy. Curr Opin HIV AIDS 2016; 11:417–23. mechanism, including whether the relationship reflects ongo- 6. Josefsson L, von Stockenstrom S, Faria NR, et  al. The HIV-1 reservoir in eight ing virus replication in lymphoid tissue. patients on long-term suppressive antiretroviral therapy is stable with few genetic changes over time. Proc Natl Acad Sci U S A 2013; 110:E4987–96. 7. Eriksson S, Graf EH, Dahl V, et al. Comparative analysis of measures of viral res- Supplementary Data ervoirs in HIV-1 eradication studies. PLoS Pathog 2013; 9:e1003174. Supplementary materials are available at Open Forum Infectious Diseases 8. Lorenzo-Redondo R, Fryer HR, Bedford T, et  al. Persistent HIV-1 replication maintains the tissue reservoir during therapy. Nature 2016; 530:51–6. online. Consisting of data provided by the authors to benefit the reader, 9. Fletcher CV, Staskus K, Wietgrefe SW, et al. Persistent HIV-1 replication is associ- the posted materials are not copyedited and are the sole responsibility of ated with lower antiretroviral drug concentrations in lymphatic tissues. Proc Natl the authors, so questions or comments should be addressed to the corre- Acad Sci U S A 2014; 111:2307–12. sponding author. 10. Nightingale S, Geretti AM, Beloukas A, et al. Discordant CSF/plasma HIV-1 RNA in patients with unexplained low-level viraemia. J Neurovirol 2016; 22:852–60. 11. Hatano H. Immune activation and HIV persistence: considerations for novel Acknowledgments therapeutic interventions. Curr Opin HIV AIDS 2013; 8:211–6. We are grateful to Dr. Giorgio Calisti (Central Manchester University 12. Cockerham LR, Siliciano JD, Sinclair E, et al. CD4+ and CD8+ T cell activation Hospitals NHS Foundation Trust, Manchester, UK), Dr. Paola Vitiello are associated with HIV DNA in resting CD4+ T cells. PLoS One 2014; 9:e110731. (Ospedali di Busto Arsizio, Varese, Italy), Ms. Helen Reynold (Royal 13. Sandler NG, Wand H, Roque A, et  al; INSIGHT SMART Study Group. Plasma Liverpool University Hospital, Liverpool, UK), and Ms. Anele Waters levels of soluble CD14 independently predict mortality in HIV infection. J Infect (Middlesex University Hospital, London, UK) for assistance with recruit- Dis 2011; 203:780–90. ments. We thank Ms. Tahami Fariba (Immunology Diagnostic Service, 14. Riddler SA, Aga E, Bosch RJ, et  al; ACTG A5276s Protocol Team. Continued slow decay of the residual plasma viremia level in HIV-1-infected adults receiving Royal Free Hospital, London, UK) for her support in producing the data on long-term antiretroviral therapy. J Infect Dis 2016; 213:556–60. cellular immune activation and Mr. Steven Lada (University of California 15. Mullins JI, Frenkel LM. Clonal expansion of human immunodeficiency virus-in- San Diego, San Diego, CA) for technical assistance with 2-LTRc DNA data fected cells and human immunodeficiency virus persistence during antiretroviral production. therapy. J Infect Dis 2017; 215:119–27. Author contributions. A.M.G.  and A.P.  designed the study. Technical 16. Klatt NR, Chomont N, Douek DC, Deeks SG. Immune activation and HIV persis- staff produced the laboratory work. A.R. analyzed the data under the super - tence: implications for curative approaches to HIV infection. Immunol Rev 2013; vision of A.C.L. and A.M.G. A.M.G. and A.R. wrote the manuscript, which 254:326–42. was reviewed by all authors. Members of the ERAS Study Group managed 17. Sereti I, Altfeld M. Immune activation and HIV: an enduring relationship. Curr recruitment. Opin HIV AIDS 2016; 11:129–30. 18. Khoury G, Fromentin R, Solomon A, et al. Human immunodeficiency virus per- Partecipating clinical centers (members of clinical study group). King’s sistence and T-cell activation in blood, rectal, and lymph node tissue in human College Hospital, London (Dr. Frank Post); North Middlesex Hospital, immunodeficiency virus-infected individuals receiving suppressive antiretroviral London (Dr. Jonathan Ainsworth); Royal Free Hospital, London and Royal therapy. J Infect Dis 2017; 215:911–9. Liverpool Hospital, Liverpool (Prof. Anna Maria Geretti); University 19. Gandhi RT, McMahon DK, Bosch RJ, et al; ACTG A5321 Team. Levels of HIV-1 College Hospital, London (Dr. Simon Edwards); St. Mary’s Hospital, persistence on antiretroviral therapy are not associated with markers of inflam- London (Dr. Nicola Mackie); St. Thomas’ Hospital, London (Dr. Julie Fox). mation or activation. PLoS Pathog 2017; 13:e1006285. Financial support. e s Th tudy was supported by research awards 20. Doyle T, Smith C, Vitiello P, et al. Plasma HIV-1 RNA detection below 50 copies/ from the European AIDS Treatment Network (NEAT), the British HIV ml and risk of virologic rebound in patients receiving highly active antiretroviral Association, the CARE and BEAT Martin Delaney Collaboratories of therapy. Clin Infect Dis 2012; 54:724–32. 21. Ruggiero A, De Spiegelaere W, Cozzi-Lepri A, et al; ERAS Study Group. During the National Institute of Allergy and Infectious Diseases (NIAID), the stably suppressive antiretroviral therapy integrated HIV-1 DNA load in periph- National Institute of Neurological Disorders and Stroke (NINDS), the eral blood is associated with the frequency of CD8 cells expressing HLA-DR/DP/ National Institute on Drug Abuse (NIDA) and the National Institute of DQ. EBioMedicine 2015; 2:1153–9. Mental Health (NIMH) of the National Institutes of Health (grant num- 22. Geretti AM, Arribas JR, Lathouwers E, et  al. Dynamics of cellular HIV-1 DNA bers 1UM1AI126619 and 1UM1Al126620), the University of California San levels over 144 weeks of darunavir/ritonavir monotherapy versus triple therapy in Diego Centre for AIDS Research (CFAR) (AI306214), the Department of the MONET trial. HIV Clin Trials 2013; 14:45–50. Veterans Affairs, and the James B. Pendleton Charitable Trust. The funding 23. Strain MC, Lada SM, Luong T, et al. Highly precise measurement of HIV DNA by sources had no role in the writing of the manuscript or the decision to sub- droplet digital PCR. PLoS One 2013; 8:e55943. mit for publication. The corresponding author (A.M.G.) had full access to 24. Poeta J, Linden R, Antunes MV, et al. Plasma concentrations of efavirenz are asso- ciated with body weight in HIV-positive individuals. J Antimicrob Chemother all the data in the study and had the final responsibility for the decision to 2011; 66:2601–4. submit for publication. 25. Havlir D, Cheeseman SH, McLaughlin M, et  al. High-dose nevirapine: safety, Potential coni fl cts of interest. A.M.G. is currently employed as Expert pharmacokinetics, and antiviral effect in patients with human immunodeficiency Scientist at Roche Pharma Research and Early Discovery (pRED); Roche virus infection. J Infect Dis 1995; 171:537–45. had no involvement in the work. The other authors have no conflicts of inter - 26. Uysal HK, Sohrabi P, Habip Z, et al. Neopterin and soluble CD14 levels as indica- est to declare. All authors have submitted the ICMJE Form for Disclosure of tors of immune activation in cases with indeterminate pattern and true positive Potential Conflicts of Interest. Conflicts that the editors consider relevant to HIV-1 infection. PLoS One 2016; 11:e0152258. the content of the manuscript have been disclosed. 27. Saing T, Valdivia A, Hussain P, et al. Data on pro-inflammatory cytokines IL-1β, IL-17, and IL-6 in the peripheral blood of HIV-infected individuals. Data Brief 2016; 8:1044–7. References 28. Gao R, Sun W, Chen Y, et al. Elevated serum levels of soluble CD30 in ankylosing 1. Guidelines E. EACS Guidelines Version 8.2. 2017. Available at: http://wwweacso- spondylitis patients and its association with disease severity-related parameters. cietyorg/files/guidelines_82-english.pdf. Accessed 22 August 2017. Biomed Res Int 2015; 2015:617282. 2. DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents – 29. De Milito A, Aleman S, Marenzi R, et al. Plasma levels of soluble CD27: a sim- A Working Group of the Office of AIDS Research Advisory Council (OARAC). ple marker to monitor immune activation during potent antiretroviral therapy in DASHH guidelines for the use of antiretroviral agents in HIV-1-infected adults HIV-1-infected subjects. Clin Exp Immunol 2002; 127:486–94. and adoloscencents. Updated 14 July 2016. Available at: https://aidsinfo.nih.gov/ 30. Marzolini C, Elzi L, Gibbons S, et  al; Swiss HIV Cohort Study. Prevalence of contentfiles/lvguidelines/adultandadolescentgl.pdf. Accessed 22 August 2017. comedications and effect of potential drug-drug interactions in the Swiss HIV 3. Palmer S, Maldarelli F, Wiegand A, et  al. Low-level viremia persists for at least Cohort Study. Antivir Ther 2010; 15:413–23. 7 years in patients on suppressive antiretroviral therapy. Proc Natl Acad Sci U S A 31. Hong FF, Mellors JW. Changes in HIV reservoirs during long-term antiretroviral 2008; 105:3879–84. therapy. Curr Opin HIV AIDS 2015; 10:43–8. 4. Doyle T, Geretti AM. Low-level viraemia on HAART: significance and manage- 32. Maldarelli F, Palmer S, King MS, et al. ART suppresses plasma HIV-1 RNA to a ment. Curr Opin Infect Dis 2012; 25:17–25. stable set point predicted by pretherapy viremia. PLoS Pathog 2007; 3:e46. 8 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 33. Besson GJ, Lalama CM, Bosch RJ, et  al. HIV-1 DNA decay dynamics in blood 42. Widney D, Gundapp G, Said JW, et  al. Aberrant expression of CD27 and sol- during more than a decade of suppressive antiretroviral therapy. Clin Infect Dis uble CD27 (sCD27) in HIV infection and in AIDS-associated lymphoma. Clin 2014; 59:1312–21. Immunol 1999; 93:114–23. 34. Butler SL, Johnson EP, Bushman FD. Human immunodeficiency virus cDNA 43. van der Vuurst de Vries RM, Mescheriakova JY, Runia TF, et  al. Soluble CD27 metabolism: notable stability of two-long terminal repeat circles. J Virol 2002; levels in cerebrospinal fluid as a prognostic biomarker in clinically isolated syn- 76:3739–47. drome. JAMA Neurol 2017; 74:286–92. 35. Pierson TC, Kieffer TL, Ruff CT, et al. Intrinsic stability of episomal circles formed 44. Speeckaert R, Lambert J, van Geel N. Clinical significance of serum solu- during human immunodeficiency virus type 1 replication. J Virol 2002; 76:4138–44. ble CD molecules to assess disease activity in vitiligo. JAMA Dermatol 2016; 36. Sharkey ME, Teo I, Greenough T, et al. Persistence of episomal HIV-1 infection inter- 152:1194–200. mediates in patients on highly active anti-retroviral therapy. Nat Med 2000; 6:76–81. 45. Abraham AG, Darilay A, McKay H, et  al. Kidney dysfunction and mark- 37. Hatano H, Jain V, Hunt PW, et  al. Cell-based measures of viral persistence are ers of inflammation in the multicenter AIDS Cohort Study. J Infect Dis 2015; associated with immune activation and programmed cell death protein 1 (PD-1)- 212:1100–10. expressing CD4+ T cells. J Infect Dis 2013; 208:50–6. 46. Atlas A, Thanh Ha TT, Lindström A, et al. Effects of potent antiretroviral therapy 38. Nolte MA, van Olffen RW, van Gisbergen KP, van Lier RA. Timing and tuning of on the immune activation marker soluble CD27 in patients infected with HIV-1 CD27-CD70 interactions: the impact of signal strength in setting the balance between subtypes A-D. J Med Virol 2004; 72:345–51. adaptive responses and immunopathology. Immunol Rev 2009; 229:216–31. 47. Wallet MA, Rodriguez CA, Yin L, et al. Microbial translocation induces persistent 39. van Oosterwijk MF, Juwana H, Arens R, et al. CD27-CD70 interactions sensitise naive macrophage activation unrelated to HIV-1 levels or T-cell activation following CD4+ T cells for IL-12-induced Th1 cell development. Int Immunol 2007; 19:713–8. therapy. AIDS 2010; 24:1281–90. 40. Hintzen RQ, de Jong R, Hack CE, et al. A soluble form of the human T cell dif- 48. Kwamena WCS, Kwabena O, Makafui S, et al. Decay of soluble CD30 and HIV-1 ferentiation antigen CD27 is released after triggering of the TCR/CD3 complex. plasma viral load during early highly active antiretroviral therapy: A short-term J Immunol 1991; 147:29–35. longitudinal study. J Microbiol Biochem Technol 2016; 8(2):090–6: 90. 41. Huang J, Jochems C, Anderson AM, et  al. Soluble CD27-pool in humans 49. Borges ÁH, O’Connor JL, Phillips AN, et al; INSIGHT SMART and ESPRIT Study may contribute to T cell activation and tumor immunity. J Immunol 2013; Groups and the SILCAAT Scientific Committee. Factors associated with plasma 190:6250–8. IL-6 levels during HIV infection. J Infect Dis 2015; 212:585–95. Residual HIV-1 RNA During ART • OFID • 9 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Open Forum Infectious Diseases Oxford University Press

Factors Associated With Persistence of Plasma HIV-1 RNA During Long-term Continuously Suppressive Firstline Antiretroviral Therapy

Free
9 pages

Loading next page...
 
/lp/ou_press/factors-associated-with-persistence-of-plasma-hiv-1-rna-during-long-FKH1Ud8HRh
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Infectious Diseases Society of America.
eISSN
2328-8957
D.O.I.
10.1093/ofid/ofy032
Publisher site
See Article on Publisher Site

Abstract

Open Forum Infectious Diseases MAJOR ARTICLE Factors Associated With Persistence of Plasma HIV-1 RNA During Long-term Continuously Suppressive Firstline Antiretroviral Therapy 1 2 1 3 4 2 1 Alessandra Ruggiero, Alessandro Cozzi-Lepri, Apostolos Beloukas, Douglas Richman, Saye Khoo, Andrew Phillips, and Anna Maria Geretti ; on behalf of the ERAS Study Group 1 2 Institute of Infection and Global Health, University of Liverpool, Liverpool, United Kingdom; Department of Infection and Population Health, University College London, London, United Kingdom; 3 4 VA San Diego Healthcare System and Center for AIDS Research, University of California San Diego, La Jolla, California; Institute of Translational Medicine, University of Liverpool, Liverpool, United Kingdom Background. Persistence of plasma HIV-1 RNA during seemingly effective antiretroviral thereapy (ART) is incompletely under - stood. Using an ultrasensitive assay, this cross-sectional study investigated residual plasma HIV-1 RNA in subjects maintained on firstline ART with continuous viral load suppression <50 copies/mL for ≤15 years without recognized viral load blips or treatment interruptions and explored its relationship with the duration of suppressive ART, efavirenz concentrations in plasma, 2-LTR circular HIV-1 DNA (2-LTRc DNA) in peripheral blood mononuclear cells, and cellular (CD4 plus CD26/CD38/CD69; CD8 plus CD38/ HLA-DR/DP/DQ) and soluble (sCD14, sCD27, sCD30, IL-6) markers of immune activation in peripheral blood.  Methods. Residual plasma HIV-1 RNA, total HIV-1 DNA and 2-LTRc DNA were quantified by real-time and digital droplet PCR. Cellular (CD4 plus CD26/CD38/CD69; CD8 plus CD38/HLA-DR/DP/DQ) and soluble (sCD14, sCD27, sCD30, IL-6) mark- ers of immune activation were measured by flow cytometry and ELISA. Results. Residual plasma HIV-1 RNA and 2-LTRc DNA were detected in 52/104 (50%) and 24/104 (23%) subjects, respectively. Among subjects with detectable HIV-1 RNA, 50/52 showed levels ≤11 copies/mL. In adjusted analyses, HIV-1 RNA levels were 0.37 log copies/mL higher with each log U/mL increase in sCD27 (95% confidence interval, 0.01–0.73; P = .02). No significant associ- 10 10 ation was found between residual plasma HIV-1 RNA and other explored parameters.  Conclusions. es Th e findings point to an ongoing relationship between plasma HIV-1 RNA and selected markers of immune activation during continuously suppressive ART. The novel direct association with levels of sCD27 warrants further investigation. Keywords. 2-LTR circular DNA; activation; drug concentration; sCD27; viral load. Maintaining plasma HIV-1 RNA suppression below the limit of heterogeneous treatment histories and varying definitions of detection of commercial viral load assays defines clinical suc- suppression, which may in part explain conflicting findings cess of antiretroviral therapy (ART) [1, 2]. Ultrasensitive assays [11–13]. can detect traces of HIV-1 RNA in the plasma of many success- Identifying the determinants of HIV-1 RNA persistence fully treated patients [3], but the source remains controversial during seemingly effective ART is central to our understand- [4, 5]. On the one hand, it is proposed that detection of residual ing of how the virus may be eradicated [14, 15]. Such studies plasma HIV-1 RNA reflects short-lived bursts of virus produc- must also consider the interplay between virus replication and tion from latently infected memory CD4 T cells, but with ongo- host immunity, and the role of chronic immune activation and ing virus replication and genetic evolution inhibited by ART [6]. inflammation in disease pathogenesis [16–18]. Effective ART An alternative model derives from evidence of ongoing virus reduces but does not always normalize inflammatory mark- replication in compartments where drug penetration or activ- ers, raising the possibility that residual immune activation may ity is suboptimal [7–10]. While the 2 models are not mutually both sustain and be sustained by residual virus replication [18]. exclusive, published studies have oen in ft cluded patients with Available data are not conclusive, however, and they rather point to virologic and immunologic events that take place before the initiation of ART as key determinants of the HIV-1 reservoirs Received 6 December 2017; editorial decision 25 January 2018; accepted 2 February 2018. and immune activation that can persist despite ART [19]. Correspondence: A. M. Geretti, MD, PhD, Department of Clinical Infection, Microbiology and We previously reported that treated patients with plasma Immunology, Institute of Infection and Global Health, University of Liverpool, 8 West Derby Street, Liverpool L69 7BE, UK (geretti@liverpool.ac.uk). HIV-1 RNA detected at levels ranging between approximately Open Forum Infectious Diseases 10 and 49 copies/mL by a commercial viral load assay were more © The Author(s) 2018. Published by Oxford University Press on behalf of Infectious Diseases likely to experience viral load rebound above 50 copies/mL and Society of America. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits above 400 copies/mL during 12  months of follow-up when unrestricted reuse, distribution, and reproduction in any medium, provided the original work compared with treated patients with undetectable HIV-1 RNA is properly cited. [20]. e Th aim of this study was to take the converse approach DOI: 10.1093/ofid/ofy032 Residual HIV-1 RNA During ART • OFID • 1 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 and assess a population that experienced continuous viral load Following centrifugation, cell pellets were lysed by 10-minute suppression <50 copies/mL for up to 15 years of firstline ART incubation at room temperature in 1× Hoffman’s lysis buffer and use ultrasensitive testing to detect HIV-1 RNA below the and incubated with FITC-, Cy5- or PE-labeled antibodies tar- 10-copies/mL cuto. Th ff e cohort was selected according to pre- geting CD4 plus CD26, CD38, or CD69, and CD8 plus CD38 or cise inclusion criteria, and residual plasma HIV-1 RNA was HLA-DR/DP/DQ (Becton Dickinson, Swindon, UK). Labeled investigated cross-sectionally in relation to the duration of sup- cells were incubated in 1% paraformaldehyde for 30 minutes pressive ART, plasma concentrations of efavirenz, 2-LTRc DNA prior to acquisition with BD FACSCalibur and analysis by in peripheral blood mononuclear cells (PBMCs), and cellular CellQuest. Soluble markers were measured in frozen plasma and soluble markers of immune activation in peripheral blood. by commercial Elisa assays (EIAs) for soluble CD14 (sCD14, −6 R&D system, London, UK; assay sensitivity ≥125 ×  10 µg/ METHODS mL), soluble CD27 (sCD27, EBiosciences; ThermoFisher, Loughborough, UK; assay sensitivity ≥0.2 U/mL), soluble CD30 Study Population (sCD30, EBiosciences, UK; assay sensitivity ≥0.3  ng/mL), and In this cross-sectional study, eligible patients had started firstline IL6 (EBiosciences, UK, assay sensitivity ≥0.9 pg/mL). ART with 2 nucleoside/nucleotide reverse transcriptase inhibi- tors (NRTIs) plus 1 non-nucleoside reverse transcriptase inhib- Plasma Drug Concentrations itor (NNRTI), had achieved viral load suppression <50 copies/ Efavirenz (EFV) and nevirapine (NVP) concentrations were mL within 6 months of starting ART, and during subsequent fol- measured in untimed plasma samples using high-performance low-up had remained on the initial NNRTI, undergone ≥2 viral liquid chromatography–tandem mass spectrometry, as previ- load measurements per year, and maintained continuous viral ously described [10]. The time of sample collection relative to load suppression <50 copies/mL without any recorded viral load last dosing was recorded. elevation >50 copies/mL or treatment interruption. To ensure balance in terms of duration of suppressive ART, recruitment was Analysis stratified to span from 1 year of treatment to 10 years or longer. Duration of suppressive ART was defined as the length of time Changes of the initial NNRTI were not allowed. Changes of following the first viral load measurement <50 copies/mL, which the initial NRTIs were allowed to reflect evolving practice, pro- in all subjects was recorded within 6  months of starting ART. vided they did not coincide with a viral load >50 copies/mL or a When analyzing residual plasma HIV-1 RNA and 2-LTRc DNA treatment interruption. The study took place at multiple centers as continuous variables, undetectable values were replaced with across the United Kingdom. It was approved by the National the midpoint value between 0 and the lower limit of detection Research Ethics Service (London-Dulwich) and included in the of the assays. EFV and NVP concentrations were categorized as National Institute for Health Research Clinical Research Network above or below the recommended minimal effective concentra- Portfolio. Patients provided written informed consent. tion (MEC) for wild-type virus: 1000  ng/mL for EFV [24] and 3400 ng/mL for NVP [25]. Participants’ characteristics were com- Residual Plasma HIV-1 RNA pared by analysis of variance (ANOVA), the Kruskal-Wallis test, Plasma was separated from venous blood in EDTA within 2 and the Fisher exact test for continuous and categorical variables, hours of collection and stored at −80°. HIV-1 RNA was quan- as appropriate. Data distribution was assessed through descrip- tified using a modified version of the RealTime HIV-1 assay tive scatter plots. The correlation between EFV concentrations (Abbott, Maidenhead, UK). Validation studies previously and levels of residual plasma HIV-1 RNA and 2-LTRc DNA was showed that the assay 50% and 95% detection rates were 1 and 3 measured by Spearman’s rank test. Given that all but 2 HIV-1 HIV-1 RNA copies/mL, respectively, with high reproducibility RNA values were ≤11 copies/mL, in order to reduce heterogene- across diverse HIV-1 subtypes [10, 20, 21]. ity, the main statistical models excluded 2 subjects with residual Total HIV-1 DNA and 2-LTRc DNA plasma HIV-1 RNA >11 copies/mL (referred to as outliers). This PBMCs were isolated by Ficoll-Hypaque gradient centrifuga- was based on our previous data indicating that patients on stable tion and cryopreserved. Total HIV-1 DNA was quantified by ART with a viral load above approximately 10 copies/mL are a real-time polymerase chain reaction (PCR) as described [22]; distinct population at risk of viral load rebound [20]. Three main the assay 50% and 95% detection rates were 20 and 40 HIV-1 models were run. The mean difference (with 95% confidence DNA copies/10 PBMC, respectively. 2-LTRc DNA was meas- interval [CI]) in residual plasma HIV-1 RNA, total HIV-1 DNA, ured by droplet digital PCR as described [23]; the assay’s lower 2-LTRc DNA, and markers of immune activation according to limit of detection was a median (interquartile range [IQR]) of 5 duration of suppressive ART (normalized over 1 year) was ana- (4–6) copies/10 PBMC.  lyzed by univariable linear regression analysis after log-transfor- Cellular and Soluble Markers of Immune Activation mation of the variables. Factors associated with levels of residual Flow cytometry was performed at the Immunology Service of plasma HIV-1 RNA or 2-LTRc DNA were investigated by uni- the Royal Free Hospital in London using freshly collected blood. variable and multivariable linear regression analysis. Variables 2 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 associated (P ≤ .20) with the outcome of interest in the univaria- <50 copies/mL for a median (IQR) of 5 (3–8) years while receiv- ble analysis were included in the multivariable models, with the ing firstline ART with 2 NRTIs plus either EFV (n = 86, 83%) following modifications: (a) when modeling 2-LTRc DNA, total or NVP (Table 1). Reflecting evolving guidelines about when to HIV-1 DNA was excluded due to high colinearity (P  <  .0001); start ART and which antiretrovirals to use, patients with longer (b) age and duration of ART were excluded in favor of duration duration of suppressive ART had a lower nadir CD4 cell count of suppressive ART. Separate models explored: (a) the effect of and a higher pre-ART viral load, were more often receiving forcing the nadir CD4 cell count and pre-ART viral load into the NVP, and were more likely to have experienced changes of the multivariable analyses and (b) the association between residual NRTI backbone after first starting ART: 48/104 (48%) patients plasma HIV-1 RNA or 2-LTRc DNA and plasma concentrations changed ≥1 NRTIs prior to recruitment, with a median (IQR) of of EFV (as a continuous variable) in EFV recipients, after adjust- 1 (1–2) NRTI change per subject. At recruitment, NRTI back- ment for the time of sampling in relation to the last EFV dose. bones comprised tenofovir disoproxil fumarate (78/104, 75%), IBM SPSS v. 22.0 (Armonk, NY, USA) and SAS v9.4 (Cary, NC, abacavir (22/104, 21%) or zidovudine (4/104, 4%), plus emtric- USA) were used for the analyses. itabine or lamivudine. Residual plasma HIV-1 RNA was detected in 52/104 (50%) RESULTS patients (Table  1). Levels ranged between 1 and 35 copies/ mL and were ≤11 copies/mL in 50/52 (96%) patients; levels in Study Population the 2 outliers were 35 and 24 copies/mL aer 2 a ft nd 3 years of The population comprised 104 patients that at the time of suppressive ART, respectively (Figure 1). 2-LTRc HIV-1 DNA recruitment had experienced continuous viral load suppression Table 1. Characteristics of the Study Population Overall and by Duration of Suppressive ART (n = 104) Duration of Suppressive ART, y Total (n = 104) <4 (n = 34) 4–7 (n = 35) >7 (n = 35) Duration of suppressive ART, median (IQR), y 5 (3–8) 2 (2–3) 5 (5–6) 9 (8–10) Age, median (IQR), y 47 (40–53) 44 (36–49) 49 (39–53) 48 (43–55) Male, n (%) 81 (78) 29 (85) 29 (83) 23 (66) Ethnicity, n (%) White 59 (57) 20 (59) 22 (63) 17 (49) Black 41 (39) 13 (38) 11 (31) 17 (49) Asian 4 (4) 1 (3) 2 (6) 1 (3) Pre-ART viral load, median (IQR), log cps/mL 4.9 (4.6–5.3) 4.7 (4.1–5.1) 4.9 (4.6–5.3) 5.1 (4.8–5.6) Nadir CD4 count, median (IQR), cells/mm 201 (110–270) 279 (216–365) 166 (126–238) 118 (79–215) EFV use, n (%) 86 (83) 30 (88) 32 (89) 24 (69) NNRTI concentration, median (IQR), ng/mL EFV 1531 (846–2235) 1492 (735–2130) 1496 (866–2154) 1966 (855–2626) NVP 5889 (3096–7419) 4545 (2798–6147) 2734 (2539–5743) 6311 (4390–7603) NNRTI concentration above target MEC , n (%) 71 (66) 22 (65) 24 (69) 25 (71) Changed initial NRTIs, n (%) 48 (46) 4 (12) 13 (37) 31 (89) Detectable residual plasma HIV-1 RNA, n (%) 52 (50) 21 (62) 18 (51) 13 (37) Residual plasma HIV-1 RNA, median (IQR), cps/mL 2 (2–4) 2 (2–4) 2 (2–4) 2 (2–2) Detectable 2-LTRc DNA, n (%) 24 (23) 12 (35) 7 (20) 5 (14) d 6 2-LTRc DNA, median (IQR), cps/10 PBMC 3 (3–3) 3 (3–7) 3 (3–3) 3 (3–3) Total HIV-1 DNA, median (IQR), log cps/10 PBMC 2.5 (2.1–2.8) 2.6 (2.3–3.0) 2.4 (1.7–2.8) 2.5 (2.1–2.8) CD4 count, median (IQR), cells/mm 581 (474–722) 548 (427–629) 580 (509–720) 632 (503–796) CD8 count, median (IQR), cells/mm 798 (656–1047) 808 (596–1060) 818 (685–1024) 679 (496–929) + + CD4 CD38 , median (IQR), % 24 (16–34) 32 (23–36) 21 (15–31) 22 (15–28) + + CD4 CD26 , median (IQR), % 58 (46–64) 56 (47–67) 59 (49–63) 58 (44–64) + + CD4 CD69 , median (IQR), % 1 (1–2) 2 (1–3) 1 (1–2) 2 (1–2) + + CD8 CD38 , median (IQR), % 5 (3–7) 6 (4–8) 4 (3–6) 4 (3–6) + + CD8 HLA-DR/DP/DQ , median (IQR), % 35 (24–42) 37 (27–47) 35 (26–40) 30 (20–39) 2.2 (1.8–2.5) 2.2 (1.8–2.4) 2.0 (1.8–2.5) 2.3 (1.9–2.5) sCD14, median (IQR), µg/mL sCD27, median (IQR), Ug/mL 62 (49–76) 64 (52–80) 61 (54–76) 60 (46–71) sCD30, median (IQR), ng/mL 11 (9–15) 11 (9–15) 12(10–17) 11 (9–13) IL6 median (IQR), pg/mL 0.8 (0.5–2.1) 0.5 (0.5–1.2) 1.1 (0.5–2.3) 0.8 (0.5–2.1) Abbreviations: 2-LTRc DNA, 2-LTR circular HIV-1 DNA; ART, antiretroviral therapy; cps, copies; EFV, efavirenz; IQR, interquartile range; MEC, minimum effective concentration; NNRTI, non-nu- cleoside reverse transcriptase inhibitor; NRTI, nucleoside/nucleotide reverse transcriptase inhibitor; NVP, nevirapine; PBMC, peripheral blood mononuclear cell. Defined as the length of time between the first viral load <50 copies/mL measured after starting ART and the date of recruitment. Minimum effective concentration recommended for wild-type virus (1000 ng/mL for EFV and 3400 ng/mL for NVP). Samples with undetectable HIV-1 RNA were assigned an arbitrary value of 1.5 copies/mL.  d 6 Samples with undetectable 2-LTRc DNA were assigned an arbitrary value of 2.5 copies/10 PBMC. Residual HIV-1 RNA During ART • OFID • 3 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 A Residual plasma HIV-1 RNA B 2-LTRc DNA 0 0 01 23 45 67 89 10 111213 1415 01 2 3456789 101112131415 Duration of suppressive ART, years Duration of suppressive ART, years C D Total HIV-1 DNA sCD27 0 12345678 9101112131415 05 10 15 Duration of suppressive ART, years Duration of suppressive ART, years Figure 1. Virologic and immunologic parameters in patients with up to 15 years of consistently suppressive antiretroviral therapy (ART; n = 104), comprising (A) residual plasma HIV-1 RNA (copies/mL), (B) 2-LTR circular HIV-1 DNA (2-LTRc DNA; copies per 10 peripheral blood mononuclear cell [PBMC]), (C) total HIV-1 DNA (log copies per 10 PBMC), and (D) sCD27 (U/mL) in relation to the duration of suppressive ART (years). In (D), the dotted line indicates the mean sCD27 levels reported in HIV-negative volunteers in published studies (164 U/mL) [25]; the solid lines indicate the median values measured in the whole study population. Each circle represents 1 participant; the 2 solid squares represent the 2 patients with residual plasma HIV-1 RNA >11 copies/mL. Virologic and Immunologic Parameters in Relation to the Duration of was detected in 24/104 (23%) patients with median levels of 3 6 6 Suppressive ART copies/10 PBMC that ranged between 4 and 119 copies/10 Differences in virologic and immunologic parameters PBMC (Figure 1). Cellular (Supplementary Figure 1) and sol- according to the duration of suppressive ART, analyzed after uble (Figure  1; Supplementary Figure  2) markers of immune exclusion of the 2 outliers, are shown in Table 2. Patients with activation were detected in all patients. Levels of soluble mark- longer duration of suppressive ART had higher CD4 counts ers were generally not dissimilar from those of HIV-negative (+19 cells/mm /y) and lower expression of activation mark- volunteers reported in published studies [26–29]. When ana- ers on CD4 cells (CD38) and CD8 cells (CD38, HLA-DR/ lyzing subjects whose levels of soluble markers fell above the DP/DQ). Levels of sCD14, sCD30, and IL-6 did not differ by upper quartile of the whole study population, a few patients duration of suppressive ART, whereas levels of sCD27 were presented concomitant elevations, but with little evidence significantly lower with longer duration of suppressive ART. of consistent patterns. Overall 5/104 (5%) patients showed A  trend was observed for lower levels of 2-LTRc DNA with concomitant elevations of >2 soluble markers, comprising longer duration of suppressive ART. There was less evidence sCD14 + sCD27 + sCD30 (3/104, 3%); CD27 + sCD30 + IL-6 of a difference in residual plasma HIV-1 RNA, and no sig- (1/104, 1%); and sCD14 + sCD27 + sCD30 + IL-6 (1/104, 1%) nificant difference was seen in total HIV-1 DNA. A  linear (Supplementary Table  1). CD8 cells counts and frequency of + + regression analysis confirmed an inverse association between CD8 HLA-DR/DP/DQ cells were oen, a ft lbeit not always, 2-LTRc DNA and duration of suppressive ART, with an increased in these patients, but there were no apparent patterns adjusted mean difference of 0.03 log copies/10 PBMC per in other parameters. 4 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Total HIV-1 DNA, HIV-1 RNA, copies/mL log copies/10 PBMC 2-LTRc HIV-1 DNA, sCD27, U/mL copies/10 PBMC Table  2. Univariable Analysis of the Mean Difference in Virologic and Factors Associated With Residual Plasma HIV-1 RNA Immunologic Parameters per 1 Year of Suppressive ART (n = 102) The univariable linear regression analysis indicated a signif- icant direct association between residual plasma HIV-1 RNA Mean b and sCD27 among the 102 subjects with HIV-1 RNA ≤11 cop- Difference 95% CI P ies/mL (Table  3). There was a weaker direct association with Residual plasma HIV-1 RNA, –0.013 –0.030 to 0.004 .12 cps/mL sCD30, whereas no association was evident with other markers 2-LTRc DNA, cps/10 PBMC –0.018 –0.039 to 0.003 .09 of immune activation. There was also no association between Total HIV-1 DNA, cps/10 PBMC 0.004 –0.033 to 0.042 .82 residual plasma HIV-1 RNA and total HIV-1 DNA and 2-LTRc CD4 count, cells/mm 0.015 0.004 to 0.026 .01 DNA. The direct association between residual plasma HIV-1 CD8 count, cells/mm –0.004 –0.018 to –0.009 .52 + + RNA and sCD27 persisted after adjusting for the duration of CD4 CD38 , % –0.022 –0.034 to –0.009 .001 + + + + suppressive ART, frequency of CD8 CD38 cells, and levels of CD4 CD26 , % –0.003 –0.012 to 0.005 .44 + + CD4 CD69 , % –0.008 –0.021 to 0.005 .21 sCD30 and IL-6 (Table  3). There was no apparent association + + CD8 CD38 , % –0.014 –0.032 to 0.004 .05 between residual plasma HIV-1 RNA and nadir CD4 cell count + + CD8 HLA-DR/DP/DQ , % –0.014 –0.027 to 0.001 .04 and pre-ART viral load in the univariable analysis. Forcing the sCD14, µg/mL 0.002 –0.004 to 0.009 .50 nadir CD4 cell count and pre-ART viral load into the multivar- sCD27, U/mL –0.010 –0.019 to 0.000 .05 iable model did not change the findings: residual plasma HIV-1 sCD30, ng/mL –0.005 –0.019 to 0.009 .44 RNA was 0.44 log copies/mL higher per each log increase IL-6, pg/mL 0.003 –0.015 to 0.022 .72 10 10 in sCD27 (95% CI, 0.09–0.79; P  =  .02; not shown). The asso- Abbreviations: 2-LTRc DNA,  2-LTR circular HIV-1 DNA; CI,  confidence interval; PBMC, peripheral blood mononuclear cell. ciation similarly persisted in analyses including the 2 outliers The analysis excluded the 2 outliers with residual plasma HIV-1 RNA >11 copies/mL.  (not shown). Variables were log-transformed prior to the analysis. Relationship Between Virologic Parameters and Plasma Drug Concentrations year longer (95% CI, –0.05 to –0.00; P = .02) (Supplementary We first explored the association between plasma EFV concen- Table 2). Forcing the nadir CD4 cell count and pre-ART viral trations and virologic parameters in 76 EFV recipients (includ- load into the multivariable model did not change the find- ing the 2 outliers) who had taken the last EFV dose a median ings: 2-LTRc DNA levels were 0.03 log copies/10 PBMC (IQR) of 12 (10–13) hours prior to sampling, and were there- lower per year of suppressive ART (95% CI, –0.05 to –0.00; fore in the range of the “mid-dose” interval, where targets have P = .02; not shown). been previously defined [30]. Although EFV concentrations Table 3. Univariable and Multivariable Linear Regression Analysis of Factors Associated With Mean Differences in Levels of Residual Plasma HIV-1 RNA (n = 102) Univariate Multivariable Mean Difference 95% CI P Mean Difference 95% CI P Nadir CD4 count per 100 cell/mm higher 0.01 –0.06 to 0.05 .58 Pre-ART viral load per log cps/mL higher 0.00 –0.08 to 0.08 .99 Duration of suppressive ART per 1 y longer –0.01 –0.03 to 0.00 .12 –0.01 –0.02 to 0.01 .39 CD4 count per 100 cells/mm higher –0.01 –0.02 to 0.02 .92 CD4/CD8 ratio per 1 unit higher 0.04 –0.11 to 0.17 .62 2-LTRc DNA per log cps/10 PBMC higher 0.05 –0.11 to 0.20 .57 Total HIV-1 DNA per log cps/10 PBMC higher –0.01 –0.09 to 0.08 .87 + + CD4 CD38 per 50% higher 0.08 –0.14 to 0.30 .47 + + CD4 CD26 per 50% higher 0.08 –0.10 to 0.27 .37 + + CD4 CD69 per 50% higher –0.61 –2.97 to 1.75 .60 CD8 count per 100 cells/mm higher 0.02 –0.01 to 0.01 .70 + + CD8 CD38 per 50% higher 0.32 –0.11 to 0.75 .16 0.26 –0.17 to 0.68 .24 + + CD8 HLA-DR/DP/DQ per 50% higher –0.03 –0.21 to 0.15 .76 sCD14 per log µg /mL higher 0.26 –0.24 to 0.75 .31 sCD27 per log U/mL higher 0.49 0.17 to 0.81 .003 0.37 0.01 to 0.73 .02 sCD30 per log ng/mL higher 0.21 –0.02 to 0.45 .07 0.02 –0.25 to 0.28 .88 IL6 per log pg/mL higher 0.13 –0.05 to 0.30 .15 0.12 –0.06 to 0.29 .19 Abbreviations: 2-LTRc DNA, 2-LTR circular HIV-1 DNA; ART, antiretroviral therapy; CI, confidence interval; cps, copies; PBMC, peripheral blood mononuclear cell. The analysis excluded the 2 outliers with residual plasma HIV-1 RNA >11 copies/mL.  Mean difference in log copies/mL.  Variables with P < .2 in the univariable analysis were included in the multivariable analysis. Residual HIV-1 RNA During ART • OFID • 5 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 varied across patients, there was no clear correlation with levels measurements made in peripheral blood. In this context, the of residual plasma HIV-1 RNA and 2-LTRc DNA (Spearman’s findings may provide support to the notion that during effect- rank test) (Figure  2). Two separate linear regression analy- ive treatment, detection of a few HIV-1 RNA copies in plasma ses explored the association between EFV concentrations and reflects transient bursts of virus production from latent reser - residual plasma HIV-1 RNA and 2-LTRc DNA among the 74 voirs that are inhibited from ongoing replication by ART [8, 32]. patients with residual plasma HIV-1 RNA ≤11 copies/mL. In In this model, a direct correlation may be anticipated between models that included both EFV concentration and timing of the likelihood of HIV-1 RNA persistence and the size of the sampling relative to last EFV dose, there was no evidence of an viral reservoir. We did not detect an association between resid- association between plasma HIV-1 RNA (P  =  .30) or 2-LTRc ual HIV-1 RNA in plasma and total HIV-1 DNA in PBMC, and DNA (P  =  .47) and EFV concentrations (Supplementary we similarly did not previously observe an association with inte- Table 3). grated HIV-1 DNA in PBMC [21]. Findings from other studies are conflicting, possibly reflecting differences in the compos- DISCUSSION ition of the study populations (eg, proportion of patients on This study demonstrated that half of subjects who had expe- protease inhibitors) and the method used for measuring HIV-1 rienced viral load suppression <50 copies/mL continuously RNA [33]. We also measured 2-LTRc DNA in PBMC as a pos- for up to 15 years while maintained on the initial NNRTI had sible indicator of viral gene expression and virus replication detectable HIV-1 RNA in plasma when tested with an ultra- and found no correlation with residual plasma HIV-1 RNA. In sensitive assay, at levels that rarely exceeded 11 copies/mL. We a previous study of 30 subjects who had received various ART previously reported that in treated patients, plasma HIV-1 RNA regimens for 7–12  years, 2-LTRc DNA was detected in 8 sub- levels above approximately 10 copies/mL were associated with jects (27%) [33], which is highly comparable to our detection a risk of viral load rebound above 50 copies/mL and above 400 rate of 23%. In our study, 2-LTRc DNA levels were lower, with copies/mL over 12  months of follow-up [20]. Taken together, longer duration of suppressive ART, with a reduction of 0.02 our findings suggest a cutoff level for plasma HIV-1 RNA below log 2-LTRc DNA copies/10 PBMC per year. This apparent, which a risk of viral load rebound is not observed. However, slow decay in 2-LTRc DNA is consistent with observations from an independent, direct association persisted between residual Bushman [34] and Siliciano [35], and potentially in line with plasma HIV-1 RNA and the immune activation marker sCD27 the hypothesis that episomal HIV-1 DNA survives for the life in this stably treated population. of the cell, rather than being a short-lived product of ongoing es Th e data do not elucidate the debated source of residual virus replication [36]. plasma HIV-1 RNA [15, 31]. The lack of a correlation with lev- Previous studies did not detect an association between els of 2-LTRc DNA, drug concentrations, or multiple markers residual plasma HIV-1 RNA and markers of immune activa- of immune activation may indicate that ongoing virus replica- tion on CD4 or CD8 cells (including CD38 and HLA-DR) in tion was an unlikely source, as far as it can be extrapolated from patients treated for up to 4 years [14, 37], and also detected no Residual plasma HIV-1 RNA 2-LTRc concentrations A B vs plasma EFV concentrations vs plasma EFV concentrations 40 120 rho = –0.01 rho = –0.07 P = .94 P = .54 0 0 0 1000 5000 10 000 15 000 0 1000 5000 10 000 15 000 EFV concentration, ng/mL EFV concentration, ng/mL Figure 2. Relationship between efavirenz (EFV) plasma concentrations and (A) residual plasma HIV-1 RNA and (B) 2-LTR circular HIV-1 DNA (2-LTRc DNA) among 76 patients who had taken the last EFV dose a median of 12 hours (interquartile range, 10–13) prior to sampling. Cutoffs in the x-axes represent the minimum effective concentration (MEC) recommended for wild-type virus (1000 ng/mL). The dotted black lines indicate the assay limit of detection (HIV-1 RNA = 3 copies/mL; 2-LTR DNA = 5 copies/10 peripheral blood mononuclear cell). Each circle represents 1 participant; the 2 solid squares represent the 2 patients with residual plasma HIV-1 RNA >11 copies/mL. 6 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 HIV-1 RNA, copies/mL 2-LTRc DNA, copies/10 PBMC association with levels of sCD14 and IL-6 [19]. Our data con- based on the infection stage at the start of ART, which could firm these observations over the longer term. However, we have introduced a bias. We observed a lower nadir CD4 cell found that levels of sCD27 were independently associated with count and a higher pre-ART viral load in subjects with longer levels of residual plasma HIV-1 RNA.  sCD27 is a 32-kD pro- duration of suppressive ART; however, the analyses did not tein identical to the extracellular domain of CD27, a transmem- identify an effect of nadir CD4 cell count or pre-ART viral load brane protein and TNF receptor that is expressed on subsets of on the measured parameters in our highly selected population. T, B, NK, and hematopoietic progenitor cells. sCD27 is released Of note, we excluded 2 outliers from the main analyses with primarily from CD4 T cells and provides a costimulatory func- residual plasma HIV-1 RNA levels >11 copies/mL. Whereas the tion that promotes activation and proliferation of T and B lym- main conclusions were not ae ff cted by their inclusion, it would phocytes [38–41]. sCD27 can be detected in healthy donors, have been of interest to determine any risk of viral load rebound and increased plasma levels have been observed in conditions with prospective follow-up in these 2 subjects. Further, we characterized by immune activation, especially auto-immune reported total HIV-1 DNA rather than integrated HIV-1 DNA disorders [42–44]. There are limited data in the setting of HIV as the indicator of the viral reservoir in this study. However, we infection. Levels of sCD27 have been shown to be generally previously described integrated HIV-1 DNA levels in a subset higher than in HIV-negative controls and found to be predic- of the same population and demonstrated that they were highly tive of the risk of HIV-related renal dysfunction and lympho- correlated with the levels of total HIV-1 DNA, not influenced mas [42, 45]. Levels have been shown to decline significantly in by the duration of suppressive ART, and not associated with the first 12 to 24 months of ART and to increase again with viral residual HIV-1 RNA [21]. We collected large volumes of blood load rebound or upon treatment discontinuation [29, 46, 47]. In from patients in order to test for multiple parameters (eg, 8 mL our study, consistent with the reported effect of treatment, levels of plasma was required for the ultrasensitive HIV-1 RNA assay of sCD27 were overall low and reduced further with longer dur- alone). Limited sample volumes meant that not all potentially ation of suppressive ART. Taken in the context of these previous relevant immune parameters were measured. When selecting observations, our findings illustrate a close association between soluble markers of immune activation, we elected to include 2 sCD27 and potential indicators of HIV replication during sup- markers that have been extensively explored in the published pressive ART that warrants further investigation. literature (sCD14 and IL-6) to provide a reference and added er Th e are scarce data on the effect of ART on sCD30 [26]. 2 markers for which there is high plausibility but scarce pub- Levels in our population were considerably lower than those lished evidence (sCD27 and sCD30). In this context, levels of measured in the first 4 weeks of ART [48], and closer to levels soluble markers of immune activation were found to be sim- published from HIV-negative volunteers [28]; however, sCD30 ilar to those historically reported in HIV-negative volunteers, showed no independent association with virologic parameters. which is consistent with the established beneficial effects of Levels of sCD14, IL-6, and cellular markers of immune activa- suppressive ART. However, published data from healthy volun- tion in our study population were comparable to those previ- teers were limited, especially for sCD27 and sCD30, and ideally, ously measured in HIV-1-positive patients receiving ART for matched HIV-negative controls should be sampled to confirm up to 6  years [19, 49]. Consistent with previous observations reference values. As an additional point, we measured plasma (reviewed in [16]), markers of immune activation on CD4 NNRTI concentrations in untimed specimens and for practical and CD8 cells (CD38 and HLA-DR/DP/DQ) were lower with reasons did not seek to obtain a measure of C . The patients trough longer duration of suppressive ART, and this apparent decline analyzed in the models had taken the last EFV dose at a consist- was paralleled by a predictable gain in CD4 cell counts. Overall, ent interval relative to sampling, and time of sampling would as also reported by Gandhi et al. [19], sCD14, IL6, and cellular be expected to have little impact overall. Finally, we recognize markers of immune activation showed no apparent association that both drug concentration and antiviral activity measured in with levels of residual plasma HIV-1 RNA and 2-LTRc DNA. peripheral blood do not necessarily reflect the situation in lym- Our study has limitations. First, findings obtained in a phoid tissue [8, 9]. cross-sectional study should be confirmed with prospectively In conclusion, this study provides evidence that residual collected data, which is challenging when aiming to study long- plasma HIV-1 RNA and PBMC-associated 2-LTRc DNA can term outcomes in populations maintained on the same regi- be detected in a subset of patients receiving long-term, con- men. To date, prospectively determined virologic indicators tinuously suppressive firstline ART, without evidence of a clear of HIV-1 persistence during long-term suppressive ART have association between the two, and without evidence of an asso- been poorly documented. Compared with published studies ciation of either parameter with plasma concentrations of EFV. addressing similar questions, one strength of our study is the While residual plasma HIV-1 RNA was not associated with relatively large population combined with stringent eligibility commonly reported markers of immune activation, there was criteria and a consistent definition of virologic suppression. a novel, independent direct association with levels of sCD27. It could be argued that our study population was not selected Further studies are warranted to confirm the findings, extend Residual HIV-1 RNA During ART • OFID • 7 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 5. Martinez-Picado J, Deeks SG. Persistent HIV-1 replication during antiretroviral the immune activation profiling, and determine the underlying therapy. Curr Opin HIV AIDS 2016; 11:417–23. mechanism, including whether the relationship reflects ongo- 6. Josefsson L, von Stockenstrom S, Faria NR, et  al. The HIV-1 reservoir in eight ing virus replication in lymphoid tissue. patients on long-term suppressive antiretroviral therapy is stable with few genetic changes over time. Proc Natl Acad Sci U S A 2013; 110:E4987–96. 7. Eriksson S, Graf EH, Dahl V, et al. Comparative analysis of measures of viral res- Supplementary Data ervoirs in HIV-1 eradication studies. PLoS Pathog 2013; 9:e1003174. Supplementary materials are available at Open Forum Infectious Diseases 8. Lorenzo-Redondo R, Fryer HR, Bedford T, et  al. Persistent HIV-1 replication maintains the tissue reservoir during therapy. Nature 2016; 530:51–6. online. Consisting of data provided by the authors to benefit the reader, 9. Fletcher CV, Staskus K, Wietgrefe SW, et al. Persistent HIV-1 replication is associ- the posted materials are not copyedited and are the sole responsibility of ated with lower antiretroviral drug concentrations in lymphatic tissues. Proc Natl the authors, so questions or comments should be addressed to the corre- Acad Sci U S A 2014; 111:2307–12. sponding author. 10. Nightingale S, Geretti AM, Beloukas A, et al. Discordant CSF/plasma HIV-1 RNA in patients with unexplained low-level viraemia. J Neurovirol 2016; 22:852–60. 11. Hatano H. Immune activation and HIV persistence: considerations for novel Acknowledgments therapeutic interventions. Curr Opin HIV AIDS 2013; 8:211–6. We are grateful to Dr. Giorgio Calisti (Central Manchester University 12. Cockerham LR, Siliciano JD, Sinclair E, et al. CD4+ and CD8+ T cell activation Hospitals NHS Foundation Trust, Manchester, UK), Dr. Paola Vitiello are associated with HIV DNA in resting CD4+ T cells. PLoS One 2014; 9:e110731. (Ospedali di Busto Arsizio, Varese, Italy), Ms. Helen Reynold (Royal 13. Sandler NG, Wand H, Roque A, et  al; INSIGHT SMART Study Group. Plasma Liverpool University Hospital, Liverpool, UK), and Ms. Anele Waters levels of soluble CD14 independently predict mortality in HIV infection. J Infect (Middlesex University Hospital, London, UK) for assistance with recruit- Dis 2011; 203:780–90. ments. We thank Ms. Tahami Fariba (Immunology Diagnostic Service, 14. Riddler SA, Aga E, Bosch RJ, et  al; ACTG A5276s Protocol Team. Continued slow decay of the residual plasma viremia level in HIV-1-infected adults receiving Royal Free Hospital, London, UK) for her support in producing the data on long-term antiretroviral therapy. J Infect Dis 2016; 213:556–60. cellular immune activation and Mr. Steven Lada (University of California 15. Mullins JI, Frenkel LM. Clonal expansion of human immunodeficiency virus-in- San Diego, San Diego, CA) for technical assistance with 2-LTRc DNA data fected cells and human immunodeficiency virus persistence during antiretroviral production. therapy. J Infect Dis 2017; 215:119–27. Author contributions. A.M.G.  and A.P.  designed the study. Technical 16. Klatt NR, Chomont N, Douek DC, Deeks SG. Immune activation and HIV persis- staff produced the laboratory work. A.R. analyzed the data under the super - tence: implications for curative approaches to HIV infection. Immunol Rev 2013; vision of A.C.L. and A.M.G. A.M.G. and A.R. wrote the manuscript, which 254:326–42. was reviewed by all authors. Members of the ERAS Study Group managed 17. Sereti I, Altfeld M. Immune activation and HIV: an enduring relationship. Curr recruitment. Opin HIV AIDS 2016; 11:129–30. 18. Khoury G, Fromentin R, Solomon A, et al. Human immunodeficiency virus per- Partecipating clinical centers (members of clinical study group). King’s sistence and T-cell activation in blood, rectal, and lymph node tissue in human College Hospital, London (Dr. Frank Post); North Middlesex Hospital, immunodeficiency virus-infected individuals receiving suppressive antiretroviral London (Dr. Jonathan Ainsworth); Royal Free Hospital, London and Royal therapy. J Infect Dis 2017; 215:911–9. Liverpool Hospital, Liverpool (Prof. Anna Maria Geretti); University 19. Gandhi RT, McMahon DK, Bosch RJ, et al; ACTG A5321 Team. Levels of HIV-1 College Hospital, London (Dr. Simon Edwards); St. Mary’s Hospital, persistence on antiretroviral therapy are not associated with markers of inflam- London (Dr. Nicola Mackie); St. Thomas’ Hospital, London (Dr. Julie Fox). mation or activation. PLoS Pathog 2017; 13:e1006285. Financial support. e s Th tudy was supported by research awards 20. Doyle T, Smith C, Vitiello P, et al. Plasma HIV-1 RNA detection below 50 copies/ from the European AIDS Treatment Network (NEAT), the British HIV ml and risk of virologic rebound in patients receiving highly active antiretroviral Association, the CARE and BEAT Martin Delaney Collaboratories of therapy. Clin Infect Dis 2012; 54:724–32. 21. Ruggiero A, De Spiegelaere W, Cozzi-Lepri A, et al; ERAS Study Group. During the National Institute of Allergy and Infectious Diseases (NIAID), the stably suppressive antiretroviral therapy integrated HIV-1 DNA load in periph- National Institute of Neurological Disorders and Stroke (NINDS), the eral blood is associated with the frequency of CD8 cells expressing HLA-DR/DP/ National Institute on Drug Abuse (NIDA) and the National Institute of DQ. EBioMedicine 2015; 2:1153–9. Mental Health (NIMH) of the National Institutes of Health (grant num- 22. Geretti AM, Arribas JR, Lathouwers E, et  al. Dynamics of cellular HIV-1 DNA bers 1UM1AI126619 and 1UM1Al126620), the University of California San levels over 144 weeks of darunavir/ritonavir monotherapy versus triple therapy in Diego Centre for AIDS Research (CFAR) (AI306214), the Department of the MONET trial. HIV Clin Trials 2013; 14:45–50. Veterans Affairs, and the James B. Pendleton Charitable Trust. The funding 23. Strain MC, Lada SM, Luong T, et al. Highly precise measurement of HIV DNA by sources had no role in the writing of the manuscript or the decision to sub- droplet digital PCR. PLoS One 2013; 8:e55943. mit for publication. The corresponding author (A.M.G.) had full access to 24. Poeta J, Linden R, Antunes MV, et al. Plasma concentrations of efavirenz are asso- ciated with body weight in HIV-positive individuals. J Antimicrob Chemother all the data in the study and had the final responsibility for the decision to 2011; 66:2601–4. submit for publication. 25. Havlir D, Cheeseman SH, McLaughlin M, et  al. High-dose nevirapine: safety, Potential coni fl cts of interest. A.M.G. is currently employed as Expert pharmacokinetics, and antiviral effect in patients with human immunodeficiency Scientist at Roche Pharma Research and Early Discovery (pRED); Roche virus infection. J Infect Dis 1995; 171:537–45. had no involvement in the work. The other authors have no conflicts of inter - 26. Uysal HK, Sohrabi P, Habip Z, et al. Neopterin and soluble CD14 levels as indica- est to declare. All authors have submitted the ICMJE Form for Disclosure of tors of immune activation in cases with indeterminate pattern and true positive Potential Conflicts of Interest. Conflicts that the editors consider relevant to HIV-1 infection. PLoS One 2016; 11:e0152258. the content of the manuscript have been disclosed. 27. Saing T, Valdivia A, Hussain P, et al. Data on pro-inflammatory cytokines IL-1β, IL-17, and IL-6 in the peripheral blood of HIV-infected individuals. Data Brief 2016; 8:1044–7. References 28. Gao R, Sun W, Chen Y, et al. Elevated serum levels of soluble CD30 in ankylosing 1. Guidelines E. EACS Guidelines Version 8.2. 2017. Available at: http://wwweacso- spondylitis patients and its association with disease severity-related parameters. cietyorg/files/guidelines_82-english.pdf. Accessed 22 August 2017. Biomed Res Int 2015; 2015:617282. 2. DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents – 29. De Milito A, Aleman S, Marenzi R, et al. Plasma levels of soluble CD27: a sim- A Working Group of the Office of AIDS Research Advisory Council (OARAC). ple marker to monitor immune activation during potent antiretroviral therapy in DASHH guidelines for the use of antiretroviral agents in HIV-1-infected adults HIV-1-infected subjects. Clin Exp Immunol 2002; 127:486–94. and adoloscencents. Updated 14 July 2016. Available at: https://aidsinfo.nih.gov/ 30. Marzolini C, Elzi L, Gibbons S, et  al; Swiss HIV Cohort Study. Prevalence of contentfiles/lvguidelines/adultandadolescentgl.pdf. Accessed 22 August 2017. comedications and effect of potential drug-drug interactions in the Swiss HIV 3. Palmer S, Maldarelli F, Wiegand A, et  al. Low-level viremia persists for at least Cohort Study. Antivir Ther 2010; 15:413–23. 7 years in patients on suppressive antiretroviral therapy. Proc Natl Acad Sci U S A 31. Hong FF, Mellors JW. Changes in HIV reservoirs during long-term antiretroviral 2008; 105:3879–84. therapy. Curr Opin HIV AIDS 2015; 10:43–8. 4. Doyle T, Geretti AM. Low-level viraemia on HAART: significance and manage- 32. Maldarelli F, Palmer S, King MS, et al. ART suppresses plasma HIV-1 RNA to a ment. Curr Opin Infect Dis 2012; 25:17–25. stable set point predicted by pretherapy viremia. PLoS Pathog 2007; 3:e46. 8 • OFID • Ruggiero et al Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018 33. Besson GJ, Lalama CM, Bosch RJ, et  al. HIV-1 DNA decay dynamics in blood 42. Widney D, Gundapp G, Said JW, et  al. Aberrant expression of CD27 and sol- during more than a decade of suppressive antiretroviral therapy. Clin Infect Dis uble CD27 (sCD27) in HIV infection and in AIDS-associated lymphoma. Clin 2014; 59:1312–21. Immunol 1999; 93:114–23. 34. Butler SL, Johnson EP, Bushman FD. Human immunodeficiency virus cDNA 43. van der Vuurst de Vries RM, Mescheriakova JY, Runia TF, et  al. Soluble CD27 metabolism: notable stability of two-long terminal repeat circles. J Virol 2002; levels in cerebrospinal fluid as a prognostic biomarker in clinically isolated syn- 76:3739–47. drome. JAMA Neurol 2017; 74:286–92. 35. Pierson TC, Kieffer TL, Ruff CT, et al. Intrinsic stability of episomal circles formed 44. Speeckaert R, Lambert J, van Geel N. Clinical significance of serum solu- during human immunodeficiency virus type 1 replication. J Virol 2002; 76:4138–44. ble CD molecules to assess disease activity in vitiligo. JAMA Dermatol 2016; 36. Sharkey ME, Teo I, Greenough T, et al. Persistence of episomal HIV-1 infection inter- 152:1194–200. mediates in patients on highly active anti-retroviral therapy. Nat Med 2000; 6:76–81. 45. Abraham AG, Darilay A, McKay H, et  al. Kidney dysfunction and mark- 37. Hatano H, Jain V, Hunt PW, et  al. Cell-based measures of viral persistence are ers of inflammation in the multicenter AIDS Cohort Study. J Infect Dis 2015; associated with immune activation and programmed cell death protein 1 (PD-1)- 212:1100–10. expressing CD4+ T cells. J Infect Dis 2013; 208:50–6. 46. Atlas A, Thanh Ha TT, Lindström A, et al. Effects of potent antiretroviral therapy 38. Nolte MA, van Olffen RW, van Gisbergen KP, van Lier RA. Timing and tuning of on the immune activation marker soluble CD27 in patients infected with HIV-1 CD27-CD70 interactions: the impact of signal strength in setting the balance between subtypes A-D. J Med Virol 2004; 72:345–51. adaptive responses and immunopathology. Immunol Rev 2009; 229:216–31. 47. Wallet MA, Rodriguez CA, Yin L, et al. Microbial translocation induces persistent 39. van Oosterwijk MF, Juwana H, Arens R, et al. CD27-CD70 interactions sensitise naive macrophage activation unrelated to HIV-1 levels or T-cell activation following CD4+ T cells for IL-12-induced Th1 cell development. Int Immunol 2007; 19:713–8. therapy. AIDS 2010; 24:1281–90. 40. Hintzen RQ, de Jong R, Hack CE, et al. A soluble form of the human T cell dif- 48. Kwamena WCS, Kwabena O, Makafui S, et al. Decay of soluble CD30 and HIV-1 ferentiation antigen CD27 is released after triggering of the TCR/CD3 complex. plasma viral load during early highly active antiretroviral therapy: A short-term J Immunol 1991; 147:29–35. longitudinal study. J Microbiol Biochem Technol 2016; 8(2):090–6: 90. 41. Huang J, Jochems C, Anderson AM, et  al. Soluble CD27-pool in humans 49. Borges ÁH, O’Connor JL, Phillips AN, et al; INSIGHT SMART and ESPRIT Study may contribute to T cell activation and tumor immunity. J Immunol 2013; Groups and the SILCAAT Scientific Committee. Factors associated with plasma 190:6250–8. IL-6 levels during HIV infection. J Infect Dis 2015; 212:585–95. Residual HIV-1 RNA During ART • OFID • 9 Downloaded from https://academic.oup.com/ofid/article-abstract/5/2/ofy032/4837293 by Ed 'DeepDyve' Gillespie user on 16 March 2018

Journal

Open Forum Infectious DiseasesOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off