The introduction of antiretroviral therapy has profoundly improved treatment outcomes in HIV-infected individuals.1 Although data on long-term treatment with antiretrovirals are becoming available from cohort studies, extended follow-up of randomized controlled trials is unusual but is of key importance, particularly for regimens recommended for initial therapy by HIV treatment guidelines.2,3 Study 934 compared the efficacy and tolerability of 2 antiretroviral regimens consisting of efavirenz (EFV) once daily with twice-daily fixed-dose zidovudine/lamivudine (ZDV/3TC) or with once-daily tenofovir disoproxil fumarate/emtricitabine (TDF/FTC). We report results of this clinical trial through 144 weeks of follow-up.4,5
Details of the study design and methodology have been previously reported.4,5 Antiretroviral-naive adults with HIV-1 RNA levels >10,000 copies/mL regardless of CD4 cell count and with an estimated glomerular filtration rate (GFR) of >50 mL/min by the Cockcroft-Gault (CG) equation were randomized to receive a once-daily regimen of 600-mg tablets of EFV, 300 mg of TDF, and 200 mg of FTC (taken without regard to meals) or 600-mg tablets of EFV and a fixed-dose of ZDV/3TC (300/150 mg) twice daily.
Clinical examinations and laboratory analyses were conducted at weeks 2, 4, and 8 and were then conducted every 8 weeks through week 48 and every 12 weeks from week 48 through week 144. A subset of patients underwent whole-body dual-energy x-ray absorptiometry (DEXA) assessments of body fat composition at week 48, and all patients underwent DEXA at 96 and 144 weeks. At 96 weeks, subjects on TDF, FTC, and EFV were switched to fixed-dose TDF/FTC (300/200 mg once daily) and EFV for a simplified dosing regimen.
The primary efficacy endpoint for this 144-week study was the achievement and maintenance of confirmed HIV-1 RNA levels <400 copies/mL at week 48 as defined by the US Food and Drug Administration (FDA) time-to-loss-of-virologic-response (TLOVR) algorithm.4,5 A finding of noninferiority required the lower bound of the 95% confidence interval for the difference in efficacy between the 2 arms to be no lower than −13%. Secondary objectives were to assess the noninferiority of TDF/FTC and EFV compared with ZDV/3TC and EFV at 96 and 144 weeks and changes in CD4 cell count.
Patients excluded from all efficacy analyses included 22 patients with baseline resistance to nonnucleoside reverse transcriptase inhibitors (NNRTIs; 11 in each arm). Excluded from the secondary efficacy analyses were 24 patients (12 from each group) who completed the week 48 study with an HIV RNA level <400 copies/mL but did not consent to participate from week 48 through 96 and 7 patients who completed the week 96 study with an HIV RNA level <400 copies/mL (5 in the TDF/FTC group and 2 in the ZDV/3TC group) but did not consent to participate from week 96 through week 144. We report the efficacy results at week 144 for 456 randomized patients (the efficacy population). The treatment groups were compared using statistical analyses previously described.4,5 Analyses of CD4 cell count and safety laboratory data included patients who were on the randomized regimen. All reported P values are 2-sided and not adjusted for multiple testing.
All patients who received at least 1 dose of study medication were included in the safety analysis, which evaluated events that occurred after initiation of the study regimen through 30 days after discontinuation of the regimen. Genotypic and phenotypic resistance analyses were performed as previously detailed.
Detailed characteristics of the study population have been published previously and were similar in the TDF/FTC and ZDV/3TC treatment arms, with an overall median age of 37 years, proportion female of 14%, median log10 viral load of 5.0, and median CD4 count of 237 cells/mm3.4,5 Significantly more patients achieved and maintained virologic suppression to an HIV RNA level <400 copies/mL through week 144 on TDF/FTC and EFV (161 [71%] of 227 patients) compared with those on ZDV/3TC and EFV (133 [58%] of 229 patients; P = 0.004) (Fig. 1). There was a trend toward greater virologic suppression to <50 copies/mL in the TDF/FTC and EFV group than in the ZDV/3TC group (146 [64%] of 227 patients vs. 130 [56%] of 231 patients; P = 0.08). Patients in the TDF/FTC arm experienced an increase in CD4 count of 312 cells/mm3 from baseline compared with an increase of 271 cells/mm3 in the ZDV/3TC arm, a trend that did not maintain statistical significance at 144 weeks (P = 0.09). Mean adherence levels as measured by pill counts were similar between the 2 groups (89% in the TDF/FTC arm compared with 87% in the ZDV/3TC arm; P = 0.15). Resistance to EFV was the most common pattern of resistance that developed in both arms, and of the 34 patients who developed resistance to EFV, 26 (77%) developed the K103N mutation in the reverse transcriptase gene (RT). There was no significant difference between the 2 arms with respect to the frequency of development of EFV-R (or of the K103N mutation, specifically). In contrast, significantly more patients in the ZDV/3TC arm developed the M184V/I mutation (10 of 243 patients vs. 2 of 244 patients; P = 0.02). No patient developed the K65R mutation. During 144 weeks of follow-up, more patients in the ZDV/3TC arm (105 patients [41%]) discontinued their study regimen than in the TDF/FTC and EFV group (75 patients [29%]; P = 0.004). Most of the discontinuations occurred before 96 weeks (detailed descriptions of which have been previously published), whereas there were only 4 discontinuations in the TDF/FTC arm and 10 in the ZDV/3TC arm between 96 and 144 weeks.4,5 Cumulatively, there were more discontinuations attributable to virologic failure in the ZDV/3TC arm (14 patients [6%]) than in the TDF/FTC arm (5 patients [2%]; P = 0.038), although only 2 patients discontinued for virologic failure between 96 and 144 weeks (both in the ZDV/3TC arm). Twenty-nine patients (11%) of the ZDV/3TC arm compared with 12 (5%) of the TDF/FTC arm (P = 0.01) discontinued their study regimens because of adverse events over 144 weeks, although only 1 patient (in the ZDV/3TC arm) discontinued because of an adverse event from week 96 through week 144. Other reasons for discontinuation were balanced between the TDF/FTC and ZDV/3TC arms (eg, loss to follow-up: 10% and 11% overall and 3% and 4% between 96 and 144 weeks).
Among the safety population of 511 patients who received at least 1 dose of study medications, grade 2 through 4 adverse events (occurring in >3% of patients regardless of relation to study drug) were reported in 60% of the TDF/FTC and EFV group and 59% of the ZDV/3TC and EFV group and were similar to those reported in detail at 96 weeks.6 There were 6 bone fractures in the TDF/FTC group compared with 8 in the ZDV/3TC group. All fractures were secondary to trauma, and none were associated with study drugs according to the investigators. Hyperpigmentation was reported in 15 patients in the TDF/FTC arm, of which 8 cases were assessed by a dermatologist as not related to study drug and 6 as related or probably related to study drug (1 patient refused referral). Hyperpigmentation was reported in 9 patients in the ZDV/3TC arm, of which 1 case was assessed by a dermatologist as not related to study drug and 5 were assessed as related or probably related to study drug (2 patients refused referral and 1 report was unavailable). Thirteen cases of hyperpigmentation involved the hands and/or feet, 8 cases involved other parts of the body, and 3 cases involved hands and/or feet and another part of the body. All episodes of hyperpigmentation were grade 1 except for 2 grade 2 events (1 in each arm).
Grade 2 through 4 laboratory abnormalities occurred in 67% of patients in the TDF/FTC arm compared with 68% in the ZDV/3TC arm. Only 3% and 4%, respectively, of these laboratory abnormalities occurred between 96 and 144 weeks. Grade 3 and 4 laboratory abnormalities occurring in 5% or greater of patients in the TDF/FTC or ZDV/3TC arm included elevated creatine kinase (9% vs. 7%), triglycerides (5% vs. 3%), neutropenia (3% vs. 5%), and amylase (8% vs. 4%).
Among renal parameters, 1 patient had confirmed grade 1 elevation of serum creatinine over 144 weeks in the TDF/FTC arm compared with the ZDV/3TC arm, in which 2 patients had confirmed grade 2 elevations in creatinine. Grade 1 and 2 proteinuria occurred at similar rates between the arms (6% and 4%, respectively, in the TDF/FTC arm and 4% in both arms), and no cases of grade 3 or 4 proteinuria occurred. The median GFR decreased in both groups over 144 weeks of follow-up as estimated by CG equation and in the TDF/FTC arm by the modification of diet in renal disease (MDRD) equation. By the CG equation, the median baseline GFR was 121 mL/min for both arms. At week 144, it was 115 mL/min in the TDF/FTC arm compared with 118 mL/min in the ZDV/3TC arm (P = 0.18). By the MDRD equation (in mL/min per 1.73 m2), the median baseline GFR was 110 for the TDF/FTC arm and 105 for the ZDV/3TC arm. At week 144, it was 98 in the TDF/FTC arm compared to 106 mL/min per 1.73 m2 in the ZDV/3TC arm (P < 0.001). Most of the decline in GFR by the MDRD equation in the TDF/FTC arm took place in the first 96 weeks, with minimal change between 96 and 144 weeks. No patients in either group discontinued because of renal events over 144 weeks.
As seen at 48 and 96 weeks, patients in the ZDV/3TC arm had significantly less limb fat at 144 weeks (median = 5.4 kg) than patients in the TDF/FTC arm (median = 7.9 kg; P < 0.001) as measured by whole-body DEXA scanning (Fig. 2). From 96 weeks to 144 weeks, 15% of patients in the ZDV/3TC arm experienced >20% decrease in limb fat compared with 5% in the TDF/FTC arm (P = 0.01). For the subset of patients with a DEXA scan at 48 and 144 weeks, there was a median gain in limb fat of 0.9 kg in the TDF/FTC arm (n = 48) compared with a median loss of 0.8 kg in the ZDV/3TC arm (n = 38). Mean fasting total cholesterol increased significantly from baseline in the TDF/FTC arm (+24 mg/dL; P < 0.0001) and the ZDV/3TC arm (+36 mg/dL; P < 0.0001), and the increase in the ZDV/3TC arm was significantly greater than that in the TDF/FTC arm (P = 0.005). There was a significant increase in mean fasting low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol in both arms (+13 and +10 mg/dL, respectively, in the TDF/FTC arm; P < 0.0001 for both) and +16 and +12 mg/dL, respectively, in the ZDV/3TC arm, P < 0.0001 for both), and the difference between arms was not statistically significant for LDL or HDL cholesterol. Patients in the TDF/FTC arm had increases in mean fasting triglyceride levels (+4 mg/dL) that were not statistically significant (P = 0.08) compared with baseline, whereas there were significant gains in the ZDV/3TC arm (+36 mg/dL; P < 0.0001) that were significantly greater than those in the TDF/FTC arm (P = 0.047).
After 144 weeks of treatment, significantly more HIV-infected patients receiving a regimen containing TDF/FTC and EFV achieved and maintained virologic suppression to <400 copies/mL (71%) than those starting a regimen of ZDV/3TC and EFV (58%). Cumulative rates of discontinuation for virologic failure and adverse events continued to be higher in the ZDV/3TC-based arm, in which occurrence of M184V was also more frequent. No patient in either arm developed the K65R mutation. Patients in the TDF/FTC arm had smaller increases in fasting total cholesterol and triglycerides than those in the ZDV/3TC arm, and total limb fat was significantly greater in those treated with TDF/FTC compared with those treated with ZDV/3TC.
These cumulative 144-week results extend the previous findings at 48 and 96 weeks of superior virologic suppression with a once-daily TDF/FTC-based regimen compared with a twice-daily ZDV/3TC-based regimen.4,5 They are also highly consistent with the 144-week results of Gilead study 903, a double-blind study comparing TDF with stavudine (d4T), in which 71% of patients receiving TDF, 3TC, and EFV had virologic suppression to <400 copies/mL (missing and switch = failure) and 68% were suppressed to <50 copies/mL (compared with 64% in the current study).6 Although reported response rates for ZDV/3TC at 144 weeks were higher in the AIDS Clinical Trials Group (ACTG) 5095 trial (using a pure intent-to-treat [ITT] method of analysis), the ITT missing data or off treatment = failure analysis of that trial was the most comparable method to the TLOVR analysis used in this study and yielded a similar outcome of approximately 60% suppressed to <50 copies/mL at 144 weeks (compared with 58% in the current study).
Overall renal safety was comparable in the 2 arms, with similar small numbers of patients experiencing graded increases in creatinine and proteinuria in both groups over 144 weeks. The clinical significance of the small declines in GFR estimated by the CG equation in both groups and a statistically significantly greater decline in GFR estimated by the MDRD equation in the TDF/FTC arm are uncertain; however, it is reassuring that there were no discontinuations for renal events in either group-a finding also seen in 6 years of follow-up in TDF-treated patients in study 903.7 Although a study from the Johns Hopkins HIV Clinical Cohort also found declines in GFR among antiretroviral-naive patients after 96 weeks of treatment, these investigators did not find significant differences between the TDF-based and non-TDF-based nucleoside reverse transcriptase inhibitor (NRTI) regimens (−15 and −14 mL/min per 1.73 m2).8 Numerous case reports, cohort studies, and clinical trials have assessed possible TDF-associated renal toxicity; on the whole, severe tubular dysfunction seems to be rare and decreases in GFR have not been clinically significant, particularly in treatment-naive patients.9
Overall, the metabolic benefits of thymidine analogue-sparing regimens are becoming clearer with results of longer term follow-up of studies such as those showing that TDF-based regimens have shown superior lipid profiles with a lower risk of lipoatrophy in comparison to ZDV- and d4T-based regimens.5,6,10 In the present study, in which TDF was compared with ZDV, and in study 903, in which TDF was compared with d4T, there were significant decreases in limb fat in the thymidine analogue-containing arms compared with increases in limb fat in the TDF arms, an important difference because of the lack of nonsurgical options for patients who experience this adverse event. Further confirmation of TDF's metabolic safety comes from preliminary data presented for the ACTG 5142 study, in which TDF-containing arms had low levels of lipoatrophy similar to those in the nucleoside-sparing arm and a significantly lower level of lipoatrophy than in the ZDV arm.10
The efficacy analyses beyond 48 weeks in this study are limited by patient dropouts among those who chose not to consent to study extensions at 48 and 96 weeks in that these patients did not contribute to the numerator or the denominator of the ITT TLOVR analysis presented in Figure 1. The number of patients who dropped out at each extension was relatively small and was similar between the 2 arms, however. Also, as noted previously, interpretation of the limb fat results of this study is limited by the lack of baseline DEXA scans and by the relatively small numbers of patients with DEXA scans at 48 weeks.5 It is notable that among the larger numbers of patients with scans at 96 and 144 weeks, the differences between the arms were maintained and the gap between them in kilograms of limb fat widened slightly at 144 weeks, findings unlikely to have been present at randomization.
Long-term follow-up of antiretroviral-naive patients starting a TDF/FTC- or ZDV/3TC-based regimen confirms the superior ability of the TDF/FTC-based regimen to suppress viral load without causing the lipoatrophy seen with ZDV/3TC-containing regimens and with less effect on lipids. These benefits may be further enhanced by the recent availability of the once-daily single-pill coformulation of TDF/FTC/EFV.
1. Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med
2. US Department of Health and Human Services (DHHS) Panel on Antiretroviral Guidelines for Adults and Adolescents Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. October 10, 2006. Available at: http://www.aidsinfo.nih.gov
. Accessed February 22, 2007.
3. May MT, Sterne JA, Costagliola D, et al. Antiretroviral Therapy (ART) Cohort Collaboration. HIV treatment response and prognosis in Europe and North America in the first decade of highly active antiretroviral therapy: a collaborative analysis. Lancet
4. Pozniak AL, Gallant JE, DeJesus E, et al. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz versus fixed-dose zidovudine/lamivudine and efavirenz in antiretroviral-naive patients: virologic, immunologic, and morphologic changes-a 96-week analysis. J Acquir Immune Defic Syndr
5. Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs. zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med
6. 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
7. Cassetti I, Madruga JVR, Suleiman JMAH, et al. The safety and efficacy of tenofovir DF in combination with lamivudine and efavirenz through 6 years in antiretroviral-naïve HIV-1-infected patients. HIV Clin Trials
8. Moore RD, Keruly JC, Gallant JE. Tenofovir and renal dysfunction in clinical practice [abstract 832]. Presented at: 14th Conference on Retroviruses and Opportunistic Infections; 2007; Los Angeles.
9. Roling J, Schmid H, Fischereder M, et al. HIV-associated renal diseases and highly active antiretroviral therapy-induced nephropathy. Clin Infect Dis
10. Haubrich R, Riddler S, DiRienzo G, et al. Metabolic outcomes of ACTG 5142: a prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection [abstract 38]. Presented at: 14th Conference on Retroviruses and Opportunistic Infections; 2007; Los Angeles.