Long-term safety and durable antiretroviral activity of lopinavir/ritonavir in treatment-naive patients: 4 year follow-up study

Objective: Combination antiretroviral therapy with lopinavir/ritonavir (LPV/r) has been highly effective in clinical trials. Results of long-term therapy with LPV/r-based regimens have not been previously reported. This study describes the 4-year (204-week) safety and antiretroviral activity of LPV/r-based treatment in antiretroviral-naive individuals.

Design: Long-term, open-label follow-up of a phase II, prospective, randomized, multicenter trial.

Methods: A group of 100 antiretroviral-naive HIV-infected patients were randomized to one of three blinded doses of LPV/r [200/100 mg (n = 16), 400/100 mg (n = 51), or 400/200 mg (n = 33)] with stavudine 40 mg and lamivudine 150 mg every 12 hours. After 48 weeks, LPV/r was dosed open-label at 400/100 mg every 12 hours with stavudine and lamivudine.

Results: Mean baseline plasma HIV-1 RNA and CD4 cell count were 4.9 log₁₀ copies/ml and 338 × 10⁶ cells/l, respectively. At week 204, 72 patients remained on study, 70 of whom had HIV-1 RNA < 50 copies/ml (70% by intent-to-treat analysis). Twenty-eight patients discontinued therapy prior to week 204 because of adverse events (n = 10), lost to follow-up (n = 9), or other reasons (n = 9). Of 15 patients who met protocol-defined criteria for virologic failure, seven remained on the study regimen and their HIV-1 RNA was re-suppressed to < 50 copies/ml at week 204. Genotypic analysis of rebound viral isolates was available from 10 patients, including all eight patients who discontinued the study prematurely. No isolate demonstrated primary or active site mutations in protease. The most common adverse events were gastrointestinal symptoms and lipid elevations.

Conclusions: LPV/r-based therapy provides durable antiretroviral response and is generally well tolerated through 204 weeks of therapy.

Introduction

The introduction of potent antiretroviral agents into clinical practice has resulted in marked improvement in survival as a consequence of improved suppression of HIV-1 replication and consequent recovery of immune function [^1-3]. Among the most potent of such agents is lopinavir (a HIV-1 protease inhibitor) co-formulated with ritonavir (a cytochrome P450 3A4 enzyme inhibitor) to enhance its pharmacokinetic profile (LPV/r) [⁴]. The resulting mean trough concentrations of lopinavir are at least 75 times as high as the protein-binding corrected IC₅₀ (concentration to inhibit 50% replication in vitro) for wild-type HIV [⁵]. In a 48-week double-blind, randomized trial, conducted in antiretroviral-naive patients, LPV/r demonstrated superior virologic efficacy when compared with nelfinavir, both dosed with stavudine and lamivudine [⁶].

Despite numerous reports of the efficacy of various combinations of antiretroviral agents, there are few long-term virologic, immunologic, and tolerability data derived from prospective clinical trials of therapeutic regimens [^7-9]. This report describes a 4-year follow-up of individuals enrolled in the first trial of LPV/r in antiretroviral-naive, HIV-infected patients.

Methods

Study design

Details of the study design and results through 48 weeks have been previously reported [¹⁰]. A group of 100 antiretroviral-naive patients were randomized to one of three blinded doses of LPV/r: 200/100 mg (n = 16), 400/100 mg (n = 51), or 400/200 mg (n = 33), given with stavudine 40 mg and lamivudine 150 mg every 12 hours. After 48 weeks, all patients received open-label LPV/r 400/100 mg every 12 hours. The 400/100 mg dose was selected in order to maximize plasma lopinavir levels while maintaining a favorable tolerability profile [¹⁰]. At baseline, patients had HIV-1 RNA > 5000 copies/ml, had no prior antiretroviral therapy, and were not acutely ill. There was no CD4 cell count restriction. The study was approved by the Institutional Review Board at each participating site and all patients provided written informed consent. Plasma HIV-1 RNA was quantified using the Roche Amplicor HIV-1 Monitor assay (Roche Molecular Systems, Branchburg, New Jersey, USA) with limits of quantification (LOQ) of 400 copies/ml and 50 copies/ml. CD4 and CD8 cell counts, hematology, and clinical chemistry evaluations were performed at each visit. Genotypic analyses of rebound viral isolates were determined by population sequencing of resistance test vector pools using dye-terminator chemistry and analysis on an ABI 3700 automated sequencer (ViroLogic, South San Francisco, California, USA).

Statistical analysis

All patients who received at least one dose of study medication were included in the analyses. The original primary endpoint for this study, the proportion of patients with HIV-1 RNA levels < 400 copies/ml at week 24, was extended for this long-term follow-up analysis. Thus, the primary efficacy variable for this analysis was the proportion of patients with HIV-1 RNA below the limit of quantification at week 204 by intent-to-treat analysis, with missing values considered failure. Secondary endpoints included the on-treatment proportion of patients with HIV-1 RNA below the limit of quantification, in which discontinuations, missing values, and those obtained during treatment interruption of at least 3 days were excluded, and the mean change from baseline in CD4 cell count. Adverse events were summarized using COSTART V [¹¹].

Results

Study patients and disposition

Demographic characteristics of this cohort have been published previously [¹⁰]. Mean age was 35 years, 96% were male, 65% were Caucasian, 29% were African-American, and 6% were Hispanic. Mean (± SD) baseline plasma HIV-1 RNA was 4.89 (± 0.68) log₁₀ copies/ml, and the mean baseline CD4 cell count was 338 (± 249) × 10⁶ cells/l. Twenty-eight (28%) patients discontinued the study prior to week 204 for the following reasons: adverse events (10), loss to follow-up (nine), non-adherence (four), relocation (three), substance abuse (one), and primary provider choice (one).

Antiretroviral and immunologic responses

By intent-to-treat analysis at week 204, HIV-1 RNA was < 400 copies/ml for 71% (71/100) and < 50 copies/ml for 70% (70/100) of patients (Fig. 1a), with corresponding on-treatment results of 99% (71/72) < 400 copies/ml and 97% (70/72) < 50 copies/ml. No important differences in antiretroviral activity were observed with respect to originally randomized dose group. Fifteen patients demonstrated loss of virologic response (defined as two consecutive rebound HIV-1 RNA values > 400 copies/ml, one rebound HIV-1 RNA value followed by discontinuation, or failure to achieve HIV RNA < 400 copies/ml), including eight patients who subsequently discontinued the study (Table 1). Genotypic analysis of viral isolates was available for 10 of these patients and revealed no primary or active site mutations in protease (amino acids at positions 8, 30, 32, 36, 47, 48, 50, 82, 84, and 90). Isolates from three of these patients demonstrated lamivudine resistance (M184V mutation in reverse transcriptase). When comparing patients with or without lamivudine resistance, no differences were observed in baseline viral load, rebound viral load level, or time since last viral load level < 400 copies/ml.

Fig. 1

Table 1

Substantial increases in CD4 cells were observed at all study visits. Among patients with samples from both baseline and week 204, the mean CD4 cell count increased from 281 to 721 × 10⁶ cells/l at week 204, a mean increase of 440 (± 249) × 10⁶ cells/l. CD4 cell count increases were similar across baseline CD4 cell strata (Fig. 1b). Among 72 patients remaining on study as of week 204, 65 (90%) had CD4 cell counts > 350 × 10⁶ cells/l. The remaining seven patients had a median CD4 cell count at week 204 of 238 × 10⁶ cells/l (range, 137-344); all seven of these patients had an increase from baseline of at least 100 × 10⁶ cells/l.

Adverse events

Adverse events led to premature study discontinuation in 10 patients. For seven patients, the event was considered at least possibly related to study medication by the investigator, including one patient who suffered a perioperative myocardial infarction and later died from unknown causes 10 days following thoracic spinal surgery. Other adverse events leading to study discontinuation thought to be probably or possibly related to the treatment regimen included asymptomatic liver function test increases (n = 2), diarrhea (n = 1), arthralgia (n = 1), symptomatic hepatomegaly with steatosis (n = 1), and elevated cholesterol (n = 1). Three patients discontinued the study because of adverse events considered by the investigator to be unrelated to study medication (lymphoma, hyperglycemia in a diabetic patient, alcohol detoxification).

Through 204 weeks of treatment, the most common adverse events of moderate or severe intensity considered probably or possibly related to LPV/r were gastrointestinal, including diarrhea (defined as more than three loose stools per day, cumulative incidence 27%), nausea (16%), abdominal pain (10%), abnormal stools (one to three loose stools per day, 8%), and vomiting (6%). Other events occurring in more than 5% of patients were asthenia (9%) and headache (7%). Most such symptoms were self-limited. At week 204, four patients noted on-going such events of moderate or greater intensity: diarrhea (n = 2), abnormal stools (n = 1), and headache (n = 1). Through week 204, events of greater than moderate intensity were uncommon; diarrhea was the only drug-related event of severe intensity reported by more than two patients (four).

Protocol-defined assessments for changes in fat distribution were not included in the study design. Through 204 weeks of treatment, events of fat loss (n = 14), fat gain (n = 5), or both (n = 6) had been reported spontaneously by investigators. Median time of onset for these events was approximately 120 weeks of treatment.

Laboratory values were obtained without regard to fasting. Through week 204, grade 3 or greater elevations in total cholesterol (> 300 mg/dl), triglycerides (> 750 mg/dl), and aspartate/alanine aminotransferases were demonstrated by 22%, 22% and 11%, respectively. At week 204, only one patient had a grade 3 cholesterol level and four patients had grade 3 triglyceride elevations, while 81% of patients on study had total cholesterol of grade 1 (< 240 mg/dl) or less (21% of whom had concomitant use of lipid-lowering agents) and 68% of patients had triglycerides of grade 1 (< 400 mg/dl) or less (18% of whom had concomitant use of lipid-lowering agents). Two patients who discontinued the study prior to week 204 had lipid measurements grade 3 or greater. After initial increases, median lipid values remained relatively stable over time; at weeks 0, 48, 96, 144, and 204, the median total cholesterol values were 154, 208, 203, 205, and 203 mg/dl, respectively, while the median triglycerides values were 133, 226, 233, 244, and 227 mg/dl, respectively. Nineteen patients initiated therapy with lipid-lowering agents during the study, 15 of whom remained on study through week 204. Aspartate and/or alanine aminotransferases elevations grade 3 or greater occurred in 5/11 (45%) patients with baseline positive hepatitis B surface antigen and/or hepatitis C antibody, compared with 6/89 patients (7%) with negative hepatitis B/C serologies (relative risk, 6.7; 95% confidence interval, 2.5-18.5). The majority of these elevations returned to baseline levels with continued dosing.

Discussion

The 4-year follow-up results in this study extend the 48-week observations reported previously [¹⁰]. LPV/r provides a potent and durable antiretroviral response in HIV-1-infected individuals when used as initial antiretroviral therapy in combination with stavudine and lamivudine. Few prospective studies of 3 or more years of highly active antiretroviral therapy have been reported [^8,12,13]. While cross-study comparisons are of limited utility partly because of differences in patient populations, concomitant antiretroviral agents, and other factors, virologic suppression results from the present study (70% of enrolled patients with HIV-1 RNA < 50 copies/ml at 204 weeks) compare favorably with long-term results for efavirenz-based [⁸], indinavir-based [¹²], and nelfinavir-based [¹³] regimens (31-55% of enrolled patients with HIV-1 RNA < 50 copies/ml after 3 to 4 years of treatment) in patients without prior exposure to highly active antiretroviral therapy. One randomized, comparative 48-week study of LPV/r- and nelfinavir-based regimens found superior virologic suppression with the LPV/r-based regimen [⁶]. However, long-term follow-up of head-to -head comparisons of antiretroviral regimens is needed to assess adequately the relative durability and safety of specific regimens.

In this cohort, no protease inhibitor resistance was observed in patients with sustained viral load rebound, although few patients had amplifiable levels of HIV-1 RNA. These results are similar to those from a larger phase III study of LPV/r [⁶]. With the increasing concerns about the prevalence of primary HIV drug resistance [¹⁴], selection of an antiretroviral regimen that minimizes resistance development by maintaining viral suppression over time and providing a high barrier to the development of resistance may be particularly important.

The low rate of treatment discontinuation attributed to study-drug-related adverse events demonstrates the favorable tolerability profile of this regimen, and it may also reflect increased attention to the importance of adherence, especially during therapy initiation. In this study, through 4 years of treatment, non-fasting lipid elevations remain the most common laboratory abnormality. Following an initial increase, cholesterol and triglyceride levels have remained stable over time. At week 204, over 80% of patients maintained total cholesterol levels below the high range (240 mg/dl) of the Adult Treatment Panel III classification (NCEP) [¹⁵], although 21% of these patients had been taking concomitant lipid-lowering agents. Patients with underlying hepatitis B or hepatitis C infection appeared to be at increased risk of developing further hepatic transaminase elevations, although these were treatment limiting in only two instances.

In summary, this long-term follow-up study of LPV/r-treated patients demonstrated durable HIV suppression and substantial increases in CD4 cell counts over 4 years. These data support the use of LPV/r in the initial treatment of HIV infection.

Acknowledgements

The authors acknowledge the invaluable support of the study coordinators and support personnel at the study sites. We would also like to express our gratitude and appreciation to all the patients who participated in this trial and particularly to those who continue to be followed.

Sponsorship: This study was supported by Abbott Laboratories, Abbott Park, Illinois, USA.

Note: Mary Albrecht (Department of Medicine, Harvard University, Boston, Massachusetts), Marshall Glesby (Weill Medical College of Cornell University, New York), Scott Hammer (Department of Medicine, Columbia University, New York), Melanie Thompson (AIDS Research Consortium of Atlanta, Atlanta, Georgia), Peter Wolfe (Pacific Oaks Research, Beverly Hills, California), and Eugene Sun (Abbott Laboratories, Abbott Park Illinois) also contributed to this work.