Antiviral Therapy 2013; 18: (doi: /IMP2667)

Transcription

1 Antiviral Therapy 3; 8:95 93 (doi:.385/imp667) Original article 96- results of abacavir/lamivudine versus tenofovir/emtricitabine, plus efavirenz, in antiretroviral-naive, HIV--infected adults: ASSERT study Graeme J Moyle, Hans-Jürgen Stellbrink, Juliet Compston 3, Chloe Orkin 4, José R Arribas 5, Pere Domingo 6, Catherine Granier 7, Helen Pearce 7, Sangeeta Sedani 7, Martin Gartland 8 *, the ASSERT Team Chelsea and Westminster Hospital, London, UK ICH Study Center, Hamburg, Germany 3 Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK 4 Royal London Hospital, Barts and The London NHS Trust, London, UK 5 Hospital La Paz, IdiPAZ, Madrid, Spain 6 Hospital Santa Creu y Sant Pau, Barcelona, Spain 7 GlaxoSmithKline, plc, London, UK 8 ViiV Healthcare, Research Triangle Park, NC, USA *Corresponding author martin.j.gartland@viivhealthcare.com Background: Abacavir/lamivudine () and tenofovir disoproxil fumarate/emtricitabine () are widely used as first-line antiretroviral therapies. However, there are limited data comparing the safety of these therapies with long-term use. The objective of this study was to assess the long-term safety of these commonly used first-line nucleoside/nucleotide combinations each administered with efavirenz (EFV). Methods: This open-label, 96-week, randomized study compared the safety (renal, bone and metabolic) and efficacy of and plus EFV in HLA-B*57- negative antiretroviral-naive adults. Results: A total of 385 subjects were enrolled, and 49 (65%) subjects completed the study. The difference in changes from baseline in estimated glomerular filtration rate (calculated by the Modified Diet in Renal Disease equation) between treatment arms was not significant. There was a significant difference between the arms (P<.) for markers of tubular dysfunction (retinol-binding protein and ß- microglobulin) favouring. Hip bone mineral density decreased from baseline in both arms, with a significantly greater decline with ( -.% and -3.5%; P<. at week 96). Subjects in the arm had greater increases from baseline in median total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol and triglycerides. Adverse events were similar between arms. The virological failure rate was low in both arms. Conclusions: and in combination with EFV minimally affected estimated glomerular filtration rate over 96 weeks. was associated with greater increases in tubular dysfunction and bone turnover marker levels, greater decreases in hip bone mineral density, and smaller increases in serum lipid levels. Introduction As antiretroviral therapy (ART) is administered lifelong, understanding the long-term safety of antiretroviral medications is crucial for informed treatment choices. Two fixed-dose nucleoside/nucleotide analogue combinations are widely recommended for use in initial ART regimens: abacavir sulfate/lamivudine () and tenofovir disoproxil fumarate/ emtricitabine () [,]. ABC is associated with a hypersensitivity reaction, which can be reliably predicted and significantly reduced with HLA-B*57 screening [3,4]. No other characteristic side effects of ABC have been clearly identified. The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) cohort study has suggested a link between ABC and increased risk of myocardial infarction [5]. The majority of recent randomized 3 International Medical Press (print) 4-58 (online) 95

2 GJ Moyle et al. clinical trial data (including meta-analyses), cohort analyses and mechanistic data have not supported an association between ABC and myocardial infarction [6 9]. However, these studies were not prospectively designed to measure risk of myocardial infarction [5,7 9]. Several long-term studies evaluating the renal safety of TDF determined that the overall incidence of nephrotoxicity with TDF was low [ ]. However, most of these data focused on glomerular function rather than proximal tubule damage and routinely excluded subjects with renal dysfunction, such as an estimated glomerular filtration rate (egfr)<5 ml/ min [,3]. Animal studies, case reports and clinical cohort studies have suggested a potential link exists between TDF and nephrotoxicity (for example, renal failure or tubular dysfunction rarely leading to Fanconi syndrome) [4 9]. Recent studies have demonstrated bone loss in subjects receiving ART, the magnitude of which varies according to the ART regimen [ 3]. Initiation of therapy with is associated with a significantly greater decrease in spine and hip bone mineral density (BMD) than with []. Also, some recent cohort studies have suggested that fracture risk is increased in HIV-infected individuals, although the contribution of ART is uncertain [4 6]. The ASSERT trial was designed to compare the 96-week safety of and, each combined with efavirenz (EFV). The primary analysis of the ASSERT trial at week 48 suggested potential differences between the two treatment arms regarding safety profiles [7]. Although there were no differences in the egfr between treatment arms, the arm was associated with significant increases in two markers of renal tubular dysfunction (that is, retinol-binding protein [RBP] and b- microglobulin [BM]). In addition, there were greater increases in bone turnover and greater decreases in hip BMD with than with [8]. Here we present the final, week 96 analysis of the ASSERT trial. Methods The full methodology for this multicentre randomized open-label study (National Clinical Trials identifier NCT54998) has been previously described [7,8]. In brief, this study, conducted in 76 centres in 3 European countries, randomized ARTnaive, HLA-B*57-negative, HIV-infected subjects to receive either or, each with EFV, for 96 weeks. Randomization was stratified by screening egfr (<9 versus 9 ml/min/.73 m ), race (Black or non-black) and body mass index (BMI; <5 or 5 kg/m ). The results of the primary safety end point, change from baseline in egfr (calculated using the Modified Diet in Renal Disease [MDRD] equation) at week 48, were previously reported, and final analysis of this parameter was conducted at week 96 [7]. Key secondary end points included change from baseline in egfr (by MDRD and Cockcroft-Gault) at week 96, proportion of subjects with decline from baseline in egfr, proportion of subjects with National Kidney Foundation chronic kidney disease or proximal renal tubule dysfunction (PRTD), changes from baseline in lumbar spine and hip BMD (measured by dual-energy X-ray absorptiometry), and proportion of subjects who met World Health Organization criteria for osteopaenia (T-score between -. and -.5) and osteoporosis (T-score <-.5). Adverse events, clinical laboratory evaluations (including fasting lipids) and efficacy (for example, proportion of subjects with HIV- RNA<5 and <4 copies/ml, change from baseline in HIV- RNA, and CD4 + T-cell counts) were also summarized. In addition, urine renal biomarkers (that is, RBP, BM, N-acetyl-b-D-glucosaminidase [NAG] and albumin), cardiovascular parameters (high-sensitivity C-reactive protein [hs-crp], interleukin 6 [IL-6] and adiponectin), lipid parameters (total cholesterol [TC], high-density lipoprotein cholesterol [HDL], low-density lipoprotein cholesterol [LDL] and triglycerides) and bone biomarkers (bone-specific alkaline phosphatase [BSAP], procollagen type amino-terminal propeptide [PNP], carboxy-terminal cross-linking telopeptide of bone collagen [CTx] and osteocalcin) were assessed in an exploratory analysis. Viral resistance patterns in protocol-defined virological failures were also evaluated. Protocol-defined virological failure was identified as either a confirmed failure to achieve a log reduction in plasma HIV- RNA by week 4 or a confirmed reduction of plasma HIV- RNA to <4 copies/ml with a subsequent increase to >4 copies/ml on two consecutive occasions, to 4 weeks apart to week 4, or an increase to >4 copies/ml on two consecutive occasions, to 4 weeks apart after week 4. A total of 38 subjects (9 per arm) was necessary to provide 9% power to detect a ml/min difference in change from baseline in egfr between the arms [7,8]. Safety and efficacy end points were assessed using the intention-to-treat exposed (ITT-E) population, which consisted of all randomized subjects who received dose of study medication. The biomarker safety population included those members of the ITT-E population with a baseline reading and post-baseline reading. Efficacy analyses were based on the proportion of subjects achieving HIV- RNA less than the predefined threshold using the time to loss of virological response (TLOVR) algorithm International Medical Press

3 Final, 96-week results of ASSERT study: a randomized trial Results Study population A total of 39 subjects were randomized between July and December 7, and 385 subjects received dose of either (n=9) or (n=93). Baseline demographics were similar in both treatment arms [7]. Most subjects were male (8%) and White/Caucasian (78%). By week 96, 36 subjects (35%) had prematurely discontinued study treatment, including 77 (4%) subjects receiving and 59 (3%) subjects receiving (Additional file ). Discontinuations were driven primarily by adverse events (AEs), which occurred at different times in the two arms, leading to a difference in mean treatment exposure time ( 49. days and 533. days). Safety Renal At week 48 (primary end point), there was no significant difference between the arms in adjusted mean changes from baseline in egfr using the MDRD equation [7]. By week 96, the adjusted mean change from baseline using the repeated-measures mixed model analysis was.48 ml/min/.73 m in the arm and -.5 ml/min/.73 m in the TDF/ FTC arm (Figure ). These results were supported by different sensitivity analyses assessing the impact of missing data. The adjusted mean difference between the treatment arms of -.64 ml/min/.73 m was not significant (95% CI -5.38,.; P=.6). Baseline egfr had a significant effect on the change from baseline in egfr by Cockcroft-Gault. Subjects with a baseline egfr<9 ml/min experienced a greater mean change in egfr (improvement) at week 96 compared with subjects who had a baseline egfr 5 ml/min (P<.5). Similarly, by MDRD subjects with a baseline egfr<9 ml/min experienced a greater mean change in egfr (improvement) at week 96 compared with subjects who had a baseline egfr ml/min (P<.). In the analysis of secondary renal end points, the proportion of subjects with a decline from baseline in egfr ml/min/.73 m at week 96 was greater in the arm (% versus 4% in the arm). The proportion of subjects experiencing a decrease in renal function of toxicity grade post-baseline was similar in both treatment arms (7% of and 9% of ). There were significant differences in change from baseline to week 96 between the two treatment arms observed for the biomarkers of tubular function examined (BM/creatinine and RBP/creatinine ratios) in favour of (Table and Figure ). No differences were seen for the glomerular function biomarkers (NAG/creatinine or albumin/creatinine ratios; Table and Figure ) between the arms. These results were similar to those of the week 48 analysis [7]. In addition, no subjects met the protocol definition of PRTD. Bone At week 96, a significantly greater decline from baseline in hip BMD, but not lumbar spine BMD, was observed for subjects in the arm compared Figure. Adjusted mean change in egfr by MDRD by visit Change in egfr from baseline, ml/min/.73 m : 9 7 : P= Adjusted mean change from baseline (95% CI) in estimated glomerular filtration rate (egfr) using the Modified Diet in Renal Disease (MDRD) equation repeatedmeasures mixed model analysis (intention-to-treat exposed population) with the following covariates: treatment, visit, baseline egfr by MDRD, baseline body mass index (BMI), race group, treatment visit interaction, baseline egfr by MDRD visit interaction and baseline BMI visit interaction., abacavir/lamivudine; TDF/ FTC, tenofovir disoproxil fumarate/emtricitabine. Antiviral Therapy

4 GJ Moyle et al. Table. Exploratory multivariate analyses of change from baseline at week 96 Subjects a Mean change from baseline Difference between Biomarker arms (95% CI) b P-value Bone biomarkers PNP, % (.,.3) <. CTx, ng/l (4.8, 85.6) c.9 Osteocalcin, mg/l (., 4.54) c.9 BSAP, mg/l (.7,.7) c.66 Renal biomarkers RBP/creatinine ratio, % (.,.65) <. BM/creatinine ratio, % (.39,.38) <. Albumin/creatinine ratio, % (.9,.37).35 NAG/creatinine ratio, % (.9,.7).333 a Values represent the number of patients. b Values represent ratios of the mean changes from baseline between each arm, unless otherwise noted. c Values represent the difference between the mean changes from baseline for each arm., abacavir/lamivudine; BSAP, bone-specific alkaline phosphatase; BM, B- microglobulin; CTx, carboxy-terminal cross-linking telopeptide of bone collagen; NAG, N-acetyl-b-D-glucosaminidase; PNP, procollagen type amino-terminal propeptide; RBP, retinol-binding protein;, tenofovir disoproxil fumarate/emtricitabine. with the arm (Figure 3). The adjusted mean percentage change from baseline in total hip BMD was -.7% in the arm and -3.55% in the arm (treatment difference -.4%, 95% CI -., -.6; P<.). The adjusted mean percentage change from baseline in lumbar spine BMD was -.87% in the arm and -.7% in the arm (treatment difference -.84%, 95% CI -.87,.; P=.). The steepest declines in BMD were observed during the first 4 weeks of treatment, after which the rate of decrease in total hip BMD lessened in both treatment arms and some recovery in lumbar spine BMD was observed. Of those covariates adjusted in the analyses, lower baseline BMI and non-black race were significantly associated with greater bone loss in both total hip and lumbar spine BMD. The adjusted percentage change from baseline in hip BMD was -4.5%, -3.3%, -.5% and -.9% for subjects with baseline BMI< kg/m, to <3 kg/ m, 3 to <5 kg/m and 5 kg/m, respectively. The adjusted percentage change from baseline was -.8% for Black subjects versus -3.% for non-black subjects. In the statistical model for spine BMD, the effect of race and baseline BMI was slightly less pronounced but still significant. A greater proportion of subjects in the arm had a decrease of 6% from baseline in both total hip and lumbar spine BMD (hip 7% and spine %) than subjects in the arm (hip % and spine 5%) at week 96. Osteoporosis was infrequent in both treatment arms. There was little change in the prevalence of osteoporosis (World Health Organization criteria of T-score -.5) with an increase in the lumbar spine from 7% to 8% in the treatment arm and from % to 4% in the treatment arm. Increases from baseline in markers of bone turnover (BSAP, CTx, osteocalcin and PNP) over the first 4 weeks were observed for both treatment arms, followed by a stabilization or decline. Overall changes from baseline to week 96 in biomarkers of both bone formation and remodelling were significantly greater in the arm than in the arm (Table and Figure ). We chose to concentrate on bone turnover markers in this study, and we did not assess the use of calcium or vitamin D supplements. Other secondary safety end points The majority of subjects (9%) reported AE during the study. Dizziness, nasopharyngitis and diarrhoea were the most common AEs, occurring at similar rates in both arms. The incidence of grade 3 and grade 4 AEs (4% versus %) and of serious AEs (6% versus %) was higher in the arm than in the arm. Despite HLA-B*57 screening, the most common AE leading to permanent discontinuation of study drug in the arm was drug hypersensitivity (n=4), including suspected ABC hypersensitivity reaction (n=6) and EFV-related reaction (n=8). In the arm, eight subjects were permanently discontinued due to decreased BMD, including one who also reported hypocalcaemia. All drug hypersensitivity reactions occurred before week, while discontinuations from BMD decreases occurred from 4 weeks through to the end of the study. A total of subjects experienced cardiac events, 6 (3%) in the arm and 5 (3%) in the TDF/3TC arm, during the entire study with 3 new cardiac events (intracardiac thrombus, angina pectoris and mitral valve prolapse) reported by 4 subjects ( in each arm) between week 48 and week 96. Subjects in the arm had greater increases from baseline in median TC, HDL cholesterol, LDL cholesterol and triglycerides at week International Medical Press

5 Final, 96-week results of ASSERT study: a randomized trial Figure. Mean change from baseline in bone and renal biomarkers A BSAP change from baseline, µg/l B CTx change from baseline, ng/l : : C Osteocalcin change from baseline, µg/l : : D Ratio of change from baseline in PNP : 4 : : : E Ratio of change from baseline in RBP/creatinine ratio.5.5 : 46 : F Ratio of change from baseline in BM/creatinine ratio.5.5 : 79 : G H Ratio of change from baseline in albumin/creatinine ratio.5.5 : 46 : Ratio of change from baseline in NAG/creatinine ratio.5.5 : 46 : Exploratory multivariate analyses of mean change from baseline (95% CI) in (A D) bone and (E H) renal biomarkers. Estimates were calculated using ANOVA models adjusted as follows: bone-specific alkaline phosphatase (BSAP; baseline value and baseline CD4 + T-cell count), carboxy-terminal cross-linking telopeptide of bone collagen (CTx; baseline value and gender), osteocalcin (baseline value, age and baseline CD4 + T-cell count), procollagen type amino-terminal propeptide (PNP; baseline value and age), retinol-binding protein (RBP)/creatinine ratio (baseline value, baseline CD4 + T-cell count and gender), b- microglobulin (BM)/creatinine ratio (baseline value, age, baseline CD4 + T-cell count and gender), albumin/creatinine ratio (baseline value, age and gender) and N-acetyl-b-D-glucosaminidase (NAG)/ creatinine ratio (baseline value and age)., abacavir/lamivudine;, tenofovir disoproxil fumarate/emtricitabine. Antiviral Therapy

6 GJ Moyle et al. Figure 3. Change from baseline in BMD of total hip and lumbar spine A Change in total hip BMD, % : 76 : 8 B Change in lumbar spine BMD, % : 8 : P< P= Multivariate analysis of change from baseline (95% CI) in bone mineral density (BMD) of (A) total hip and (B) lumbar spine using a repeated-measures model including the following covariates: for hip, treatment, visit, baseline hip BMD, baseline body mass index (BMI), race group, risk factor, prohibited medication, previous fracture, treatment visit interaction, baseline hip BMD visit interaction and baseline BMI visit interaction; for spine, treatment, visit, baseline spine BMD, baseline BMI, race group, age group, treatment visit interaction, baseline spine BMD visit interaction, and baseline BMI visit interaction., abacavir/lamivudine;, tenofovir disoproxil fumarate/emtricitabine. (Figure 4). However, there was no difference in the median change in TC:HDL ratio between treatment arms. Inflammatory markers hs-crp, IL-6 and adiponectin were evaluated in an exploratory analysis. In both treatment arms there was a decrease in hs-crp and IL-6 levels over time with no difference between the treatment arms. There was a significant difference (P=.54) observed in the improvement from baseline for adiponectin. Efficacy In the TLOVR analysis at week 96 (Additional file ), 59% of subjects receiving and 5% of those receiving achieved HIV- RNA<5 copies/ml (difference 7.5%, 95% CI -.4, 7.4). Similarly, 65% of subjects receiving and 57% receiving achieved HIV- RNA<4 copies/ ml (difference 8.%, 95% CI -.7, 7.7). Treatment response in subjects in both arms was similar regardless of baseline viral load. The proportions of subjects achieving plasma HIV- RNA<5 copies/ml for the low (<, copies/ml) and high (, copies/ ml) viral subgroups were 49% and 53%, respectively, in the arm and 59% and 58%, respectively, in the arm. Similar increases from baseline to week 96 in median CD4 + T-cell counts were observed in both treatment arms ( 35 cells/mm 3 and cells/mm 3 ). Overall, (3%) subjects met the criteria for protocol-defined virological failure (7 subjects receiving and 3 subjects receiving ). In the arm, 4 subjects developed EFV-associated mutations (K3N, V6M, V79D, Y88L and G9A/G) and subjects developed nucleoside reverse transcriptase inhibitor-associated mutations (K65R, D67N and M84V). No treatment-emergent mutations were seen with. Discussion This study compared the safety of and TDF/ FTC, each administered with EFV, over 96 weeks. No significant changes from baseline in egfr were observed in either treatment arm at week 96 nor did any subjects develop PRTD or grade 3/4 renal toxicity. Significant differences in BM/creatinine and RBP/ creatinine ratios were observed between and at week 96 in favour of. RBP and BM are low molecular weight proteins, and increased urinary loss reflects reduced tubular reabsorption of these proteins [9]. Excretion of RBP has been proposed as the most sensitive marker of early renal tubular damage [9] and thus may substantially precede clinical disease. Of note, BM declines with HIV suppression; therefore, the observed decline in urinary loss of this protein with but not was in line with known changes in plasma BM with successful ART [3]. Changes in these tubular proteins may occur in the absence of reductions in glomerular filtration, 9 3 International Medical Press

7 Final, 96-week results of ASSERT study: a randomized trial Figure 4. Mean change from baseline in fasting lipids TC a n=3 n= P<. HDL a LDL a n=3 n=97 n= n= P<.7 P=.7 TG b n=3 n= P= Mean change from baseline, mg/dl Difference, - Comparison of the mean change (95% CI), unless otherwise noted, in fasting lipids from baseline to week 96 (safety population) between abacavir/lamivudine () and tenofovir disoproxil fumarate/emtricitabine () treatment arms. CIs were analysed by two-sample Student s t-test. a Mean, 95% CI and two-sample Student s t-test are presented. b Median, IQR and Mann-Whitney test are presented. HDL, high-density lipoprotein; LDL, low-density lipoprotein; TC, total cholesterol; TG, triglycerides. which is a reflection of the active and passive clearance of creatinine. It is therefore not surprising that these differences in tubular protein loss were observed in the absence of differences in egfr. PRTD and severe osteomalacia have been described during TDF therapy in the absence of reductions in egfr [6 9]. The clinical significance of mild-to-moderate increases in urinary RBP and BM loss remains unclear but requires further investigation as a potentially more targeted means of monitoring for TDF-related tubular injury than the current standard of egfr plus urine dipstick. A trial conducted by the AIDS Clinical Trial Group (ACTG; study A54, a substudy of ACTG A5) indicated that was associated with significantly greater reductions in lumbar spine and hip BMD compared with [,3]. In a powered secondary analysis in the current study, the arm had significantly greater loss of hip BMD compared with the arm at week 96, whereas the significantly greater loss of spine BMD observed at week 48 in subjects in the arm was no longer significant at week 96. The results confirm those of the A54 study in terms of the pattern of BMD reduction and the change in bone biomarkers. Both studies showed significantly greater increases in markers of bone resorption and formation indicating higher bone turnover in subjects in the arm than in those in the arm [,3]. Despite the significantly greater decreases in BMD with than with, the longterm clinical significance of these findings is not known. In their Veterans Affairs study, Bedimo et al. [4] observed a significantly increased risk of low trauma fractures and an association with TDF use. It must be noted however that BMD measurements were not obtained in that study. Thus, the correlation between parameters such as low BMD and fracture rates, specific to ARV regimens, has not been established. Consistent with previous initial therapy studies (ACTG 5 and HEAT), small differences in changes in lipids were observed with greater increase in both TC and HDL cholesterol with [3,33], although there was no difference in the median change in TC:HDL ratio between the treatment arms. Only one myocardial infarction was observed in the ASSERT study in a subject receiving [7]. Recent meta-analyses of randomized clinical trial data suggest no association between ABC and myocardial infarction or major cardiovascular events [7 9]. The incidence of grade 3 and grade 4 AEs (4% versus %) and of serious AEs (6% versus %) was higher in the arm than in the arm, the main reason for this difference being recorded as either drug hypersensitivity or hypersensitivity. Since the week 48 analysis from this study, the proportion of patients achieving treatment success declined in both arms, and by a similar amount (7% to 59% achieved <5 copies/ml in the arm, and 59% to 5% in the arm) [7]. The study was not powered to assess efficacy, and the evaluation of virological response rates is complicated by the high withdrawal rate observed in both treatment arms, primarily as a result of early discontinuations due to AEs and administrative reasons in the arm and later discontinuations due to AEs in the arm. Interestingly, there were no changes in the numbers of patients withdrawing from the study due to virological reasons between weeks 48 and 96. Previous comparative studies have not routinely included HLA-B*57 testing or full Antiviral Therapy 8.7 9

8 GJ Moyle et al. genotypic screening. However, the more well-powered HEAT study using TLOVR end points did not observe a difference between and in treatment efficacy [33], whereas ACTG A5, which reported initial results during the conduct of this study and used different end points indicated differences in subjects with high (>, copies/ml) but not lower screening viral loads, favouring [3]. Overall this study demonstrates that and, when used in combination with EFV, minimally affected egfr, although the arm was associated with greater increases in markers of tubular dysfunction (BM and RBP). Furthermore, these data suggest that may be associated with greater increases in bone turnover and greater loss of BMD than. Because patients are remaining on their initial regimens longer, continued study of the long-term safety profiles of these regimens (including bone effects) is warranted. Acknowledgements We gratefully acknowledge all subjects who participated in the study and all contributing investigators. All listed authors meet the criteria for authorship set forth by the International Committee for Medical Journal Editors. The authors also wish to acknowledge the following individual for his editorial assistance during the development of this manuscript: Todd Parker. Funding for this study was provided by GlaxoSmithKline. The data reported herein were presented in part at the th International Workshop on Adverse Drug Reactions and Comorbidities in HIV, 4 6 November, London, UK. Disclosure statement The authors CG, HP and SS are employees of Glaxo- SmithKline and receive company stock as part of their incentive packages. The author MG is an employee of ViiV Healthcare and receives company stock as part of his incentive package. Additional potential conflicts of interest are detailed as follows: funds have been received directly through grants by CO and PD, through consultancy by GJM, PD, JRA and H-JS, through speaker s bureaus by GJM, CO and JRA, through travel support by CO and H-JS, through board membership by CO, PD, JRA and H-JS, and through development of educational presentations by CO, JRA and H-JS; funds have been received by institutions through grants to GJM, CO and JRA. Additional file Additional file : A figure illustrating a summary of subject disposition and a figure illustrating the proportion of subjects with HIV RNA<4 and <5 copies/ml per visit by ITT-E analysis can be accessed via 3-OA-847_Moyle_Add_file.pdf References. European AIDS Clinical Society. European AIDS Clinical Society (EACS) Guidelines. Version 6. October. (Accessed 5 May.) Available from EACSGuidelines-v6.-English.pdf. Thompson MA, Aberg JA, Cahn P, et al. Antiretroviral treatment of adult HIV infection: recommendations of the International AIDS Society USA panel. JAMA ; 34: Hetherington S, McGuirk S, Powell G, et al. Hypersensitivity reactions during therapy with the nucleoside reverse transcriptase inhibitor abacavir. Clin Ther ; 3: Mallal S, Phillips E, Carosi G, et al. HLA-B*57 screening for hypersensitivity to abacavir. N Engl J Med 8; 358: D:A:D Study Group. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: a multi-cohort collaboration. Lancet 8; 37: Kim C, Gupta SK, Green L, et al. Abacavir, didanosine and tenofovir do not induce inflammatory, apoptotic or oxidative stress genes in coronary endothelial cells. Antivir Ther ; 6: Ding X, Andraca-Carrera E, Cooper C, et al. No association of myocardial infarction with abacavir use: findings of an FDA meta-analysis. 8th Conference on Retroviruses and Opportunistic Infections. 7 February March, Boston, MA, USA. Abstract O Cruciani M, Zanichelli V, Serpelloni G, et al. Abacavir use and cardiovascular disease events: a meta-analysis of published and unpublished data. AIDS ; 5: Brothers CH, Hernandez JE, Cutrell AG, et al. Risk of myocardial infarction and abacavir therapy: no increased risk across 5 GlaxoSmithKline-sponsored clinical trials in adult subjects. J Acquir Immune Defic Syndr 9; 5: 8.. Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz versus zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med 6; 354:5 6.. Cassetti I, Madruga JV, Suleiman JM, et al. The safety and efficacy of tenofovir DF in combination with lamivudine and efavirenz through 6 years in antiretroviral-naive HIV--infected patients. HIV Clin Trials 7; 8: Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA 4; 9:9. 3. Arribas JR, Pozniak AL, Gallant JE, et al. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz compared with zidovudine/lamivudine and efavirenz in treatment-naive patients: 44-week analysis. J Acquir Immune Defic Syndr 8; 47: Viread (tenofovir disoproxil fumarate). Package insert. Gilead Sciences, Inc., Foster City, CA, USA. 5. Verhelst D, Monge M, Meynard JL, et al. Fanconi syndrome and renal failure induced by tenofovir: a first case report. Am J Kidney Dis ; 4: Karras A, Lafaurie M, Furco A, et al. Tenofovir-related nephrotoxicity in human immunodeficiency virus-infected patients: three cases of renal failure, Fanconi syndrome, and nephrogenic diabetes insipidus. Clin Infect Dis 3; 36: International Medical Press

9 Final, 96-week results of ASSERT study: a randomized trial 7. Créput C, Gonzalez-Canali G, Hill G, et al. Renal lesions in HIV--positive patient treated with tenofovir. AIDS 3; 7: Calza L, Trapani F, Tedeschi S, et al. Tenofovir-induced renal toxicity in 34 HIV-infected, antiretroviral-naive patients. Scand J Infect Dis ; 43: Horberg M, Tang B, Towner W, et al. Impact of tenofovir on renal function in HIV-infected, antiretroviral-naive patients. J Acquir Immune Defic Syndr ; 53: Duvivier C, Kolta S, Assoumou L, et al. Greater decrease in bone mineral density with protease inhibitor regimens compared with nonnucleoside reverse transcriptase inhibitor regimens in HIV- infected naive patients. AIDS 9; 7: Hansen AB, Obel N, Nielsen H, Pedersen C, Gerstoft J. Bone mineral density changes in protease inhibitor-sparing vs nucleoside reverse transcriptase inhibitor-sparing highly active antiretroviral therapy: data from a randomized trial. HIV Med ; : McComsey GA, Kitch D, Daar ES, et al. Bone mineral density and fractures in antiretroviral-naive persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarateemtricitabine along with efavirenz or atazanavir-ritonavir: AIDS Clinical Trials Group A54s, a substudy of ACTG A5. J Infect Dis ; 3: van Vonderen MG, Lips P, van Agtmael MA, et al. First line zidovudine/lamivudine/lopinavir/ritonavir leads to greater bone loss compared to nevirapine/lopinavir/ritonavir. AIDS 9; 3: Bedimo R, Maalouf NM, Zhang S, Drechsler H, Tebas P. Osteoporotic fracture risk associated with cumulative exposure to tenofovir and other antiretroviral agents. AIDS ; 6: Triant VA, Brown TT, Lee H, Grinspoon SK. Fracture prevalence among human immunodeficiency virus (HIV)- infected versus non-hiv-infected patients in a large US healthcare system. J Clin Endocrinol Metab 8; 93: Womack JA, Goulet JL, Gibert C, et al. Increased risk of fragility fractures among HIV infected compared to uninfected male veterans. PLoS ONE ; 6:e Post FA, Moyle GJ, Stellbrink HJ, et al. Randomized comparison of renal effects, efficacy, and safety with oncedaily abacavir/lamivudine versus tenofovir/emtricitabine, administered with efavirenz, in antiretroviral-naive, HIV- infected adults: 48-week results from the ASSERT study. J Acquir Immune Defic Syndr ; 55: Stellbrink H-J, Orkin C, Arribas JR, et al. Comparison of changes in bone density and turnover with abacavirlamivudine versus tenofovir-emtricitabine in HIV-infected adults: 48-week results from the ASSERT study. Clin Infect Dis ; 5: Bernard AM, Vyskocil AA, Mahieu P, Lauwerys RR. Assessment of urinary retinol-binding protein as an index of proximal tubular injury. Clin Chem 987; 33: Chitra P, Bakthavatsalam B, Palvannan T. Beta- microglobulin as an immunological marker to assess the progression of human immunodeficiency virus infected patients on highly active antiretroviral therapy. Clin Chim Acta ; 4: McComsey G, Kitch D, Daar E, et al. Bone and limb fat outcomes of ACTG A54s, a substudy of ACTG A5: a prospective, randomized, partially blinded phase III trial of or with EFV or ATV/r for initial treatment of HIV- infection. 7th Conference on Retroviruses and Opportunistic Infections 6 9 February, San Francisco, CA, USA. Abstract 6LB. 3. Sax PE, Tierney C, Collier AC, et al. Abacavir lamivudine versus tenofovir emtricitabine for initial HIV- therapy. N Engl J Med 9; 36: Smith KY, Patel P, Fine D, et al. Randomized, double-blind, placebo-matched, multicenter trial of abacavir/lamivudine or tenofovir/emtricitabine with lopinavir/ritonavir for initial HIV treatment. AIDS 9; 3: Accepted 9 May ; published online 3 July 3 Antiviral Therapy