HIV/AIDS CID 2014:59 (1 August) 425

Transcription

1 MAJOR ARTICLE HIV/AIDS Lipid Levels and Changes in Body Fat Distribution in Treatment-Naive, HIV-1 Infected Adults Treated With Rilpivirine or Efavirenz for 96 Weeks in the ECHO and THRIVE Trials Pablo Tebas, 1 Michael Sension, 2 José Arribas, 4 Dan Duiculescu, 5 Eric Florence, 6 Chien-Ching Hung, 8 Timothy Wilkin, 3 Simon Vanveggel, 7 Marita Stevens, 7 and Henri Deckx 7 ; on behalf of the ECHO and THRIVE Study Groups 1 University of Pennsylvania, Philadelphia; 2 Comprehensive Care Center, Fort Lauderdale, Florida; 3 Weill Cornell Medical College, New York, New York; 4 Hospital Universitario La Paz, IdiPAZ, Madrid, Spain; 5 Titu Maiorescu University of Medicine and Dr Victor Babes Hospital for Infectious and Tropical Diseases, Bucharest, Romania; 6 Institute of Tropical Medicine, Antwerp, and 7 Janssen Infectious Diseases BVBA, Beerse, Belgium; and 8 National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei Background. Pooled ECHO/THRIVE lipid and body fat data are presented from the ECHO (Efficacy Comparison in Treatment-Naïve, HIV-Infected Subjects of TMC278 and Efavirenz) and THRIVE (TMC278 Against HIV, in a Once-Daily Regimen Versus Efavirenz) trials. Methods. We assessed the 96-week effects on lipids, adverse events (AEs), and body fat distribution (dual-energy x-ray absorptiometry) of rilpivirine (RPV) and EFV plus 2 nucleoside/nucleotide reverse transcriptase inhibitors (N[t]RTIs) in treatment-naive adults infected with human immunodeficiency virus type 1 (HIV-1). Results. Rilpivirine produced minimal changes in total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglycerides. Compared with RPV, EFV significantly (P <.001) increased lipid levels. Decreases in the TC/HDL-C ratio were similar with RPV and EFV. Background N[t] RTI affected RPV-induced lipid changes; all levels increased with zidovudine/lamivudine (3TC) and abacavir/3tc (except triglycerides, which were unchanged). With emtricitabine/tenofovir, levels of HDL-C were increased, TC and LDL-C were unchanged, and triglycerides were decreased. With EFV, lipid levels increased in each N[t]RTI subgroup (except triglycerides were unchanged with abacavir/3tc). Fewer (P <.001) RPV-treated patients than EFV-treated patients had TC, LDL-C, and triglyceride levels above National Cholesterol Education Program cutoffs. More RPVthan EFV-treated patients had HDL-C values below these cutoffs (P =.02). Dyslipidemia AEs were less common with RPV than with EFV. Similar proportions of patients had a 10% decrease in limb fat (16% with RPV and 17% with EFV). Limb fat was significantly (P <.001) increased to a similar extent (by 12% with RPV and 11% with EFV). At week 96, patients receiving zidovudine/3tc had lost limb fat, and those receiving emtricitabine/ tenofovir had gained it. Conclusions. Over the course of 96 weeks, RPV-based therapy was associated with lower increases in lipid parameters and fewer dyslipidemia AEs than EFV-based treatment. Body fat distribution changes were similar between treatments. The N[t]RTI regimen affected lipid and body fat distribution changes. Keywords. rilpivirine; efavirenz; lipids; lipodystrophy; dual-energy x-ray absorptiometry. Received 26 August 2013; accepted 4 March 2014; electronically published 11 April Presented in part: 18th Conference on Retroviruses and Opportunistic Infections, Boston, Massachusetts, 27 February to 2 March 2011 (poster O-304); 13th International Workshop on Adverse Drug Reactions and Co-morbidities in HIV, Rome, Italy,14 16 July 2011 (abstract O23); 49th Annual Meeting of the Infectious Disease Society of America and the HIV Medicine Association, Boston, October 2011 (posters 403 and 408); 14th International Workshop on Adverse Drug Reactions and Co-morbidities in HIV, Washington, DC, July 2012 (poster P029). Correspondence: Pablo Tebas, MD, University of Pennsylvania, Division of Infectious Diseases/Clinical Trial Unit, 502 Johnson Pavilion/3610 Hamilton Walk, Philadelphia, PA (pablo.tebas@uphs.upenn.edu). Clinical Infectious Diseases 2014;59(3): The Author Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please journals.permissions@oup.com. DOI: /cid/ciu234 HIV/AIDS CID 2014:59 (1 August) 425

2 Two trials, ECHO (Efficacy Comparison in Treatment-Naïve, HIV-Infected Subjects of TMC278 and Efavirenz [1]) and THRIVE (TMC278 Against HIV, in a Once-Daily Regimen Versus Efavirenz [2]), compared the efficacy and safety of the nonnucleoside reverse transcriptase inhibitors (NNRTIs), rilpivirine (once-daily RPV; 25 mg), and efavirenz (once-daily EFV; 600 mg) with those of 2 nucleoside/nucleotide reverse transcriptase inhibitors (N[t]RTIs) in treatment-naive adults infected with human immunodeficiency virus type 1 (HIV-1). Pooled 48- week results demonstrated noninferior efficacy of RPV compared with EFV [3]. RPV was associated with significantly lower incidences of drug-related grade 2 4 adverse events (AEs) and fewer grade 2 4 lipid abnormalities than EFV [1 3]. The 96-week results confirmed and extended the 48-week findings [4]. RPV is licensed for use with other antiretroviral (ARV) agents [5, 6] and as a single-tablet regimen with tenofovir disoproxil fumarate (TDF) and emtricitabine (FTC) [7, 8] in ARV-naive, HIV-1 infected adults with HIV-1 RNA copies/ml. Acceptable tolerability and lack of chronic adverse effects are relevant for long-term success of ARV therapy [9, 10]. HIV infection and its treatment have been associated with dyslipidemia and abnormal body fat distribution [11 13]. These metabolic abnormalities are important because of their association with cardiovascular disease [14 17]. Dyslipidemia has been associated with protease inhibitors (PIs) [9, 18] and some nucleoside reverse transcriptase inhibitors (NRTIs), particularly stavudine and zidovudine (AZT) [9, 12, 19]. NNRTIs tend to have more favorable lipid profiles [12]. Lipid abnormalities (increased levels of low-density lipoprotein-cholesterol [LDL-C] and high-density lipoprotein cholesterol [HDL-C]) have been seen with EFV, although not clearly associated with increased cardiovascular risk [20, 21]. The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study indicated that PIs, not NNRTIs, were associated with an increased risk of myocardial infarction [22]. However, atazanavir was reported not to be associated with an increased risk of cardiovascular or cerebrovascular events [23]. Body fat abnormalities (subcutaneous lipoatrophy and/or accumulation of central fat) have been associated with some NRTIs, particularly thymidine analogues [9, 24], and less commonly with PIs [9, 25 27] and EFV [26, 27]. The current 96-week subanalyses of ECHO/THRIVE provide a comparative analysis of the relative long-term effects of RPV and EFV on serum lipids and body fat redistribution; they also enable the effect of the N[t]RTI background regimen to be evaluated. MATERIALS AND METHODS Trial Designs and Treatment Details of the ECHO (NCT ) and THRIVE (NCT ) trials have been reported. Both were multicenter, multinational, phase 3, randomized, double-blind, double-dummy trials in treatment-naive, HIV-1 infected adults. Patients received either once-daily RPV (25 mg; n = 686) or once-daily EFV (600 mg; n = 682), plus 2 N[t]RTIs, namely, TDF/FTC in ECHO and TDF/FTC (60% in both RPV and EFV groups), AZT/lamivudine (3TC; 30% in RPV and 31% in EFV group), or abacavir (ABC)/3TC (10% in both RPV and EFV groups) in THRIVE [1 4]. Four patients with grade 3 or 4 lipid abnormalities at screening were excluded. The use of lipid-lowering medications (except probucol) was allowed. Assessment of Lipid Parameters Fasting (overnight) lipid levels were evaluated at screening (weeks 6 to 4), at randomization/baseline (day 1), at every visit up to week 96, and 4 weeks after trial completion. Serum total cholesterol (TC), LDL-C, HDL-C, and triglyceride levels were measured. The TC/HDL-C ratio was calculated. LDL-C was calculated as follows: LDL-C = TC HDL-C (triglycerides/5) [28]; if triglyceride levels were >400 mg/dl, LDL-C was measured directly. Monitoring of AEs AEs, including dyslipidemia and lipodystrophy, were monitored by screening and classified according to the Medical Dictionary for Regulatory Activities (version 11.0, available at: meddra.org). The severity of treatment-emergent, lipid-related abnormalities was graded using the Division of AIDS scale [29]. Dual-Energy X-ray Absorptiometry Substudies Limb, trunk, and total body fat distribution were analyzed with whole-body dual-energy x-ray absorptiometry (DXA; GE [Lunar] or Hologic) at baseline, at weeks 48 and 96 (±7 days), and at discontinuation from the main trial if after 64 weeks, in 413 patients enrolled into the body fat distribution substudy. Data Analyses All analyses were done on the intent-to-treat population, using ECHO/THRIVE pooled data after all patients had either received 96 weeks of treatment or discontinued the study early. Data for lipids and body fat were pooled for all patients who received treatment for 96 weeks. We calculated changes from baseline to week 96 by treatment group and by N[t]RTI background regimen for TC, LDL-C, HDL-C, TC/HDL-C ratio, and triglycerides. Data are presented as medians and interquartile ranges (IQRs; first to third quartile), and percentage changes. Differences between treatment groups were analyzed using the Wilcoxon rank sum test. Lipid parameters were categorized according to the US National Cholesterol Education Program (NCEP) [30] or the American Heart Association (AHA) cutoff values. The proportion of patients with 1 abnormal fasting lipid value by NCEP cutoffs at any time was compared between treatments and according to 426 CID 2014:59 (1 August) HIV/AIDS

3 Table 1. Baseline Patient Characteristics and Fasting Lipid Parameters Overall Population Characteristic or Parameter RPV (n = 686) EFV (n = 682) Female, No. (%) 168 (24) 163 (24) Age, median (range), y 36 (18 87) 36 (19 69) Race, No. (%) White 420 (61) 410 (60) Black/African 165 (24) 156 (23) American Asian 78 (11) 97 (14) Other race/not stated 21 (3) 19 (3) Initial background regimen, No. (%) ABC/3TC 35 (5) 33 (5) AZT/3TC 101 (15) 103 (15) TDF/FTC 550 (80) 546 (80) Lipid level, median (IQR), mg/dl a TC ( ) ( ) LDL-C 93.0 ( ) 95.1 ( ) HDL-C 39.1 ( ) 38.0 ( ) Triglycerides ( ) ( ) TC/HDL-C ratio 4.0 ( ) 4.0 ( ) Abbreviations: 3TC, lamivudine; ABC, abacavir; AZT, zidovudine; EFV, efavirenz; FTC, emtricitabine; HDL-C, high-density lipoprotein-cholesterol; IQR, interquartile range; LDL, low-density lipoprotein; RPV, rilpivirine; TC, total cholesterol; TDF, tenofovir disoproxil fumarate. a For lipid parameters, n = 685 and 681 for RPV and EFV (except n = 681 and 679 for LDL-C). N[t]RTI regimen, using the Fisher exact test. The change from baseline to week 96 in the Framingham coronary heart disease (CHD) relative risk score was calculated (10-year risk [31]). For limb, trunk, and total body fat, data were summarized descriptively as absolute (median and IQR) and percentage changes from baseline at weeks 48 and 96. The proportion of participants with decreased limb fat relative to baseline ( 10% [ primary end point], 20%, or 30%) was calculated. Within-treatment and between-group comparisons were made using the Wilcoxon signed rank and rank sum tests, respectively. Body fat data were analyzed according to N[t]RTI regimen. RESULTS Baseline Patient Characteristics and Disposition Table 1 summarizes patient demographic characteristics. Overall, patient demographics were well balanced between the groups. Across N[t]RTI regimens, patient demographics were also generally well balanced, except for the proportions of female and Asian patients, both of which were higher in the AZT/3TCsubgroupthanintheABC/3TCandTDF/FTC subgroups (Table 2). Baseline patient characteristics for DXA substudies were well balanced between the RPV (n = 209) and EFV (n = 204) groups (Table 3). In the DXA substudies, 18 patients in the RPV group and 13 in the EFV group discontinued the trials owing to AEs (3 vs 6 patients, respectively); other reasons (0 vs 2 patients), loss to follow-up (5 vs 1 patient), noncompliance (0 vs 1 patient), virologic failure (8 vs 3 patients), and withdrawal of consent (2 vs 0 patients). By week 96, no withdrawals were due to treatment-emergent dyslipidemia or lipodystrophy [4]. Use of Lipid-Modifying Agents At baseline, lipid-modifying therapy use was low, with RPV treatment in 11 patients (1.6%) and EFV in 8 (1.2%). At week 96, significantly fewer RPV-treated patients (18; 2.6%) received lipid-modifying drugs compared with the EFV group (38 patients; 6%; P =.006). The most common agents were simvastatin, atorvastatin, and rosuvastatin in the ECHO trial and gemfibrozil and atorvastatin in the THRIVE trial. For patients with dyslipidemia (at least grade 2 4 for TC, LDL-C, or triglycerides [29]), 8 of 83 (10%) and 32 of 194 (16%) in the RPV and EFV groups, respectively, received lipid-modifying drugs (P =.19; Fisher exact test). Changes in Fasting Lipid Parameters Baseline fasting lipid parameters were similar between the groups (Table 1). Figure 1 shows the changes in lipid levels by treatment. RPV resulted in minimal changes from baseline in TC (Figure 1A), LDL-C (Figure 1B), HDL-C (Figure 1C), and triglyceride (Figure 1D) levels at week 96, whereas these levels were all increased with EFV. The differences between RPV and EFV were statistically significant for all of these lipid parameters. By week 96, the median (IQR) TC increases were 2.0 mg/dl ( 14.3 to 22.0; 1.3% change) for RPV and 25.9 mg/dl ( ; 16.8% change) for EFV (P <.001). The increases in LDL-C were 0.0 mg/dl ( 13.9 to 14.3; 0% change) and 12.4 mg/dl ( 4.6 to 31.0; 12.7% change) for RPV and EFV, respectively (P <.001). RPV increased HDL-C by 3.9 mg/dl (IQR, 1.0 to 9.7; 9.7% change), compared with an increase of 10.0 mg/dl (4.6 to 16.2; 26.0% change) for EFV (P <.001). Triglyceride levels were decreased slightly with RPV ( 7.1 mg/dl; IQR, 33.8 to 17.9; 8.7% change) and increased with EFV (5.0 mg/dl; IQR, 24.8 to 43.2; 5.7% change; P <.001). The TC/HDL-C ratio (Figure 1E) was decreased from baseline at week 96, with no difference (P =.17) between RPV ( 0.24; IQR, 0.75 to 0.16; 6.7% change) and EFV ( 0.32; IQR, 0.90 to 0.19; 8.8% change). Lipid Values Outside NCEP and AHA Categories Baseline fasting lipid parameters for patients with values within NCEP categories were similar between the treatment groups HIV/AIDS CID 2014:59 (1 August) 427

4 428 CID 2014:59 (1 August) HIV/AIDS Table 2. Baseline Patient Characteristics and Fasting Lipid Parameters by N[t]RTI Background Regimen N[t]RTI Background Regimen ABC/3TC AZT/3TC TDF/FTC Characteristic or Parameter RPV (n = 35) EFV (n = 33) RPV (n = 101) a EFV (n = 103) RPV (n = 550) b EFV (n = 546) c Female, No. (%) 7 (20) 6 (18) 40 (40) 42 (41) 121 (22) 115 (21) Age, median (range), y 39 (22 59) 36 (23 63) 34 (19 57) 33 (19 54) 36 (18 78) 36 (19 69) Race, No. (%) White 21 (60) 25 (76) 51 (50) 51 (50) 348 (63) 334 (61) Black/African American 10 (28) 8 (24) 21 (21) 20 (19) 134 (24) 128 (23) Asian 2 (6) 0 (0) 22 (22) 27 (26) 54 (10) 70 (13) Other d 2 (6) 0 (0) 7 (7) 5 (5) 5 (1) 7 (1) Not allowed to ask/missing 0 (0) 0 (0) 0 (0) 0 (0) 9 (2) 7 (1) Lipid parameter, median (IQR) mg/dl TC ( ) ( ) ( ) ( ) ( ) ( ) LDL-C ( ) 92.0 ( ) 89.5 ( ) 90.9 ( ) 94.0 ( ) 95.9 ( ) HDL-C 37.0 ( ) 40.0 ( ) 39.8 ( ) 39.1 ( ) 39.4 ( ) 38.0 ( ) Triglycerides ( ) ( ) 95.6 ( ) ( ) ( ) ( ) TC/HDL-C ratio 4.1 ( ) 4.3 ( ) 3.9 ( ) 3.9 ( ) 4.0 ( ) 4.2 ( ) Abbreviations: 3TC, lamivudine; ABC, abacavir; AZT, zidovudine; EFV, efavirenz; FTC, emtricitabine; HDL-C, high-density lipoprotein-cholesterol; IQR, interquartile range; LDL, low-density lipoprotein; N[t]RTI, nucleoside/ nucleotide reverse transcriptase inhibitor; RPV, rilpivirine; TC, total cholesterol; TDF, tenofovir disoproxil fumarate. a n = 100 for LDL-C. b n = 549 for lipid parameters except for LDL-C (n = 546). c n = 545 for lipid parameters except for LDL-C (n = 543). d American Indian or Alaskan Native, Native Hawaiian or other Pacific Islander, Asian/black, or African American/white.

5 Table 3. Baseline Patient Characteristics in DXA Substudies DXA Substudies Characteristic or Parameter RPV (n = 209) EFV (n = 204) Female, No. (%) 40 (19) 41 (20) Age, median (IQR), y 36 (29 42) 36 (28 43) Limb fat, median, kg a 6.6 ( ) b 6.9 ( ) c Background regimen, No. (%) ABC/3TC 9 (4) 6 (3) AZT/3TC 29 (14) 23 (11) TDF/FTC 171 (82) 175 (86) Abbreviations: 3TC, lamivudine; ABC, abacavir; AZT, zidovudine; DXA, dualenergy x-ray absorptiometry; EFV, efavirenz; FTC, emtricitabine; IQR, interquartile range; RPV, rilpivirine; TDF, tenofovir disoproxil fumarate. a There were too few patients for an ABC/3TC subset evaluation. b n = 205. c n = 201. (Table 4, footnote). In patients whose lipid parameters were within NCEP cutoffs at baseline, a significantly (P <.001) lower proportion of RPV-treated than EFV-treated patients had abnormal values for TC, LDL-C, and triglycerides any time during treatment (Table 4). For HDL-C, a significantly (P =.02) higher percentage of RPV-treated patients than EFV-treated patients had values below the NCEP cutoff. For patients with an optimal TC/HDL-C ratio ( 3.5 according to the AHA cutoff) at baseline, 82% (148 of 180) in the RPV group and 83% (120 of 145) in the EFV group still had an optimal ratio at week 96, whereas 2% (3 of 180) and 3% (5 of 145), respectively, had a TC/HDLC ratio >5. Changes in Fasting Lipid Parameters by N[t]RTI Background Regimen Baseline fasting lipid parameters were similar between the RPV and EFV groups for each N[t]RTI background regimen (Table 2). With RPV, changes in lipid parameters at week 96 from baseline varied with N[t]RTI background (Figure 2). Median TC, HDL-C, LDL-C, and triglyceride levels were increased with RPV plus AZT/3TC. With ABC/3TC plus RPV, triglyceride levels were unaffected, and the other lipid parameters were increased. Only HDL-C levels were increased with RPV plus TDF/FTC, whereas TC and LDL-C levels were almost unchanged and triglyceride levels were decreased. With EFV, median lipid parameters at week 96 were all increased from baseline in each N[t]RTI subgroup, except for triglycerides in the ABC/3TC group, which were unchanged (Figure 2). All observed increases in lipid parameters were lower with RPV than with EFV (Figure 2). CHD Risk The change from baseline to week 96 in the Framingham CHD relative risk score (10-year risk) based on fasting TC data was similar in RPV and EFV groups; the mean scores were 0.04 (95% confidence interval [CI],.105 to.019) for RPV (n = 407) and 0.01 ( ) for EFV (n = 408). With fasting LDL-C data, the mean scores were 0.15 (95% CI,.205 to.088) for RPV (n = 410) and 0.10 (.168 to.032) for EFV (n = 407). Lipid-Related Laboratory Abnormalities and AEs For laboratory abnormalities of TC, LDL-C, and triglycerides, the overall incidences of all grades and grade 3 or 4 treatment-emergent abnormalities were significantly lower with RPV than with EFV (see Supplementary Table 1). The incidences of grade 3 or 4 treatment-emergent TC, LDL-C, and triglyceride values as laboratory abnormalities by N[t]RTI background regimen are in Supplementary Table 2. Incidences of all of these abnormalities were significantly lower with RPV than EFV in the TDF/FTC subgroup and were similar between treatment groups in the AZT/3TC subgroup. In the ABC/3TC subgroup, the only significant difference was a lower incidence of grade 3 or 4 TC abnormalities with RPV than with EFV. The AEs of dyslipidemia reported as treatment-related AEs (occurring in 1% of patients) were less common in patients receiving RPV than in those receiving EFV; the AEs included hypercholesterolemia (4 [0.6%] vs 10 [1.5%] patients), hyperlipidemia (2 [0.3%] vs 10 [1.5%]), and hypertriglyceridemia (2 [0.3%] vs 9 [1.3%]). Most of these AEs were severity grade 1 or 2. Of grade 3 or 4 hyperlipidemia occurred only in the EFV group (5 patients [0.7%]). Changes in Body Fat Distribution At 96 weeks, there were no significant differences in the proportions of patients with a decrease from baseline in limb fat of 10% (RPV group, 16%; EFV group, 17%), 20% (RPV, 7%; EFV, 10%), or 30% (RPV, 3.5%; EFV, 4.4%). Overall, with both treatments, median limb (Figure 3A) and trunk fat (Supplementary Figure 1) weresignificantly (P <.001) increased from baseline at weeks 48 and 96. At week 96, the median percentage increases from baseline in limb, trunk, and total body fat were 12%, 16%, and 14%, respectively, with RPV and 11%, 14%, and 11%, respectively, with EFV; all increases differed significantly from baseline but not between treatments. With TDF/FTC, limb (Figure 3B), trunk (data not shown), and total body fat (data not shown) were significantly increased from baseline at week 96 within the groups, with no differences between RPV and EFV. A nonsignificant decrease in median limb fat was observed for RPV and EFV in the AZT/3TC subgroup at week 96 (Figure 3B). Small increases from baseline were seen in trunk fat (statistically significant with RPV at weeks 48 [P =.01] and 96 [P =.02] but not significant with EFV) and total body fat (not significant for either drug) at week 96 for patients in the AZT/3TC subgroup (data not HIV/AIDS CID 2014:59 (1 August) 429

6 Figure 1. Median changes in fasting lipid parameters from baseline to week 96, with interquartile ranges. A, Total cholesterol (TC). B, Low-density lipoprotein cholesterol (LDL-C). C, High-density lipoprotein cholesterol (HDL-C). D, Triglycerides. E, TC/HDL-C ratio. P values denote comparison between rilpivirine (RPV) and efavirenz (EFV) groups at week 96 (Wilcoxon rank sum test, preplanned analysis). shown). DXA data were available for only 10 patients receiving ABC/3TC, so these observations are too limited to allow conclusions to be drawn (data not shown). DISCUSSION Our subanalyses of ECHO/THRIVE show that RPV (25 mg once daily) is associated with fewer lipid changes than EFV (600 mg once daily), when both are coadministered with 2 N [t]rtis in treatment-naive, HIV-1 infected adults. Both drugs are associated with comparable body fat changes from baseline to week 96. The background N[t]RTI regimen affected changes in lipids and body fat distribution. These pooled analyses are particularly interesting because they involve a large number of patients over a prolonged follow-up period, allowing a direct comparison of the long-term effects of RPV and EFV on these parameters. EFV was selected as the control, because guidelines recommend EFV combined with 2 N[t]RTIs (TDF/FTC or ABC/3TC) as a preferred regimen for ARV treatment-naive, HIV-1 infected adults [9, 10]. Our results demonstrated that RPV plus 2 N[t]RTIs had less effect on serum lipid levels, the proportion of patients with 430 CID 2014:59 (1 August) HIV/AIDS

7 Table 4. Patients With 1 Abnormal Fasting Lipid Value During Treatment a Patients, % With Abnormal Value (No./Total Sampled) Lipid Parameter b RPV EFV P Value c TC (above cutoff) 22 (134/599) 52 (305/581) <.001 LDL-C (above cutoff) 21 (124/602) 44 (256/581) <.001 HDL-C (below cutoff) 58 (157/270) 47 (106/224).02 Triglycerides (above cutoff) 40 (207/523) 55 (271/497) <.001 Abbreviations: EFV, efavirenz; HDL-C, high-density lipoprotein-cholesterol; LDL, low-density lipoprotein; NCEP, National Cholesterol Education Program; RPV, rilpivirine; TC, total cholesterol. a Abnormal values were based on NCEP cutoffs. Patients included are those whose lipid parameters were within NCEP cutoffs at baseline. The median values at baseline for RPV vs EFV groups were as follows for patients with normal fasting lipid values: TC, (interquartile range, ) vs ( ) mg/dl; LDL-C, 89.0 ( ) vs 90.5 ( ) mg/dl; HDL-C, 49.5 ( ) vs 48.7 ( ) mg/dl; and triglycerides, 90.3 ( ) vs 92.1 ( ) mg/dl. b NCEP cutoff values are as follows: TC, <200 mg/dl; LDL-C, <130 mg/dl; HDL-C, >40 mg/dl (male patients) and >50 mg/dl (female patients); and triglycerides, <150 mg/dl. c RPV vs EFV. abnormal lipid values, and lipid-related AEs than EFV. This difference was reflected in the number of patients requiring hypolipidemic drugs for dyslipidemia, because fewer RPV-treated patients received such intervention (given during the doubleblinded phase of the study) than in the EFV group. Although more EFV-treated than RPV-treated patients received hypolipidemic drugs, serum lipid levels were still higher with EFV than with RPV. The effect of EFV plus 2 N[t]RTIs on serum lipid parameters in our ECHO/THRIVE analyses are consistent with those from the Gilead 934 [32], STARTMRK [33], and Adult Clinical Trials Group (ACTG) 5202 [34] trials. Moreover, a less favorable lipid profile of EFV-based therapy has been reported compared with other NNRTIs, that is, NVP- and ETR-based therapy [35 37]. Collectively, such comparative studies suggest that EFV is associated with more lipid perturbations than other available NNRTIs in HIV-1 infected adults. Our analyses also confirm that particular N[t]RTIs affect the lipid profile of a given ARV regimen in HIV-1 infected patients [12, 38]. We were able to compare the impact of TDF/FTC, AZT/3TC, and ABC/3TC when combined with RPV or EFV. However, data for ABC/3TC should be interpreted with caution because only 68 patients received this combination. All 4 serum lipid parameters were increased with RPV/AZT/3TC, as were TC, HDL-C, and LDL-C with RPV/ABC/3TC, whereas RPV/ TDF/FTC increased HDL-C and decreased triglyceride levels. However, all 3 N[t]RTI combinations with EFV resulted in increased serum lipid parameters (except triglycerides, which were unaffected with EFV plus ABC/3TC). The lipid increases with RPV plus 2 N[t]RTIs were smaller than with EFV plus 2 N Figure 2. Median changes in fasting lipid parameters from baseline to week 96, with interquartile ranges, by nucleoside/nucleotide reverse transcriptase inhibitor background regimen. Median changes in TC/HDL-C ratios are shown for completeness; the unit of measure does not apply. Abbreviations: ABC/ 3TC, abacavir/lamivudine; AZT/3TC, zidovudine/lamivudine; EFV, efavirenz; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; RPV, rilpivirine; TC, total cholesterol; TDF/FTC, tenofovir disoproxil fumarate/emtricitabine. HIV/AIDS CID 2014:59 (1 August) 431

8 Figure 3. Changes from baseline in limb fat. A, Pooled data. B, By nucleoside/nucleotide reverse transcriptase inhibitor background regimen. Withingroup changes at weeks 48 and 96 after baseline were significant (P <.001) overall and for tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) and not significant for zidovudine/lamivudine (AZT/3TC); differences between treatment groups were not significant at each time point (Wilcoxon rank sum test). Abbreviations: EFV, efavirenz; IQR, interquartile range; RPV, rilpivirine. [t]rtis, and differences between these NNRTI/N[t]RTI combinations were more pronounced with TDF/FTC. Currently, for treatment-naive HIV-1 infected patients, the preferred background regimens are TDF/FTC in US guidelines [9] and TDF/FTC or ABC/3TC in European guidelines [10]. Overall, RPV had less impact on lipid profiles than EFV, independent of the backbone used. Dyslipidemia contributes to an increased CHD risk [14, 15]. At 96 weeks, although the HDL-C levels increased with RPV and EFV compared with baseline, this increase was lower with RPV, resulting in significantly more RPV-treated patients having HDL- C values below the NCEP cutoff considered beneficial in lowering CHD risk. However, the TC/HDL-C ratio, a key lipoprotein predictor of future CHD, declined from baseline to a similar extent for both RPV and EFV. Although dyslipidemia (changes in lipids and AEs) was seen with RPV and EFV in our analyses, the implication of these differences for CHD is currently unclear. The Framingham CHD relative risk scores based on fasting TC and LDL-C perturbations did not change significantly for either group, probably because of the categorization of cholesterol values used in the calculation of the Framingham risk score. Limb lipoatrophy is typically seen with thymidine analogue NRTIs [9, 24], and less so with PIs [9, 25 27] and NNRTIs [26, 27]. The DXA substudy showed that the proportion of patients with a 10% decrease in limb fat from baseline at week 96 was small and similar between RPV and EFV groups. Other studies of EFV plus 2 NRTIs found a higher incidence of patients with 20% loss in extremity fat than the 10% seen in our study (eg, 32% in ACTG A5142 [27] and 34% in M [39]). This difference may be explained by the more frequent use of NRTIs, such as stavudine and AZT in ACTG A5142 [27] and AZT/3TC in M [39]. In our analyses, limb fat was increased to a similar extent with RPV and EFV. Theincrease(11%)inlimbfatwithEFVplus2N[t]RTIsat week 96 is higher than that seen in other studies (eg, 1.4% in ACTG A5142 [27] and 2.4% in ACTG 5005s [40]). This difference probably reflects the NRTI contribution of the regimens evaluated. In our subanalyses, limb fat was increased with TDF/ FTC but decreased over time with AZT/FTC in both RPV and EFV groups. Increased trunk fat is part of ARV-induced lipodystrophy [11, 14, 15, 18] and is probably a manifestation of the return-to-health phenomenon. RPV and EVF had similar 432 CID 2014:59 (1 August) HIV/AIDS

9 effects on increasing trunk fat in our subanalyses of ECHO/ THRIVE. In conclusion, the 96-week pooled analyses of ECHO/ THRIVE demonstrate that RPV results in significantly lower increases in serum lipid parameters and fewer treatmentemergent lipid abnormalities than EFV, whereas it was associated with similar changes in Framingham CHD risk scores and body fat distribution changes. These analyses support the overall findings from ECHO/THRIVE, indicating that RPV has a more favorable safety profile than EFV at week 96 [4]. Supplementary Data Supplementary materials are available at Clinical Infectious Diseases online ( Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author. Notes Acknowledgments. We are very grateful to the patients and their families for their participation and support during the trials, the ECHO and THRIVE study teams from Janssen, the study center staff and principal investigators, and members of the Janssen RPV team, in particular Ralph DeMasi, Peter Williams, and Eric Wong, for their input. Jackie Phillipson of Gardiner-Caldwell Communications, Macclesfield, United Kingdom, provided medical writing support and assistance in coordinating and collating author contributions, with funding by Janssen. Financial support. This work was supported by Janssen Research & Development. Potential conflicts of interest. P. T. has received consulting fees from Merck, Gilead, AstraZeneca, was a member of an adjudication committed for GlaxoSmithKline (GSK), and was on a data safety monitoring board for Cytheris. M. Sension has served on speaker bureaus for Gilead Sciences, Janssen, and Bristol-Myers Squibb (BMS) and acted as a consultant for Janssen; he has received research support and speaker honoraria from Janssen and Gilead Sciences, speaker honoraria from BMS, and consulting fees from Janssen. J. A. has received grant research support, advisory, and speaker fees from Janssen, ViiV Healthcare, and BMS. E. F. has served on advisory boards for and has received educational and research funding and consulting fees from Janssen. C. C. H. has received research support from Janssen. T. W. has received research grant support (paid to Weill-Cornell) from Janssen, Gilead, and GSK and has served as an ad hoc consultant to Merck. S. V., M. Stevens, and H. D. are full-time employees of Janssen. D. D. reports no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. References 1. Molina JM, Cahn P, Grinsztejn B, et al. Rilpivirine versus efavirenz with tenofovir and emtricitabine in treatment-naïve adults infected with HIV-1 (ECHO): a phase 3 randomised double-blind active-controlled trial. Lancet 2011; 378: Cohen CJ, Andrade-Villanueva J, Clotet B, et al. Rilpivirine versus efavirenz with two background nucleoside or nucleotide reverse transcriptase inhibitors in treatment-naïve adults infected with HIV-1 (THRIVE): a phase 3, randomised, non-inferiority trial. Lancet 2011; 378: Cohen C, Molina JM, Cahn P, et al. Efficacy and safety of rilpivirine (TMC278) versus efavirenz at 48 weeks in treatment-naïve, HIV-1-infected patients: pooled results from the phase 3 double-blind, randomized ECHO and THRIVE trials. J Acquir Immune Defic Syndr 2012; 60: Cohen C, Molina JM, Cassetti I, et al. Week 96 efficacy and safety of rilpivirine in treatment-naïve, HIV-1 patients in two phase III randomized trials. AIDS 2013; 27: Edurant (rilpivirine) [tablets]. Full prescribing information. Janssen- Cilag, Available at: EDURANT-PI.pdf. Accessed 15 May Edurant (rilpivirine) tablets, EU summary of product characteristics. Janssen-Cilag, February Available at: EMC/medicine/25490/SPC/Edurant+25+mg+tablets/#DOCREVISION. Accessed 15 May COMPLERA (emtricitabine/rilpivirine/tenofovir disoproxil fumarate) tablets. Full prescribing information. Gilead Sciences, January Available at: /media/files/pdfs/medi cines/hiv/complera/complera_pi.pdf. Accessed 15 May EVIPLERA 200 mg/25 mg/245 mg film-coated tablets. Summary of Product Characteristics. Gilead Sciences, February Available at: Accessed 15 May Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. US Department of Health and Human Services (updated February 2013). Available at: lvguidelines/adultandadolescentgl.pdf. Accessed 15 May EACS guidelines: EACS European Guidelines for treatment of HIV infected adults in Europe. Version 6.1 (Updated November 2012). Available at: pdf. Accessed 15 May Lichtenstein KA, Ward DJ, Moorman AC, et al. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. AIDS 2001; 15: Estrada V, Portilla J. Dyslipidemia related to antiretroviral therapy. AIDS Rev 2011; 13: Domingo P, López-Aldeguer J, Villaroya F, et al. Fat redistribution syndromes associated with HIV-1 infection and combination antiretroviral therapy. AIDS Rev 2012; 14: Grinspoon S, Carr A. Cardiovascular risk and body-fat abnormalities in HIV-infected adults. N Engl J Med 2005; 352: Anuurad E, Semrad A, Berglund L. Human immunodeficiency virus and highly active antiretroviral therapy-associated metabolic disorders and risk factors for cardiovascular disease. Metab Syndr Relat Disord 2009; 7: Friis-Møller N, Weber R, Reiss R, et al. Cardiovascular disease risk factors in HIV patients association with antiretroviral therapy: results from the DAD study. AIDS 2003; 17: Worm SW, Kamara DA, Reiss P, et al. Elevated triglycerides and risk of myocardial infarction in HIV-positive persons. AIDS 2011; 25: Flint OP, Noor MA, Hruz PW, et al. The role of protease inhibitors in the pathogenesis of HIV-associated lipodystrophy: cellular mechanisms and clinical implications. Toxicol Pathol 2009; 37: Crane HM, Grunefeld C, Willig JH, et al. Impact of NRTIs on lipid levels among a large HIV-infected cohort initiating antiretroviral therapy in clinical care. AIDS 2011; 24: Worm SW, Sabin C, Weber R, et al. Risk of myocardial infarction in patients with HIV infection exposed to specific individual antiretroviral drugs from the 3 major drug classes: the data collection on adverse events of anti-hiv drugs (D:A:D) study. J Infect Dis 2010; 201: Maggi P, Bellacosa C, Carito V, et al. Cardiovascular risk factors in patients on long-term treatment with nevirapine- or efavirenz-based regimens. J Antimicrob Chemother 2011; 66: DAD Study GroupFriis-Møller N, Reiss P, et al. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med 2007; 356: HIV/AIDS CID 2014:59 (1 August) 433

10 23. Monforte AD, Reiss P, Ryom L, et al. Atazanavir is not associated with an increased risk of cardio or cerebrovascular disease events. AIDS 2013; 27: Mallal SA, John M, Moore CB, et al. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 2000; 14: Tsiodras S, Mantzoros C, Hammer S, et al. Effects of protease inhibitors on hyperglycemia, hyperlipidemia, and lipodystrophy: a 5-year cohort study. Arch Intern Med 2000; 160: Pérez-Molina JA, Domingo P, Martínez E, et al. The role of efavirenz compared with protease inhibitors in the body fat changes associated with highly active antiretroviral therapy. J Antimicrob Chemother 2008; 62: Haubrich RH, DiRienzo AG, Komarow L, et al. Metabolic outcomes in a randomized trial of nucleoside, nonnucleoside and protease inhibitorsparing regimens for initial HIV treatment. AIDS 2009; 23: Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972; 18: Division of AIDS table for grading the severity of adult and pediatric adverse events. Version 1.0. December 2004 (clarification August 2009). Available at: macovigilance/table_for_grading_severity_of_adult_pediatric_ Adverse_Events.pdf. Accessed 22 August Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002; 106: D Agostino RB Sr, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008; 117: Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med 2006; 354: Lennox JL, DeJesus E, Lazzarin A, et al. Safety and efficacy of raltegravir-based versus efavirenz-based combination therapy in treatmentnaïve patients with HIV-1 infection: a multicentre, double-blind randomised controlled trial. Lancet 2009; 374: Sax PE, Tierney C, Collier AC, et al. Abacavir-lamivudine versus tenofovir-emtricitabine for initial HIV-1 therapy. N Engl J Med 2009; 361: van Leth F, Phanuphak P, Stroes E, et al. Nevirapine and efavirenz elicit different changes in lipid profiles in antiretroviral-therapy-naïve patients infected with HIV-1. PLoS Med 2004; 1:e Young J, Weber R, Rickenbach M, et al. Lipid profiles for antiretroviralnaïve patients starting PI- and NNRTI-based therapy in the Swiss HIV cohort study. Antivir Ther 2005; 10: Fätkenheuer G, Duvivier C, Rieger A, et al. Lipid profiles for etravirine versus efavirenz in treatment-naïve patients in the randomized, double-blind SENSE trial. J Antimicrob Chemother 2012; 67: Feeney ER, Mallon PW. HIV and HAART-associated dyslipidemia. Open Cardiovasc Med J 2011; 5: Cameron DW, da Silva BA, Arribas JR, et al. A 96-week comparison of lopinavir-ritonavir combination therapy followed by lopinavir-ritonavir monotherapy versus efavirenz combination therapy. J Infect Dis 2008; 198: Dubé MP, Komarow L, Mulligan K, et al. Long-term body fat outcomes in antiretroviral-naive participants randomized to nelfinavir or efavirenz or both plus dual nucleosides. dual x-ray absorptiometry results from A5005s, a substudy of Adult Clinical Trials Group 384. J Acquir Immune Defic Syndr 2007; 45: CID 2014:59 (1 August) HIV/AIDS