Abstract

Background

Efavirenz and lopinavir/ritonavir are both recommended antiretroviral agents for combination first-line therapy, although information on direct comparisons between them is scarce. A retrospective longitudinal study from the VACH cohort comparing both regimens was performed.

Methods

Efficacy was examined comparing time to virological failure, CD4 recovery and clinical progression. Tolerability was examined comparing time to treatment discontinuation for any reason and for toxicity. Survival analysis was conducted using the Kaplan–Meier method, and standard and weighted Cox regression models.

Results

A total of 1550 antiretroviral-naive patients starting a two-nucleoside reverse transcriptase inhibitor regimen plus either efavirenz (n = 1159) or lopinavir/ritonavir (n = 391) were included in the study. At baseline, patients starting lopinavir/ritonavir had higher HIV-1 RNA and lower CD4+ cell counts. There was no difference in the adjusted hazards of virological failure [efavirenz versus lopinavir/ritonavir hazard ratio (HR) = 0.93, 95% confidence interval (CI): 0.77–1.12, P = 0.43], CD4 recovery (HR = 1.11, 95% CI: 0.95–1.30, P = 0.19) and clinical progression (HR = 0.71, 95% CI: 0.39–1.31, P = 0.27). There was an increased risk of discontinuation for any reason or for toxicity for lopinavir/ritonavir (HR = 2.10, 95% CI: 1.40–3.15, P = 0.0003). CD4 recovery with both drugs was also similar in the lowest CD4 strata. A higher risk of early hypertriglyceridaemia was associated with lopinavir/ritonavir-based regimens.

Conclusions

Our study suggests similar virological efficacy for efavirenz- or lopinavir/ritonavir-based first-line antiretroviral regimens, but an increased risk of discontinuation because of toxicity in case of lopinavir/ritonavir-based therapy. Immunological outcome appeared similar with both regimens.

Introduction

The primary goal of antiretroviral therapy in treatment-naive patients beginning therapy is the maintenance of an undetectable viral load using an ultrasensitive assay as long as possible, as anything less is associated with decreased durability of the regimen and the development of drug resistance.1 Based on its demonstrated efficacy and tolerability, the preferred regimens to start therapy within antiretroviral-naive patients are those based on combinations of two nucleoside reverse transcriptase inhibitors (NRTIs) and either a boosted protease inhibitor or a non-nucleoside reverse transcriptase inhibitor (NNRTI).2–5 A number of antiretroviral treatment guidelines recommend lopinavir/ritonavir or efavirenz as the preferred drugs for designing a protease inhibitor-based or NNRTI-based highly active antiretroviral therapy regimen, respectively.1,6 Recently, the results of a randomized trial comparing the efficacy and tolerability of efavirenz and lopinavir/ritonavir in treatment-naive patients have been made available, demonstrating a superior virological efficacy for efavirenz, albeit immunological recovery was superior in the lopinavir/ritonavir-treated patients.7 However, recent studies in ‘real life’ have shown controversial results; efavirenz and lopinavir/ritonavir-based first-line antiretroviral regimens performing equally with respect to virological efficacy, immunological recovery and tolerability8,9 or lopinavir/ritonavir-based regimens being superior to efavirenz-based ones.10

The aim of the SUSKA study was to compare the long-term efficacy and tolerability of potent antiretroviral combinations based on two NRTIs plus either efavirenz or lopinavir/ritonavir in antiretroviral-naive HIV-1-infected subjects in a real-life setting such as that represented by patients enrolled in the Spanish VACH (VIH-Aplicación de Control Hospitalario) Cohort.

Patients and methods

The Spanish VACH Cohort

The members of the Spanish VACH Cohort are followed in 19 tertiary-level hospitals, geographically representative of the different regions of Spain. All participating institutions belong to the National Health Service, which provides medical care to all the country’s population, including free antiretroviral medication when needed. The VACH Cohort now accounts for 10 776 patients who represent 14% of the Spanish HIV-1-infected population.

Data collection

In the VACH Cohort, data are prospectively collected according to standardized criteria and are electronically stored in the Aplicación de Control Hospitalario (AC&H™), an application specifically developed for the management of the cohort data. All study-related data and variables were collected during routine, scheduled, clinical appointments and via a review of the subjects’ electronic medical records. An electronic case record form was used to collect the information, following the standard operation procedures of the VACH group. The specific data of the study were sent via e-mail to the coordinator and the data management centre and were imported into the central database.

On enrolment, standardized data collection electronic forms were completed at the sites providing information, which for this study included: gender, age at enrolment, risk factors for the transmission of HIV-1 infection Centers of Disease Control and Prevention (CDC) stage, the date of HIV-1 treatment initiation, the specific antiretroviral regimen used, the dates of changes of that regimen and the reasons for those changes, CD4+ T cell counts, plasma HIV-RNA levels, alanine aminotransferase (ALT), total and low-density lipoprotein (LDL) cholesterol and triglyceride levels at initiation of antiretroviral therapy and every 3 months thereafter. Further, all cumulative data characterizing the patient’s underlying HIV-1 infection since inclusion in the VACH Cohort were collected, including information on demography, antiretroviral therapy, CD4 cell counts and HIV-1 viral loads. Dates of diagnosis of all AIDS-defining diseases were recorded using the 1993 clinical definition of AIDS from the CDC.11 All collected information was transformed into a standardized format and merged into a central data set. The data passed an internal duplicate control at the central data centre that identifies patients by a unique code. Internal validation controls and quality controls of the data were used.

Study population

To be included in the SUSKA study, patients belonging to the VACH Cohort had to meet the following criteria: (i) having started a potent antiretroviral regimen based on the combination of two NRTIs and either efavirenz or lopinavir/ritonavir, without previous exposure to any treatment for HIV-1 infection; (ii) having at least one measurement of HIV-1 RNA after initiation; and (iii) having no treatment with immunomodulating agents before baseline or during the follow-up. The study was approved by all the Ethics Committees of the participating hospitals. Written informed consent was obtained from all the participants in the study.

Methods

Plasma HIV-1 RNA was measured using quantitative RT–PCR (Amplicor 1.5; Roche Molecular Systems, Branchburg, NJ, USA). The lower limit of detection of these assays is 50 copies/mL. For values below the limit of detection, a value of 49 copies/mL was imputed. Peripheral blood CD4+ T cell counts were measured using standard flow cytometry techniques at baseline and every 3 months.

Outcomes definition

Patients were compared with respect to several outcomes using ‘time-to-event’ type analyses, which we defined as follows: dates of the first of two consecutive viral load measurements >50 copies/mL after at least 6 months of therapy and of first CD4+ T cell count >100 cells/mm3 above baseline count were used to define virological failure and immune reconstitution events, whereas the dates of new or recurring AIDS-defining conditions11 or death from any cause were used to analyse clinical progression events. Toxicity events were defined as follows: discontinuation of efavirenz and lopinavir/ritonavir because of any toxicity; reaching grade III ALT elevation (i.e. 205 IU/mL in those with baseline levels ≤40 IU/mL or >3.5× baseline levels in those with pre-therapy levels >40 IU/mL; and experiencing ≥1 step up in the NCEP Adult Treatment Panel III category definitions for total fasting cholesterol, LDL cholesterol and triglycerides.12 For patients who did not experience such events, follow-up was right-truncated at last visit.

Statistical analysis

The characteristics of patients who started efavirenz or lopinavir/ritonavir were compared using the two-sample Wilcoxon rank test for continuous measurements and the χ2 test for categorical variables. ANCOVA adjusted for baseline values was used to assess changes from baseline of CD4, viral load and plasma lipids.

The Kaplan–Meier method was used to describe time-to-event variables, and comparisons between the starting efavirenz or lopinavir/ritonavir were made by means of the log-rank test in the univariable analyses; adjustment for confounders was conducted by means of the Cox proportional hazards model. We used logistic regression models with the baseline CD4 count as covariate to compare CD4 gain at 6, 12 and 18 months. The analysis was performed using SAS version 9.1.3 software (SAS Institute Inc., Cary, NC, USA) and the significance was established at the 0.05 level (two-sided).

Results

Baseline patient characteristics

The study included 1550 patients who met the eligibility criteria. The number of subjects available for the analysis and the reasons for censoring are shown in Table 1. Baseline characteristics of the study population according to the third drug used in addition to the NRTI backbone are shown in Table 2. Efavirenz was given to 74.8% of patients, whereas 25.2% received lopinavir/ritonavir as the third drug. Globally, the most frequently employed NRTI backbones were zidovudine/lamivudine (n = 550, 35.5%), followed by didanosine/lamivudine (n = 283, 18.3%), tenofovir/other nucleoside analogues (n = 244, 15.7%), stavudine/lamivudine (n = 215, 13.9%) and stavudine/didanosine (n = 141, 9.1%). There were statistically significant differences between both lopinavir/ritonavir- and efavirenz-based regimens with respect to nucleoside analogues backbone (P < 0.0001). Patients receiving a lopinavir/ritonavir-based first-line regimen had, on average, lower baseline CD4+ T cell counts (P < 0.0001) and a higher viral load (P = 0.0043) compared with those receiving efavirenz-based regimens. Also, the lopinavir/ritonavir-based regimens were initiated at later calendar years compared with the efavirenz-based regimens (P < 0.0001) (Table 2).

Table 1

Flow of study participants, number of patients available and reasons for censoring per time-point

Follow-up (months) EFV
(n = 1159) 
LPV/r
(n = 391) 
Total
(n = 1550) 
0 to 3    
 available for analysis 1159 391 1550 
 reason for censoring 113 54 167 
  change of treatment 
  lost to follow-up 
  on follow-up 43 20 63 
  other reasons 10 
  safety/tolerability 24 13 37 
  treatment failure 30 14 44 
>3 to 6    
 available for analysis 1046 337 1383 
 reason for censoring 167 57 224 
  change of treatment 13 22 
  lost to follow-up 
  on follow-up 59 18 77 
  other reasons 13 14 
  safety/tolerability 35 15 50 
  treatment failure 41 14 55 
>6 to 12    
 available for analysis 879 280 1159 
 reason for censoring 260 77 337 
  change of treatment 20 15 35 
  exitus 
  lost to follow-up 12 
  on follow-up 102 28 130 
  other reasons 20 25 
  safety/tolerability 55 12 67 
  treatment failure 56 11 67 
>12 to 18    
 available for analysis 619 203 822 
 reason for censoring 156 52 208 
  change of treatment 11 15 26 
  lost to follow-up 
  on follow-up 56 23 79 
  other reasons 10 11 
  safety/tolerability 50 59 
  treatment failure 27 30 
>18    
 available for analysis 463 151 614 
Follow-up (months) EFV
(n = 1159) 
LPV/r
(n = 391) 
Total
(n = 1550) 
0 to 3    
 available for analysis 1159 391 1550 
 reason for censoring 113 54 167 
  change of treatment 
  lost to follow-up 
  on follow-up 43 20 63 
  other reasons 10 
  safety/tolerability 24 13 37 
  treatment failure 30 14 44 
>3 to 6    
 available for analysis 1046 337 1383 
 reason for censoring 167 57 224 
  change of treatment 13 22 
  lost to follow-up 
  on follow-up 59 18 77 
  other reasons 13 14 
  safety/tolerability 35 15 50 
  treatment failure 41 14 55 
>6 to 12    
 available for analysis 879 280 1159 
 reason for censoring 260 77 337 
  change of treatment 20 15 35 
  exitus 
  lost to follow-up 12 
  on follow-up 102 28 130 
  other reasons 20 25 
  safety/tolerability 55 12 67 
  treatment failure 56 11 67 
>12 to 18    
 available for analysis 619 203 822 
 reason for censoring 156 52 208 
  change of treatment 11 15 26 
  lost to follow-up 
  on follow-up 56 23 79 
  other reasons 10 11 
  safety/tolerability 50 59 
  treatment failure 27 30 
>18    
 available for analysis 463 151 614 

EFV, efavirenz; LPV/r, lopinavir/ritonavir.

Table 2

Baseline characteristics of study population according to the use of efavirenz or lopinavir/ritonavir

Characteristic Treatment P value 
EFV (n = 1159) LPV/r (n = 391) 
Age, years 36 (32, 42) 39 (33, 44) 0.0001 
Male, n (%) 904 (78) 305 (78) 0.9977 
HIV exposure, n (%)   0.4234 
 IVDU 441 (38.2) 147 (37.6)  
 MsM 275 (23.8) 88 (22.5)  
 heterosexuals 388 (33.6) 129 (33)  
 other/unknown 55 (4.7) 27 (6.9)  
CDC C events, n (%) 314 (27.1) 142 (36.3) 0.0005 
NRTI backbone, n (%)   <0.0001 
 ZDV/3TC 412 (35.5) 138 (35.3)  
 d4T/3TC 169 (14.6) 46 (11.8)  
 d4T/ddI 122 (10.5) 19 (4.9)  
 ddI/3TC 219 (18.9) 64 (16.4)  
 ZDV/ddI 24 (2.1) 11 (2.8)  
 TDF/other NRTI 155 (13.4) 89 (22.8)  
 ABC/3TC 58 (5) 24 (6.1)  
Year of starting, n (%)   <0.0001 
 1999–2000 229 (19.8) 2 (0.5)  
 2001 191 (16.5) 24 (6.1)  
 2002 152 (13.1) 63 (16.1)  
 2003–04 338 (29.2) 213 (54.5)  
 2005–06 249 (21.5) 89 (22.8)  
HCV infection, n (%) 467 (45.1) 156 (45.5) 0.9075 
HBV infection, n (%) 62 (6.2) 13 (4.0) 0.1296 
HIV-RNA, log10 copies/mL 5.03 (4.69, 5.53) 5.16 (4.72, 5.70) 0.0043 
CD4 cell count, cells/mm3 187 (67, 285) 120 (35, 230) <0.0001 
ALT, IU/mL 34 (20, 61) 36 (23, 60) 0.538 
AST, IU/mL 32 (23, 54) 34 (25, 58.5) 0.1109 
Total cholesterol, mg/dL 154.44 (131, 181) 155.5 (132, 184) 0.6458 
HDL cholesterol, mg/dL 36.5 (29, 45) 35 (28, 44) 0.154 
LDL cholesterol, mg/dL 95 (73, 116) 97 (71, 121) 0.5809 
Triglycerides, mg/dL 112 (84, 155) 128 (89, 178.5) 0.0021 
Characteristic Treatment P value 
EFV (n = 1159) LPV/r (n = 391) 
Age, years 36 (32, 42) 39 (33, 44) 0.0001 
Male, n (%) 904 (78) 305 (78) 0.9977 
HIV exposure, n (%)   0.4234 
 IVDU 441 (38.2) 147 (37.6)  
 MsM 275 (23.8) 88 (22.5)  
 heterosexuals 388 (33.6) 129 (33)  
 other/unknown 55 (4.7) 27 (6.9)  
CDC C events, n (%) 314 (27.1) 142 (36.3) 0.0005 
NRTI backbone, n (%)   <0.0001 
 ZDV/3TC 412 (35.5) 138 (35.3)  
 d4T/3TC 169 (14.6) 46 (11.8)  
 d4T/ddI 122 (10.5) 19 (4.9)  
 ddI/3TC 219 (18.9) 64 (16.4)  
 ZDV/ddI 24 (2.1) 11 (2.8)  
 TDF/other NRTI 155 (13.4) 89 (22.8)  
 ABC/3TC 58 (5) 24 (6.1)  
Year of starting, n (%)   <0.0001 
 1999–2000 229 (19.8) 2 (0.5)  
 2001 191 (16.5) 24 (6.1)  
 2002 152 (13.1) 63 (16.1)  
 2003–04 338 (29.2) 213 (54.5)  
 2005–06 249 (21.5) 89 (22.8)  
HCV infection, n (%) 467 (45.1) 156 (45.5) 0.9075 
HBV infection, n (%) 62 (6.2) 13 (4.0) 0.1296 
HIV-RNA, log10 copies/mL 5.03 (4.69, 5.53) 5.16 (4.72, 5.70) 0.0043 
CD4 cell count, cells/mm3 187 (67, 285) 120 (35, 230) <0.0001 
ALT, IU/mL 34 (20, 61) 36 (23, 60) 0.538 
AST, IU/mL 32 (23, 54) 34 (25, 58.5) 0.1109 
Total cholesterol, mg/dL 154.44 (131, 181) 155.5 (132, 184) 0.6458 
HDL cholesterol, mg/dL 36.5 (29, 45) 35 (28, 44) 0.154 
LDL cholesterol, mg/dL 95 (73, 116) 97 (71, 121) 0.5809 
Triglycerides, mg/dL 112 (84, 155) 128 (89, 178.5) 0.0021 

EFV, efavirenz; LPV/r, lopinavir/ritonavir; CDC, Centers for Disease Control; IVDU, intravenous drug user; MsM, men who have sex with men; NRTI, nucleoside reverse transcriptase inhibitor; d4T, stavudine; 3TC, lamivudine; ddI, didanosine; ZDV, zidovudine; TDF, tenofovir; FTC, emtricitabine; ABC, abacavir; HCV, hepatitis C virus; HBV, hepatitis B virus; ALT, alanine aminotransferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Reported values are medians (interquartile range) unless specified.

Virological failure and predictors

Over a cumulative virological follow-up of 1170 years, 658 (42.5%) patients, 489 (42.1%) in the efavirenz group and 169 (43.2%) in the lopinavir/ritonavir group, experienced a virological failure. The Kaplan–Meier estimate of the time to virological failure according to which third drug was originally started is shown in Figure 1. The risk of virological failure in the group of patients started on a lopinavir/ritonavir-based regimen was 16% higher, although not statistically significant (log-rank, P = 0.09), compared with those started on an efavirenz-based regimen. The percentages of patients with virological failure by the median (mean) follow-up of 71 (81) weeks were 43.2% in the lopinavir/ritonavir group versus 43.2% in the efavirenz group. In the multivariable analysis, the use of efavirenz or lopinavir/ritonavir as the third drug was not associated with differing risks of virological failure [hazard ratio (HR) = 0.93, 95% CI: 0.77–1.12, P = 0.43] (Table 3). Independent predictors of virological failure were: baseline HIV-1 RNA and CD4 count, prior CDC C status, age, year of treatment onset, having acquired HIV-1 infection through intravenous drug use when compared with heterosexual contacts and initial nucleoside analogue backbone. There was evidence of the association of regimens containing stavudine/didanosine, stavudine/lamivudine and tenofovir/other nucleoside analogues with an increased risk of virological failure (Table 3).

Figure 1

Kaplan–Meier curves illustrating the cumulative proportion of patients with virological failure (>50 copies/mL), according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Figure 1

Kaplan–Meier curves illustrating the cumulative proportion of patients with virological failure (>50 copies/mL), according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Table 3

Relative hazards of virological failure (>50 copies/mL) from fitting a traditional Cox proportional hazards model and a weighted Cox regression

Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0904 1.16 0.98 1.39 0.4283 0.93 0.77 1.12 
Baseline HIV-RNA (per 1 log10 higher) 0.0028 1.20 1.06 1.35 0.0005 1.25 1.10 1.41 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.4177 1.02 0.98 1.06 0.0110 1.06 1.01 1.11 
CDC C event 0.1479 1.13 0.96 1.33 0.0349 1.21 1.01 1.45 
Age (per 10 years older) 0.0146 0.90 0.83 0.98 0.0094 0.89 0.81 0.97 
Male (reference = female) 0.8425 0.98 0.81 1.18 0.4219 0.92 0.75 1.13 
Mode of transmission (reference = IVDU) 0.0001    0.0002    
 MsM 0.0332 0.81 0.66 0.98 0.0069 0.68 0.52 0.90 
 heterosexual 0.0002 0.71 0.59 0.85 0.0001 0.62 0.49 0.79 
 other/unknown 0.0014 0.51 0.34 0.77 0.0005 0.46 0.30 0.72 
NRTI backbone (reference = ZDV/3TC) 0.0000    0.0000    
 d4T/3TC 0.0168 1.33 1.05 1.69 0.0014 1.48 1.16 1.88 
 d4T/ddI 0.0016 1.57 1.19 2.08 0.0000 1.85 1.39 2.46 
 ddI/3TC 0.1469 1.20 0.94 1.52 0.3165 0.88 0.68 1.13 
 ZDV/ddI 0.4774 1.20 0.72 2.00 0.4901 1.20 0.72 2.00 
 TDF/other NRTI 0.0000 2.29 1.82 2.87 0.0008 1.53 1.19 1.95 
 ABC/3TC 0.0089 1.58 1.12 2.23 0.0272 1.48 1.05 2.09 
Year of starting (per more recent year) 0.0000 1.23 1.17 1.29 0.0000 1.29 1.22 1.37 
HCV infection 0.0828 1.16 0.98 1.36 0.4647 0.91 0.72 1.16 
HBV infection 0.0016 1.69 1.22 2.34 0.0069 1.58 1.13 2.20 
Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0904 1.16 0.98 1.39 0.4283 0.93 0.77 1.12 
Baseline HIV-RNA (per 1 log10 higher) 0.0028 1.20 1.06 1.35 0.0005 1.25 1.10 1.41 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.4177 1.02 0.98 1.06 0.0110 1.06 1.01 1.11 
CDC C event 0.1479 1.13 0.96 1.33 0.0349 1.21 1.01 1.45 
Age (per 10 years older) 0.0146 0.90 0.83 0.98 0.0094 0.89 0.81 0.97 
Male (reference = female) 0.8425 0.98 0.81 1.18 0.4219 0.92 0.75 1.13 
Mode of transmission (reference = IVDU) 0.0001    0.0002    
 MsM 0.0332 0.81 0.66 0.98 0.0069 0.68 0.52 0.90 
 heterosexual 0.0002 0.71 0.59 0.85 0.0001 0.62 0.49 0.79 
 other/unknown 0.0014 0.51 0.34 0.77 0.0005 0.46 0.30 0.72 
NRTI backbone (reference = ZDV/3TC) 0.0000    0.0000    
 d4T/3TC 0.0168 1.33 1.05 1.69 0.0014 1.48 1.16 1.88 
 d4T/ddI 0.0016 1.57 1.19 2.08 0.0000 1.85 1.39 2.46 
 ddI/3TC 0.1469 1.20 0.94 1.52 0.3165 0.88 0.68 1.13 
 ZDV/ddI 0.4774 1.20 0.72 2.00 0.4901 1.20 0.72 2.00 
 TDF/other NRTI 0.0000 2.29 1.82 2.87 0.0008 1.53 1.19 1.95 
 ABC/3TC 0.0089 1.58 1.12 2.23 0.0272 1.48 1.05 2.09 
Year of starting (per more recent year) 0.0000 1.23 1.17 1.29 0.0000 1.29 1.22 1.37 
HCV infection 0.0828 1.16 0.98 1.36 0.4647 0.91 0.72 1.16 
HBV infection 0.0016 1.69 1.22 2.34 0.0069 1.58 1.13 2.20 

EFV, efavirenz; LPV/r, lopinavir/ritonavir; CDC, Centers for Disease Control; IVDU, intravenous drug user; MsM, men who have sex with men; NRTI, nucleoside reverse transcriptase inhibitor; d4T, stavudine; 3TC, lamivudine; ddI, didanosine; ZDV, zidovudine; TDF, tenofovir; ABC, abacavir; HCV, hepatitis C virus; HBV, hepatitis B virus; HR, hazard ratio; LL, lower limit of 95% CI; UL, upper limit of 95% CI.

aAll variables in the list were fitted for the multivariate analysis.

Clinical outcome and predictors

The Kaplan–Meier estimates of the time to clinical progression over a cumulative follow-up of 1194 years (log-rank P = 0.1966), according to the initial treatment group, are shown in Figure 2. There was not a statistically significantly higher cumulative proportion of patients with clinical progression in the group starting an efavirenz-based regimen (5.5%) by 72 weeks, compared with the group initially receiving a lopinavir/ritonavir-based regimen (3.6%). After adjusting for other baseline characteristics, the results remained unchanged (HR = 0.71, 95% CI: 0.39–1.31, P = 0.2073) (Table 4). Finally, independent predictors of clinical progression were: baseline CD4 count, prior CDC C status, year of treatment onset and hepatitis B virus (HBV) infection at baseline.

Figure 2

Kaplan–Meier curves illustrating the cumulative proportion of patients with clinical progression (new or recurrent AIDS-defining conditions or death), according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Figure 2

Kaplan–Meier curves illustrating the cumulative proportion of patients with clinical progression (new or recurrent AIDS-defining conditions or death), according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Table 4

Relative hazards of clinical progression from fitting a traditional Cox proportional hazards model and a weighted Cox regression

Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.1966 0.69 0.38 1.22 0.2703 0.71 0.39 1.31 
Baseline HIV-RNA (per 1 log10 higher) 0.0123 1.57 1.10 2.23 0.1472 1.31 0.91 1.89 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0003 0.71 0.59 0.85 0.0152 0.78 0.64 0.95 
CDC C event 0.0000 2.68 1.72 4.19 0.0132 1.86 1.14 3.05 
Age (per 10 years older) 0.4849 1.08 0.86 1.36 0.6290 1.07 0.82 1.38 
Male (reference = female) 0.4132 1.27 0.71 2.27 0.6490 1.15 0.62 2.15 
Mode of transmission (reference = IVDU) 0.1303    0.7739    
 MsM 0.1024 0.61 0.34 1.10 0.5450 0.78 0.35 1.75 
 heterosexuals 0.0345 0.56 0.33 0.96 0.3172 0.70 0.34 1.41 
 other/unknown 0.4195 0.65 0.23 1.83 0.4954 0.67 0.22 2.09 
NRTI backbone (reference = ZDV/3TC) 0.5486    0.8521    
 d4T/3TC 0.5382 0.77 0.33 1.78 0.9692 1.02 0.43 2.41 
 d4T/ddI 0.1002 0.53 0.25 1.13 0.8310 1.10 0.46 2.65 
 ddI/3TC 0.1995 0.68 0.37 1.23 0.6666 0.87 0.47 1.61 
 ZDV/ddI 0.6937 0.74 0.17 3.22 0.9155 0.92 0.21 4.05 
 TDF/other NRTI 0.3805 0.70 0.32 1.54 0.3899 1.50 0.60 3.78 
 ABC/3TC 0.0949 0.18 0.02 1.35 0.3384 0.37 0.05 2.84 
Year of starting (per more recent year) 0.0014 0.81 0.72 0.92 0.0080 0.79 0.66 0.94 
HCV infection 0.0470 1.61 1.01 2.57 0.5756 1.22 0.61 2.43 
HBV infection 0.0004 3.39 1.73 6.63 0.0020 2.99 1.49 6.01 
Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.1966 0.69 0.38 1.22 0.2703 0.71 0.39 1.31 
Baseline HIV-RNA (per 1 log10 higher) 0.0123 1.57 1.10 2.23 0.1472 1.31 0.91 1.89 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0003 0.71 0.59 0.85 0.0152 0.78 0.64 0.95 
CDC C event 0.0000 2.68 1.72 4.19 0.0132 1.86 1.14 3.05 
Age (per 10 years older) 0.4849 1.08 0.86 1.36 0.6290 1.07 0.82 1.38 
Male (reference = female) 0.4132 1.27 0.71 2.27 0.6490 1.15 0.62 2.15 
Mode of transmission (reference = IVDU) 0.1303    0.7739    
 MsM 0.1024 0.61 0.34 1.10 0.5450 0.78 0.35 1.75 
 heterosexuals 0.0345 0.56 0.33 0.96 0.3172 0.70 0.34 1.41 
 other/unknown 0.4195 0.65 0.23 1.83 0.4954 0.67 0.22 2.09 
NRTI backbone (reference = ZDV/3TC) 0.5486    0.8521    
 d4T/3TC 0.5382 0.77 0.33 1.78 0.9692 1.02 0.43 2.41 
 d4T/ddI 0.1002 0.53 0.25 1.13 0.8310 1.10 0.46 2.65 
 ddI/3TC 0.1995 0.68 0.37 1.23 0.6666 0.87 0.47 1.61 
 ZDV/ddI 0.6937 0.74 0.17 3.22 0.9155 0.92 0.21 4.05 
 TDF/other NRTI 0.3805 0.70 0.32 1.54 0.3899 1.50 0.60 3.78 
 ABC/3TC 0.0949 0.18 0.02 1.35 0.3384 0.37 0.05 2.84 
Year of starting (per more recent year) 0.0014 0.81 0.72 0.92 0.0080 0.79 0.66 0.94 
HCV infection 0.0470 1.61 1.01 2.57 0.5756 1.22 0.61 2.43 
HBV infection 0.0004 3.39 1.73 6.63 0.0020 2.99 1.49 6.01 

EFV, efavirenz; LPV/r, lopinavir/ritonavir; CDC, Centers for Disease Control; IVDU, intravenous drug user; MsM, men who have sex with men; NRTI, nucleoside reverse transcriptase inhibitor; d4T, stavudine; 3TC, lamivudine; ddI, didanosine; ZDV, zidovudine; TDF, tenofovir; ABC, abacavir; HCV, hepatitis C virus; HBV, hepatitis B virus; HR, hazard ratio; LL, lower limit of 95% CI; UL, upper limit of 95% CI.

aAll variables in the list were fitted for the multivariate analysis.

Immunological outcome and predictors

Regarding immunological response, CD4+ T cell recovery was achieved earlier in patients in whom lopinavir/ritonavir was prescribed: the percentage of patients who gained at least 100 cells/mm3 in the efavirenz- and lopinavir/ritonavir-treated groups was 61.4% and 71.2% at 6 months (P = 0.0053), 73.6% and 82.8% at 12 months (P = 0.0128) and 81.6% and 87.0% at 18 months (P = 0.1502), respectively. Although the trend was similar in the adjusted mean change analysis, no statistically significant differences were observed between both groups (Figure 3). CD4 gain led to the same conclusions when adjusting by baseline CD4 count using a logistic regression model (data not shown). Independent predictors of CD4+ cell recovery were: baseline HIV-1 RNA, having acquired HIV-1 infection through homosexual contact when compared with heterosexual contact, year of treatment onset and hepatitis C virus infection (Table 5).

Figure 3

Kaplan–Meier curves illustrating the cumulative proportion of patients with CD4 > 100 cells/mm3, according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Figure 3

Kaplan–Meier curves illustrating the cumulative proportion of patients with CD4 > 100 cells/mm3, according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Table 5

Relative hazards of CD4 recovery (at least a gain of 100 cells/mm3 from baseline) from fitting a Cox proportional hazards model

Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0078 1.23 1.06 1.43 0.1936 1.11 0.95 1.30 
Baseline HIV-RNA (per 1 log10 higher) 0.0000 1.28 1.15 1.42 0.0000 1.25 1.13 1.40 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0105 0.95 0.91 0.99 0.2137 0.97 0.93 1.02 
CDC C event 0.5659 1.04 0.90 1.20 0.7223 0.97 0.83 1.14 
Age (per 10 years older) 0.7125 0.99 0.92 1.06 0.2400 0.96 0.89 1.03 
Male (reference = female) 0.4237 0.94 0.80 1.10 0.3284 0.92 0.77 1.09 
Mode of transmission (reference = IVDU) 0.0000    0.0553    
 MsM 0.0000 1.69 1.41 2.01 0.0440 1.31 1.01 1.69 
 heterosexuals 0.0010 1.32 1.12 1.55 0.7013 1.05 0.83 1.31 
 other/unknown 0.2324 1.19 0.89 1.59 0.7120 0.94 0.68 1.30 
NRTI backbone (reference = ZDV/3TC) 0.0003    0.2408    
 d4T/3TC 0.9915 1.00 0.82 1.23 0.4782 1.08 0.88 1.33 
 d4T/ddI 0.2932 1.15 0.89 1.49 0.1034 1.25 0.96 1.62 
 ddI/3TC 0.0001 1.43 1.19 1.72 0.0271 1.25 1.03 1.53 
 ZDV/ddI 0.2733 0.76 0.47 1.24 0.5449 0.86 0.53 1.40 
 TDF/other NRTI 0.0011 1.45 1.16 1.82 0.0955 1.23 0.96 1.56 
 ABC/3TC 0.1876 1.23 0.90 1.67 0.4196 1.14 0.83 1.55 
Year of starting (per more recent year) 0.0000 1.13 1.09 1.18 0.0013 1.09 1.03 1.14 
HCV infection 0.0000 0.65 0.56 0.75 0.0134 0.75 0.60 0.94 
HBV infection 0.2555 0.80 0.54 1.18 0.2705 0.80 0.54 1.19 
Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0078 1.23 1.06 1.43 0.1936 1.11 0.95 1.30 
Baseline HIV-RNA (per 1 log10 higher) 0.0000 1.28 1.15 1.42 0.0000 1.25 1.13 1.40 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0105 0.95 0.91 0.99 0.2137 0.97 0.93 1.02 
CDC C event 0.5659 1.04 0.90 1.20 0.7223 0.97 0.83 1.14 
Age (per 10 years older) 0.7125 0.99 0.92 1.06 0.2400 0.96 0.89 1.03 
Male (reference = female) 0.4237 0.94 0.80 1.10 0.3284 0.92 0.77 1.09 
Mode of transmission (reference = IVDU) 0.0000    0.0553    
 MsM 0.0000 1.69 1.41 2.01 0.0440 1.31 1.01 1.69 
 heterosexuals 0.0010 1.32 1.12 1.55 0.7013 1.05 0.83 1.31 
 other/unknown 0.2324 1.19 0.89 1.59 0.7120 0.94 0.68 1.30 
NRTI backbone (reference = ZDV/3TC) 0.0003    0.2408    
 d4T/3TC 0.9915 1.00 0.82 1.23 0.4782 1.08 0.88 1.33 
 d4T/ddI 0.2932 1.15 0.89 1.49 0.1034 1.25 0.96 1.62 
 ddI/3TC 0.0001 1.43 1.19 1.72 0.0271 1.25 1.03 1.53 
 ZDV/ddI 0.2733 0.76 0.47 1.24 0.5449 0.86 0.53 1.40 
 TDF/other NRTI 0.0011 1.45 1.16 1.82 0.0955 1.23 0.96 1.56 
 ABC/3TC 0.1876 1.23 0.90 1.67 0.4196 1.14 0.83 1.55 
Year of starting (per more recent year) 0.0000 1.13 1.09 1.18 0.0013 1.09 1.03 1.14 
HCV infection 0.0000 0.65 0.56 0.75 0.0134 0.75 0.60 0.94 
HBV infection 0.2555 0.80 0.54 1.18 0.2705 0.80 0.54 1.19 

EFV, efavirenz; LPV/r, lopinavir/ritonavir; CDC, Centers for Disease Control; IVDU, intravenous drug user; MsM, men who have sex with men; NRTI, nucleoside reverse transcriptase inhibitor; d4T, stavudine; 3TC, lamivudine; ddI, didanosine; ZDV, zidovudine; TDF, tenofovir; ABC, abacavir; HCV, hepatitis C virus; HBV, hepatitis B virus; HR, hazard ratio; LL, lower limit of 95% CI; UL, upper limit of 95% CI.

aAll variables in the list were fitted for the multivariate analysis.

Discontinuation of lopinavir/ritonavir- or efavirenz-based initial therapy due to toxicity

A total of 119 patients interrupted efavirenz or lopinavir/ritonavir due to drug-related toxicity over a cumulated time period of 1202 years. The risk of treatment interruption was 98% greater in lopinavir/ritonavir-based regimens (log-rank P = 0.0003, Figure 4). The adjusted HR (95% CI) comparing the rate of discontinuation of lopinavir/ritonavir with that of efavirenz remained increased and statistically significant 2.10 (1.40–3.15), suggesting that a difference in the rate of discontinuation between these two drugs seems to exist. Independent predictors of treatment interruption secondary to drug-related toxicity were: baseline CD4 counts, gender and HBV infection (Table 6).

Figure 4

Kaplan–Meier curves illustrating the cumulative proportion of patients discontinuing treatment for toxicity, according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Figure 4

Kaplan–Meier curves illustrating the cumulative proportion of patients discontinuing treatment for toxicity, according to treatment with efavirenz (EFV; continuous line) or lopinavir/ritonavir (LPV/r; broken line).

Table 6

Relative hazards of discontinuing efavirenz or lopinavir/ritonavir for toxicity from fitting a Cox proportional hazards model

Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0003 1.98 1.35 2.89 0.0003 2.10 1.40 3.15 
Baseline HIV-RNA (per 1 log10 higher) 0.7803 0.96 0.73 1.26 0.7096 1.05 0.80 1.39 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0703 1.08 0.99 1.18 0.0210 1.12 1.02 1.24 
CDC C event 0.1691 0.75 0.49 1.13 0.2795 0.78 0.50 1.22 
Age (per 10 years older) 0.7340 0.97 0.80 1.17 0.7877 1.03 0.84 1.26 
Male (reference = female) 0.0957 0.71 0.47 1.06 0.0121 0.55 0.35 0.88 
Mode of transmission (reference = IVDU) 0.15290    0.1191    
 MsM 0.29898 0.783 0.493 1.243 0.6228 0.85 0.44 1.63 
 heterosexuals 0.02299 0.599 0.385 0.932 0.0260 0.51 0.29 0.92 
 other/unknown 0.49516 0.760 0.346 1.671 0.3459 0.65 0.27 1.58 
NRTI backbone (reference = ZDV/3TC) 0.2249    0.1651    
 d4T/3TC 0.2673 1.35 0.79 2.29 0.2046 1.42 0.83 2.44 
 d4T/ddI 0.0099 2.12 1.20 3.76 0.0074 2.24 1.24 4.05 
 ddI/3TC 0.0846 1.57 0.94 2.63 0.2017 1.43 0.82 2.49 
 ZDV/ddI 0.9600 NE NE NE 0.9610 NE NE NE 
 TDF/other NRTI 0.3263 1.37 0.73 2.60 0.9785 0.99 0.51 1.94 
 ABC/3TC 0.7746 0.86 0.31 2.41 0.7013 0.82 0.29 2.30 
Year of starting (per more recent year) 0.0945 1.10 0.98 1.24 0.1501 1.11 0.96 1.29 
HCV infection 0.1943 1.29 0.88 1.88 0.9405 0.98 0.56 1.71 
HBV infection 0.1767 1.70 0.79 3.67 0.0855 1.99 0.91 4.34 
Factor Univariate analysis Multivariate analysisa 
P value HR LL UL P value HR LL UL 
EFV versus LPV/r 0.0003 1.98 1.35 2.89 0.0003 2.10 1.40 3.15 
Baseline HIV-RNA (per 1 log10 higher) 0.7803 0.96 0.73 1.26 0.7096 1.05 0.80 1.39 
Baseline CD4 cell count (per 100 cells/mm3 higher) 0.0703 1.08 0.99 1.18 0.0210 1.12 1.02 1.24 
CDC C event 0.1691 0.75 0.49 1.13 0.2795 0.78 0.50 1.22 
Age (per 10 years older) 0.7340 0.97 0.80 1.17 0.7877 1.03 0.84 1.26 
Male (reference = female) 0.0957 0.71 0.47 1.06 0.0121 0.55 0.35 0.88 
Mode of transmission (reference = IVDU) 0.15290    0.1191    
 MsM 0.29898 0.783 0.493 1.243 0.6228 0.85 0.44 1.63 
 heterosexuals 0.02299 0.599 0.385 0.932 0.0260 0.51 0.29 0.92 
 other/unknown 0.49516 0.760 0.346 1.671 0.3459 0.65 0.27 1.58 
NRTI backbone (reference = ZDV/3TC) 0.2249    0.1651    
 d4T/3TC 0.2673 1.35 0.79 2.29 0.2046 1.42 0.83 2.44 
 d4T/ddI 0.0099 2.12 1.20 3.76 0.0074 2.24 1.24 4.05 
 ddI/3TC 0.0846 1.57 0.94 2.63 0.2017 1.43 0.82 2.49 
 ZDV/ddI 0.9600 NE NE NE 0.9610 NE NE NE 
 TDF/other NRTI 0.3263 1.37 0.73 2.60 0.9785 0.99 0.51 1.94 
 ABC/3TC 0.7746 0.86 0.31 2.41 0.7013 0.82 0.29 2.30 
Year of starting (per more recent year) 0.0945 1.10 0.98 1.24 0.1501 1.11 0.96 1.29 
HCV infection 0.1943 1.29 0.88 1.88 0.9405 0.98 0.56 1.71 
HBV infection 0.1767 1.70 0.79 3.67 0.0855 1.99 0.91 4.34 

EFV, efavirenz; LPV/r, lopinavir/ritonavir; CDC, Centers for Disease Control; IVDU, intravenous drug user; MsM, men who have sex with men; NRTI, nucleoside reverse transcriptase inhibitor; d4T, stavudine; 3TC, lamivudine; ddI, didanosine; ZDV, zidovudine; TDF, tenofovir; ABC, abacavir; HCV, hepatitis C virus; HBV, hepatitis B virus; HR, hazard ratio; LL, lower limit of 95% CI; UL, upper limit of 95% CI; NE, not estimated.

aAll variables in the list were fitted for the multivariate analysis.

Total fasting cholesterol, LDL cholesterol and high-density lipoprotein cholesterol increased in both efavirenz- and lopinavir/ritonavir-treated patients without statistically significant differences at any time-point (Figure 5). Although triglycerides increased in both groups, the increase was more marked in lopinavir/ritonavir-treated patients reaching statistical significance at month 6. However, differences between both treatment groups vanished after month 6 (Figure 5).

Figure 5

Baseline-adjusted mean changes and 95% CI by time in plasma lipids for the efavirenz (EFV; continuous lines) and lopinavir/ritonavir (LPV/r; broken lines), according to the ANCOVA analysis with the LOCF approach using the baseline measurements as a covariate. No statistically significant between-treatment differences were found except for triglycerides at month 6 (P < 0.001).

Figure 5

Baseline-adjusted mean changes and 95% CI by time in plasma lipids for the efavirenz (EFV; continuous lines) and lopinavir/ritonavir (LPV/r; broken lines), according to the ANCOVA analysis with the LOCF approach using the baseline measurements as a covariate. No statistically significant between-treatment differences were found except for triglycerides at month 6 (P < 0.001).

Discussion

Lopinavir/ritonavir- and efavirenz-based regimens are currently recommended as first choice initial antiretroviral regimens, although until recently comparative data arising from a randomized trial comparing both types of regimens was lacking.7 Efavirenz-based regimens have usually outperformed boosted protease inhibitor regimens such as ritonavir-boosted soft-gel saquinavir in the FOCUS trial13 and ritonavir-boosted amprenavir,14 resulting in better virological control and less toxicity. The only protease inhibitor that has shown a comparable efficacy rate and toxicity profile at 48 weeks was non-boosted atazanavir in BMS 034 study, although the results of this study are difficult to interpret because of technical problems.15

On the other hand, lopinavir/ritonavir has had until recently the undisputed leadership among protease inhibitors, despite that ritonavir-boosted fosamprenavir has recently showed non-inferior efficacy and tolerability profiles.16 Studies comparing efavirenz- and lopinavir/ritonavir-based regimens include four observational studies and the randomized trial ACTG 5142.7–10,17 The latter has shown that efavirenz-based regimens were superior to lopinavir/ritonavir-based ones in terms of virological efficacy, although the CD4+ cell increase was significantly greater in lopinavir/ritonavir-based regimens.7 The four smaller observational studies reported so far have given conflicting results. Two of them8,17 showed similar virological efficacy between efavirenz- and lopinavir/ritonavir-based regimens, whereas one9 showed better virological response with efavirenz and the other found a superior virological outcome in lopinavir/ritonavir-treated patients.10 All these observational studies, except two,8,10 found a better CD4+ T cell response with lopinavir/ritonavir. Results from our study show similar virological efficacy and long-term CD4+ T cell responses between both regimens.

Although data required to set up recommendations are derived from randomized trials, observational studies more closely reflect the effect of a specific regimen in clinical practice, given that study patients are not selected as in randomized trials and treatment is prescribed with a specific clinical indication. However, this is at the same time their major limitation together with the existence of a confounding-by-indication bias. Therefore, the fact that patients were not randomly assigned to receive either lopinavir/ritonavir or efavirenz is the major limitation of our study as well. However, this methodological drawback was partly limited because of the broad use of national and international guidelines that suggested both efavirenz and lopinavir/ritonavir as first-line therapies in the treatment of HIV-1-infected patients.1,18,19 Notwithstanding that, there were some differences between the two groups of patients that may influence the final outcome, and therefore confounding by indication is still of concern. For instance, patients who were prescribed lopinavir/ritonavir were older, had a more advanced disease, with a higher viral load and a lower CD4+ cell count at baseline. Since both baseline HIV-1 RNA and CD4+ cell count were independent predictors for virological failure, these differences may favour a better outcome for efavirenz-treated patients. Furthermore, independent predictors of clinical progression such as baseline low CD4+ cells and prior C events were also significantly more frequent among lopinavir/ritonavir-treated patients, which may penalize the outcome in this group. Despite all these baseline differences, both regimens performed in a similar way both virologically and immunologically. In addition, patients who were prescribed efavirenz received nucleos(t)ide analogues—which are prescribed ‘once daily’—more frequently than those who were prescribed lopinavir/ritonavir, therefore allowing the construction of ‘once daily’ regimens. Such a treatment strategy may have allowed patients to adhere to treatment more easily, therefore compensating (and masking) possible differences in intrinsic potency between these regimens.20

In our study, similar to the results of ACTG 384, the initial use of a nucleoside backbone with stavudine/didanosine, which was prescribed in around 9% of patients in both groups, led to an increased hazard of virological failure when compared with zidovudine/lamivudine.3 However, in the ACTG 384, this difference was only observed when combined with efavirenz but not with nelfinavir.3 We and others have found that patients starting stavudine/didanosine had a poor virological outcome irrespective of whether efavirenz or lopinavir/ritonavir was used in combination with this backbone.8 In addition, stavudine/lamivudine or tenofovir/other nucleoside analogue backbones were also associated with an excess virological failure. Furthermore, the excess toxicity associated with these backbones supports avoiding their use in any first-line combination. Thus, the inclusion of stavudine in the regimen appeared to exert an unfavourable impact on the virological treatment response in our study. The known association between tenofovir use and the lower probability of achieving virological suppression may be explained by higher rates of virological failure in patients on tenofovir/didanosine/efavirenz21 or by the toxic effects of this combination on CD4+ cells.22 Finally, didanosine is the only nucleoside analogue whose consumption is subordinated to diet requirements, thus rendering treatment regimens including this drug more complicated, and this may have decreased adherence, although this variable was not captured in this study.

Except for two studies,8,10 all the observational studies and ACTG 5142 found a better CD4+ T cell response with lopinavir/ritonavir.7,9,10,17 In our study, we did not found a better immunological outcome in the lopinavir/ritonavir-based arm measured as change from baseline in the long run, although lopinavir/ritonavir-treated patients had a better CD4+ cell recovery during the first year of treatment. The better immunological response in protease inhibitor-based regimens compared with NNRTI-based ones has been attributed to the ability of protease inhibitors to inhibit uninfected CD4+ apoptosis, thus rendering more circulating CD4+ cells available in the blood. However, the functional capabilities of these cells have not been ascertained yet.23

Our data do support a difference between the rate of lopinavir/ritonavir and efavirenz discontinuation because of toxicity, favouring efavirenz. We did not find age differences in the rate of regimen discontinuation due to toxicity. However, females showed an increased risk of discontinuation due to antiretroviral toxicity, a finding that has been described previously.24 With respect to metabolic toxicities and similar to findings reported from the ACTG 514225 and two observational studies,8 a higher risk of triglyceride increase was observed in patients taking lopinavir/ritonavir, although among our patients, this higher risk vanished over time. However, it should be noted that patients treated with lopinavir/ritonavir had a higher baseline triglyceride level, compared with those treated with efavirenz.

Efavirenz has the advantage of a lower pill burden and the use of lopinavir/ritonavir may be hampered by triglyceride elevation. However, the undoubted advantages of efavirenz-based regimens should be balanced against its lower genetic barrier compared with lopinavir/ritonavir as the ACTG 5142 has exemplified.7 Observations from clinical trials and cohort studies show very little or an absence of protease inhibitor resistance and limited nucleoside analogue resistance in viral isolates from patients failing a lopinavir/ritonavir-built first-line regimen, whereas failures from efavirenz-based first-line regimens are associated with the selection of viral mutants showing higher rates of cross-resistance. Although this study showed no difference in the outcome of surrogate and clinical markers between the initial treatment strategies, a higher initial rate of drug resistance selection due to the use of drugs with lower genetic barrier could potentially lead to a more rapid exhaustion of drug options over a longer period of time.

In summary, analysis of virological and immunological efficacy of efavirenz versus lopinavir/ritonavir in antiretroviral-naive patients in this large non-randomized study did not show relevant differences between both drugs. However, a higher risk of discontinuing therapy as well as a higher risk of early hypertriglyceridaemia was associated with the use of lopinavir/ritonavir-based regimens. Therefore, effective first-line antiretroviral regimens may be built based on both drugs, and individual treatment choices must rely on other factors such as simplicity, toxicity profile, co-morbidities and patient’s preferences and lifestyle.

Funding

No specific funding was received for this work.

Transparency declarations

None to declare.

References

1
Hammer
SM
Saag
MS
Schechter
M
, et al.  . 
Treatment for adult HIV infection: 2006 recommendations of the International AIDS Society-USA panel
Top HIV Med
 , 
2006
, vol. 
14
 (pg. 
827
-
43
)
2
Walmsley
S
Bernstein
B
King
M
, et al.  . 
Lopinavir-ritonavir versus nelfinavir for the initial treatment of HIV infection
N Engl J Med
 , 
2002
, vol. 
346
 (pg. 
2039
-
46
)
3
Robbins
GK
De Gruttola
V
Shafer
RW
, et al.  . 
Comparison of sequential three-drug regimens as initial therapy for HIV-1 infection
N Engl J Med
 , 
2003
, vol. 
349
 (pg. 
2293
-
303
)
4
Gulick
RM
Ribaudo
HJ
Shikuma
CM
, et al.  . 
Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV-1 infection
N Engl J Med
 , 
2004
, vol. 
350
 (pg. 
1850
-
61
)
5
van Leth
F
Phanuphak
P
Ruxrungtham
K
, et al.  . 
Comparison of first-line antiretroviral therapy with regimens including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: a randomised open-label trial, the 2NN Study
Lancet
 , 
2004
, vol. 
363
 (pg. 
1253
-
63
)
6
Gazzard
B
British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy (2006)
HIV Med
 , 
2006
, vol. 
7
 (pg. 
487
-
503
)
7
Riddler
SA
Haubrich
R
DiRienzo
G
A prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection: ACTG 5142
In: Abstracts of the Sixteenth International AIDS Conference, Toronto, Canada, 2006
  
Abstract THLB0204
8
De Luca
A
Cozzi-Lepri
A
Antinori
A
, et al.  . 
Lopinavir/ritonavir or efavirenz plus two nucleoside analogues as first-line antiretroviral therapy: a non-randomized comparison
Antivir Ther
 , 
2006
, vol. 
11
 (pg. 
609
-
18
)
9
Torti
C
Maggiolo
F
Patroni
A
, et al.  . 
Exploratory analysis for the evaluation of lopinavir/ritonavir-versus efavirenz-based HAART regimens in antiretroviral-naive HIV-positive patients: results from the Italian MASTER Cohort
J Antimicrob Chemother
 , 
2005
, vol. 
56
 (pg. 
190
-
5
)
10
Panagopoulos
P
Tsiodras
S
Antoniadou
A
, et al.  . 
Efficacy and safety of an anti-retroviral combination regimen including either efavirenz or lopinavir-ritonavir with a backbone of two nucleoside reverse transcriptase inhibitors
Clin Microbiol Infect
 , 
2006
, vol. 
12
 (pg. 
486
-
9
)
11
Centers for Disease Control
1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults
MMWR
 , 
1992
, vol. 
41
 (pg. 
1
-
19
)
12
Expert Panel on Detection Evaluation and Treatment of High Blood Cholesterol in Adults
National Cholesterol Education Program: Adult Treatment Panel III Report
 , 
2001
 
Publication 01-3095
13
Montaner
JS
Schutz
M
Schwartz
R
, et al.  . 
Efficacy, safety and pharmacokinetics of once-daily saquinavir soft-gelatin capsule/ritonavir in antiretroviral-naive, HIV-infected patients
MedGenMed
 , 
2006
, vol. 
8
 pg. 
36
 
14
Bartlett
JA
Johnson
J
Herrera
G
, et al.  . 
Long-term results of initial therapy with abacavir and lamivudine combined with efavirenz, amprenavir/ritonavir, or stavudine
J Acquir Immune Defic Syndr
 , 
2006
, vol. 
43
 (pg. 
284
-
92
)
15
Squires
K
Lazzarin
A
Gatell
JM
, et al.  . 
Comparison of once-daily atazanavir with efavirenz, each in combination with fixed-dose zidovudine and lamivudine, as initial therapy for patients infected with HIV
J Acquir Immune Defic Syndr
 , 
2004
, vol. 
36
 (pg. 
1011
-
9
)
16
Eron
J
Jr
Yeni
P
Gathe
J
Jr
, et al.  . 
The KLEAN study of fosamprenavir-ritonavir versus lopinavir-ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: a randomised non-inferiority trial
Lancet
 , 
2006
, vol. 
368
 (pg. 
476
-
82
)
17
Manfredi
R
Calza
L
Chiodo
F
First-line efavirenz versus lopinavir-ritonavir-based highly active antiretroviral therapy for naive patients
AIDS
 , 
2004
, vol. 
18
 (pg. 
2331
-
3
)
18
DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents
Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents
  
(10 October 2006, date last accessed)
19
Iribarren
JA
Labarga
P
Rubio
R
, et al.  . 
Spanish GESIDA/Nacional AIDS Plan Recommendations for antiretroviral therapy in HIV-infected adults (October 2004)
Enferm Infecc Microbiol Clin
 , 
2004
, vol. 
22
 (pg. 
564
-
642
)
20
Maggiolo
F
Ravasio
L
Ripamonti
D
, et al.  . 
Similar adherence rates favor different virologic outcomes for patients treated with nonnucleoside analogues or protease inhibitors
Clin Infect Dis
 , 
2005
, vol. 
40
 (pg. 
158
-
63
)
21
Podzamczer
D
Ferrer
E
Gatell
JM
, et al.  . 
Early virological failure with a combination of tenofovir, didanosine and efavirenz
Antivir Ther
 , 
2005
, vol. 
10
 (pg. 
171
-
7
)
22
Barrios
A
Rendon
A
Negredo
E
, et al.  . 
Paradoxical CD4+ T-cell decline in HIV-infected patients with complete virus suppression taking tenofovir and didanosine
AIDS
 , 
2005
, vol. 
19
 (pg. 
569
-
75
)
23
Matarrese
P
Gambardella
L
Cassone
A
, et al.  . 
Mitochondrial membrane hyperpolarization hijacks activated T lymphocytes toward the apoptotic-prone phenotype: homeostatic mechanisms of HIV protease inhibitors
J Immunol
 , 
2003
, vol. 
170
 (pg. 
6006
-
15
)
24
Anderson
PL
Kakuda
TN
Lichtenstein
KA
The cellular pharmacology of nucleoside- and nucleotide-analogue reverse-transcriptase inhibitors and its relationship to clinical toxicities
Clin Infect Dis
 , 
2004
, vol. 
38
 (pg. 
743
-
53
)
25
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
In: Abstracts of the Fourteenth Conference on Retroviruses and Opportunistic Infections, Los Angeles, CA, 2007
 
Alexandria, VA, USA
Foundation for Retrovirology and Human Health
 
Abstract 38